http://localhost:4503/content/lw/en/blogs/lightwave-blog.html2016-09-02T01:37:16.883ZThe Lightwave Blog ArchivesRead detailed & penetrating opinions & analysis of fiber optic market, network & communications.Adobe Experience ManagerSCTE Cable-Tec Expo 2015 Reporter's Notebooknoemail@noemail.orgStephen Hardy<p>The 2015 edition of the SCTE Cable-Tec Expo took place October 13-16 in New Orleans. While there were some notable optical communications related announcements at the show, the major topic was the upcoming coax-based alternative to fiber to the home, DOCSIS 3.1.</p> <p>I moderated a panel on the subject for sister site Broadband Technology Report (and you can watch it at <a adhocenable="false" href="" target="_blank"></a>). Here were the key takeaways:</p> <p><li>· The specifications, while not yet etched in stone, are solid enough to create working hardware.</li> <li>· And that working hardware is well down the road toward generation availability.</li> <li>· A certification process is in place and waiting for products to run the gauntlet. I was told we'd see certified products “very soon,” which is as specific as anyone was willing to be.</li> <li>· Field trials are underway and showing good results. There have been the usual bumps one encounters when testing a new technology, but nothing that appears to be a showstopper.</li> <li>· Comcast's Jorge Salinger reports that his company's trials indicate the technology is on track for deployments in 2016.<br> The technology is designed to work on existing DOCSIS 3.0 plant, and so far is proving to work along those lines. In fact, bonding with DOCSIS 3.0 channels will prove to be a major application.</li> <li>· But operators need to open the necessary spectrum to get the full benefit of DOCSIS 3.1. Salinger said that Comcast has already begun this process. How well that operator and its fellow cable companies perform this task may prove to be a major driver of deployment schedules.</li> </p> <p>CableLabs held an interop/demo shortly before the show. As Ron Hendrickson of <a adhocenable="false" href="" target="_blank">BTR reported</a>, several of the demos showed multi-gigabit capability, typically with 3 to 4 Gbps downstream and approximately 1 Gbps.</p> <p>But one could find some new optical communications technology on the show floor. As I described elsewhere, Fujitsu Network Communications unveiled the A100, a 10G EPON optical line terminal (OLT) that's part of its new 1FINITY line. Sources at Fujitsu's booth said they already have two customers for the platform on the hook.</p> <p>Calix announced the E3-8G , a node-based GPON OLT for fiber-deep applications. The DOCSIS-friendly platform is designed to support gigabit services initially, with an upgrade path to 10G. The company also launched the 856G GigaCenter, a CPE platform that supports MoCA 2.0, a variety of services and resident/cloud-based applications, and gigabit Wi-Fi.</p> <p>CommScope showed off upgrades to some of the technology the company acquired along with the Broadband Network Solutions (BNS) business <a adhocenable="false" href="">it bought from TE Connectivity</a>. I've written <a adhocenable="false" href="">a separate story</a> on it.<br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2015/10/scte-cable-tec-expo-2015-reporters-notebook.html2015-10-23T18:00:00.000Z2015-10-23T21:57:56.601ZECOC Reporter's Notebook: Day 3noemail@noemail.orgStephen Hardy<p>So let's wrap things up quickly with a few odds and ends...</p> <p>The story I posted on <a adhocenable="false" href="">Finisar showing off a CFP8</a> garnered a lot of interest, so I'm start with that. First, I should emphasize that what they displayed was not a working 400 Gigabit Ethernet transceiver. It wasn't even a prototype of such a transceiver. It was, as the company's Rafik Ward described it, &quot;a mock up running SR16 in optical loopback mode.&quot; The idea was to show off the CFP8 format and see what customers thought of it; the format is an official proposal within the CFP MSA. The package would accommodate all of the 400 Gigabit Ethernet PMDs, including the 4x100G parallel singlemode one.</p> <p>Finisar, of course, is working on 400GbE modules, Rafik said. When I brought up that I had heard someone had already done field trials of a 400GbE transceiver and if he cared to comment on that, Rafik just kind of chuckled and said, &quot;No.&quot;</p> <p>The company also showed off a 100GbE shortwave WDM (SWDM) transceiver, a follow-on to the 40GbE module displayed at OFC. SWDM technology enables the transmission of four wavelengths down multimode fiber. (<a adhocenable="false" href="">There's an alliance</a> built around the technology.) I've associated SWDM with wideband multimode fiber (WBMMF), but Rafik pointed out that Finisar's modules will work with conventional multimode fiber as well. The device demonstrated in Valencia will support 75 m on conventional OM3 and 100 m on conventional OM4. For comparison, CommScope was showing 300-m reach at their booth using <a adhocenable="false" href="">their newly announced WBMMF</a>.</p> <p>The Finisar device is in alpha sampling; Rafik declined to say when general availability is expected. He did reveal that Finisar has licensed the technology to fellow SWDM Alliance member Lumentum to enable a second source.<br> </p> <p>I had a nice chat with Oclaro's Adam Carter for a podcast I'm composing on data center optics (there will be another one on 100G in the data center; I hope to have both completed and up on the site next week). Among other things, Carter said that he thought the single-lambda 100G that will emerge from the 400 Gigabit Ethernet specification process will prove to be a significant building block for other applications going forward. That includes data center interconnect (naturally) but also whatever speeds happen to succeed 400G.</p> <p>Earlier on Wednesday, Jorg-Peter Elbers of ADVA Optical Networking pointed out that if PAM4 is selected for data center interconnect, you might need dispersion compensation to support all of the target reaches. DMT transmission would be more immune to impairments, but more expensive. (At the Lumentum booth, CTO Brandon Collings said his company continues to work on DMT for this sort of requirement; he said a development ecosystem for the necessary enablers is being created.)<br> </p> <p>Carter also said there may be a divergence between OTN systems and packet-optical transport platforms regarding interfaces, both in terms of format (CFP2 for OTN, CFP4 for POTS) and data rate (400G for OTN and multiple 100G, and maybe 200G, for POTS).</p> <p>Meanwhile, proving that there are a lot of things to consider when introducing a new data rate into the world, a source at USConec explained to me the pains taken to avoid having cable with the company's 400GbE-targeted <a adhocenable="false" href="">MTP-16 connector</a> plugged into a standard MTP receptacle. The company has shifted the locking key on the MTP-16 to one side and changed the ferrule pitch, meaning a technician will have to be very determined to make such a mistake.</p> <p>There were of course more conversations at the show than I've had time to report in these Notebooks. I hope to hit these in the <a adhocenable="false" href="">&quot;ECOC 2015 Wrap-up&quot; webcast</a> I'll present this coming Wednesday, October 7.<br> </p> <p>And now I have to finish packing...</p> <p>Read the other ECOC 2015 Notebooks: <a adhocenable="false" href="">Day 1</a>&nbsp;&nbsp; <a adhocenable="false" href="">Day 2</a>.</p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2015/09/ecoc-2015-reporters-notebook-day-3.html2015-10-01T07:15:00.000Z2015-10-02T19:59:43.765ZECOC 2015 Reporter's Notebook: Day 2noemail@noemail.orgStephen Hardy<p>Having proclaimed PAM4 The Official Data Center Modulation Format of ECOC 2015 <a href="" adhocenable="false">in yesterday's Notebook</a>, let's talk about it in more detail -- which is what I did with Inphi's Siddharth Sheth earlier today. Having announced <a href="" adhocenable="false">PAM4 PHYs and a companion TIA</a> in August, Sheth says Inphi has had plenty of customer conversations at the show about possible uses for the chips.</p> <p>Sheth says the conversations are wide ranging, encompassing single-lambda 40 Gigabit Ethernet as a lower cost alternative to LR4 lite, 50G for a possible optical transceiver targeting server access, 100 Gigabit Ethernet via 2x50G for a potential MSA, 400 Gigabit Ethernet, and data center interconnect. (There's interest in backplane applications as well.)</p> <p>In terms of what might reach the field soonest, Sheth surprised me by saying that he has a customer who has tested a 400 Gigabit Ethernet transceiver in a customer network already. (Although perhaps I shouldn't have been surprised -- <a href="" adhocenable="false">isn't someone showing of a 400 Gigabit Ethernet CFP8 at the show?</a>) The data center interconnect use should see commercialization fairly soon, he added, as will the single-lambda 40 Gigabit Ethernet module. He also said we should keep our eyes peeled for 50G and 100G modules based on PAM4 at OFC 2016.</p> <p>He says Inphi has more than 30 potential customers evaluating the technology.</p> <p>Speaking of data center interconnect, IHS analyst Andrew Schmitt pointed out today in a Market Focus presentation that the 100G 80-km reach application most commonly called &quot;data center interconnect&quot; actually encompasses a wide range of other applications. But whatever you call it (Andrew calls it &quot;metro access&quot;), the application promises to explode in six to eight years, with demand beginning to ramp next year (or at least everyone who is marketing data center interconnect systems hopes so). Schmitt believes coherent technology will dominate initially, but that direct-detect options based on PAM4, DMT, or TBD (that's &quot;too be determined,&quot; not a new modulation format) will challenge coherent's dominance around 2018.</p> <p>The pricing of such direct-detect devices will be a tricky issue, Ovum's Daryl Inniss pointed out to me in a later conversation. They will of course have to be cheap enough to appeal as an alternative to 10G. But as coherent prices decline, their cost advantage versus coherent may not be as pronounced as their suppliers might hope.</p> <p>Uwe Fischer, CTO at Coriant, said he sees the metro access in a light similar to Andrew's. The question is where direct detect will play. Coriant, of course, is working hard right now to reduce the cost of coherent transmission. But he sees the economic appeal of approaches based on modulation formats such as PAM4 for applications of 40 km or less. The two approaches will battle over applications between 40 km and, say, 80 km; that niche is very much up in the air, he said.<br> </p> <p>Read the other ECOC 2015 Notebooks: <a adhocenable="false" href="">Day 1</a>&nbsp;&nbsp; <a adhocenable="false" href="">Day 3</a></p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2015/09/ecoc-2015-reporters-notebook-day-2.html2015-09-29T22:00:00.000Z2015-10-02T19:41:46.798ZECOC 2015 Reporter's Notebook: Day 1noemail@noemail.orgStephen Hardy<p>Bienvenidos a Valencia, home of the European Conference on Optical Communication (ECOC) for 2015. Sunny Spain has not been so sunny so far this week, and rain is forecast for the remaining days of the event. Which I guess means there may be fewer distractions from ECOC's content.</p> <p>Being Day 1, it's of course early in the event -- but PAM4 is making a strong bid to become the Official Data Center Modulation Format of the ECOC 2015 Exhibition. For example, there's a joint demonstration of 400 Gigabit Ethernet LR8 transmission between Inphi and NeoPhotonics that pairs the former's <a adhocenable="false" href="">PAM4 ICs</a> with the latter's new Q-TOSA 100G EML-based modules. Discovery Semiconductor is showing off receiver technology that has been used in several PAM4-related demonstrations. TeraXion staff are talking about the progress they've made on a silicon photonics based PAM4 modulator that the subject of a paper at ECOC 2014. Test and measurement vendors such as Tektronix, Anritsu (as part of what is being touted as <a href="" adhocenable="false">&quot;a complete 400G ecosystem&quot;</a> for test and measurement that also includes Coherent Solutions' new <a adhocenable="false" href="">IQTransmitter</a>, plus oscilloscope technology from Teledyne LeCroy), and Keysight Technologies are touting their PAM4 test capabilities as well. And, again, it's just Day 1.<br> </p> <p>PAM4 also is being touted as a means of supporting shorter-length <a adhocenable="false" href="">data center interconnect</a> applications at lower cost than coherent transmission. However, OE Solutions remains interested in discrete multitone (DMT) technology for single-lambda 100G for such applications, a source there confirmed.</p> <p>Elsewhere in the data center realm, 100G QSFP28s remain the rage, particularly for CWDM4, CLR4 Alliance, and -- in the case of demonstrations from the likes of Molex (via its Oplink acquisition), Oclaro, and Source Photonics -- both in the same device. InnoLight has a CLR4 it's delivering exclusively to what a source at their stand described as &quot;our largest customer.&quot; Kaiam says it is sampling CWDM4 and CLR4 QSFP28 devices. And Sumitomo Electric Lightwave and Lumentum are showing off new CWDM4 optical modules; a source at Lumentum said their device will also support CLR4, and I suspect Sumitomo will tell me the same thing once I have a chance to swing by their booth (excuse me, &quot;stand&quot;).<br> </p> <p>Meanwhile, on the line side, it's all about coherent. We're seeing enabling laser and detector/receiver technology for 400G from NeoPhotonics as well as the previously reported 400G modulators from TeraXion and Oclaro.</p> <p>And 100G has not been forgotten. TeraXion staff briefed me on a silicon photonics enabled coherent receiver development that should produce samples by the end of the year/early 2016.</p> <p>Not to be overlooked in between, of course, is the fact that <a adhocenable="false" href="">NEL finally came up with a DSP</a> to support 200G via 16QAM (not to mention 150G via 8QAM).<br> </p> <p>Read the other ECOC 2015 Notebooks: <a adhocenable="false" href="">Day 2</a>&nbsp; <a adhocenable="false" href="">Day 3</a></p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2015/09/ecoc-2015-reporters-notebook-day-1.html2015-09-28T20:04:00.000Z2015-10-02T19:40:32.670Z2015 FTTH Connect Reporter’s Notebook, Day 1noemail@noemail.orgStephen Hardy<p>The FTTH Connect conference (nee the FTTH Council Americas' FTTH Conference) is in full swing at the Anaheim Convention Center. So is a high school girls volleyball tournament. At both events, we find:<br> </p> <h2>Bump</h2> <p>Roland Montagne of IDATE, who does a lot of the FTTH Council Europe's market research work, reported that FTTH activity in Latin America bumped up significantly in 2014 over the prior year. For example, FTTH/B subscriber number in the region jumped 57%. Brazil is the largest <a href="/content/lw/en/fttx/ftth-b.html" adhocenable="false">FTTH/B</a> market, with Telefonica Vivo Brazil doubled its coverage. In terms of total number of FTTH/B subscribers, Mexico falls right behind Brazil. Meanwhile, Uruguay's Antel receives a gold star for passing 95% of the country's subscribers with fiber-optic broadband access infrastructure.<br> <br> Michael Render of RVA LLC, who handles market research for the North American group, said that the annual official FTTH/B tally won't be available until September. However, he provided the results of a recent consumer survey that covered perceptions of FTTH. Among the tidbits are that most consumers think of &quot;fiber-optic Internet&quot; rather than &quot;fiber to the home&quot; and that the &quot;time and hassle&quot; involved and concerns with new service contracts were reported as barriers to switching to FTTH by 67% of survey respondents.<br> </p> <h2>Set</h2> <p>With more cable MSOs openly discussing use of FTTH to serve customers (with <a href="" adhocenable="false">Comcast's 2-Gbps announcement</a> a primary example), this demographic was the focus of a pair of presentations on Monday. In one, representatives from CableLabs talked about the cable operator landscape and how CableLabs helps out through such initiatives as DOCSIS Provisioning of EPON (DPoE) and GPON (DPoG). The organization also has a modeling system to help cable operators figure out how to optimize oversubscription strategies, determine appropriate service tiers for given architectures, architecture life expectancy, etc.<br> <br> While the CableLabs speakers were careful to avoid discussing what specific operators might do in the future, both they and ADTRAN's Kurt Raaflaub, who ran the other cable MSO focused session, said it was unlikely FTTH deployments would go away once DOCSIS 3.1 technology became generally available. (They're both tools in the tool kit, Raaflaub said to me after reading this blog earlier today.) However, the CableLabs speakers also didn't predict a wholesale adoption of FTTH anytime soon, either (no suprise there).</p> <p> The question, of course, is how many of today's infrastructure upgrades to meet competitive threats would leverage DOCSIS 3.1 if it were available, rather than FTTH. With DOCSIS 3.1 deployments expected to be underway next year, I guess we'll get an indication fairly soon.<br> </p> <h2>Spike</h2> <p>As I wrote on Lightwave's sister site <a href="" adhocenable="false">Broadband Technology Report</a>, there is general agreement that likely will see its first major use within MDUs as part of a fiber to the building deployment. However, panelists on the conference's opening MDU-focused session -- including Michael Weston of Verizon and Andre Kriger of Telefonica Vivo Brazil -- emphatically spiked the idea that we'll see in their MDU roll outs. The business case for any sort of copper-based optical access isn't attractive, Kriger said, while Weston said that Verizon wants to enter a building just once to enable high-speed broadband services, so future-proofing with fiber was the preferred option.<br> <br> That said, <a href="" adhocenable="false">as was the case last year</a>, systems vendors weren't shy about showing off capabilities on the show floor. For example, Cisco sources discussed the company's plans to support the technology, in addition to the ME 4600 combination GPON/point-to-point Ethernet OLT that was the focus of the company's booth. Cisco has it technology, which should be generally available this November, in the labs of a U.S. Tier 1 carrier. Meanwhile, the ME 4600 has been shipping for over a year to customers; Cisco appears to be having significant success with the platform in the Middle East.<br> <br> Dasan Networks also is highlighting its capabilities.<br> <br> And speaking of spikes, we continue to see a spike in support of NG-PON2 and 10G PON. Calix is debuting <a href="" adhocenable="false">its NG-PON2 capabilities</a> at the show, and the Cisco sources said both U.S. Tier 1 carriers are kicking the tires on their technology. Meanwhile, DPoE friendly 10G EPON should be available in the second half of this year, they added.<br> <a href="" adhocenable="false"><br> Alcatel-Lucent</a> and <a href="" adhocenable="false">ADTRAN</a> also are showing off their previously announced next-gen FTTH capabilities.<br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2015/06/2015-ftth-connect-reporters-notebook-day-1.html2015-06-30T14:15:00.000Z2015-07-01T18:25:16.205ZWe've made a few tweaksnoemail@noemail.orgStephen Hardy<p>The sharper-eyed among you may have noticed we've tweaked the website over the past couple of weeks. We've added a pair of new topic centers, moved another one, and dropped one.</p> <p>The two new ones are <a adhocenable="false" href="">SDN/NFV</a> and <a adhocenable="false" href="">Data Center</a>. Yes, both are long past due. The topic center titles are self-explanatory, as far as what you'll find in each. You'll also see <a adhocenable="false" href="">Data Center Interconnectivity</a> as a standalone subtopic resource within the Data Center domain.</p> <p>All three areas, of course, represent growing opportunities for optical communications. We've been covering these spaces for a while now, but it seemed proper to give them the visibility on the website that they've earned in the field.</p> <p>To make room, we've moved our <a adhocenable="false" href="">Mobile Backhaul</a> topic center underneath our <a adhocenable="false" href="">Network Design</a> domain. it's not that we have anything against mobile backhaul; it just seemed that technology development there had matured to the point that it was safe to move our coverage of this space to the domain where we cover similar applications. (And I reserve the right to change my mind once mobile fronthaul evolves from a topic of conversation to matter of immediate need.)</p> <p>Finally, we also dropped Cable/MSO as a standalone topic center. Again, we're going to fold that coverage into the Network Design domain. Besides, our sister site, <a adhocenable="false" href="" target="_blank">Broadband Technology Report</a>, does a pretty fair job of covering this space. You can look toward future collaborations between the two of us in the very near future.</p> <p>As always, we make changes like these to help meet your evolving information needs. If you have ideas about how we can better achieve that goal, I'm all ears.<br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2015/04/weve-made-a-few-tweaks.html2015-04-13T19:00:00.000Z2015-04-30T21:22:47.136ZOFC 2015 Reporter's Notebook: Day 2noemail@noemail.orgStephen Hardy<p>I don't know if it's a fully coordinated effort or merely coincidence, but I've met with several companies so far (Finisar, OFS, and Avago among them, with CommScope joining the list on Thursday) who want you all to know that <a href="" adhocenable="false">multimode fiber</a> isn't dead yet. Sure, Microsoft, Facebook, and other data center operators are touting their preferences for <a href="" adhocenable="false">single-mode fiber</a>. However, the majority of the data center networks don't require the reaches of such mega-scale infrastructures, the various sources point out. The 300 m or so that multimode fiber links can support -- particularly with extended reach optical transceivers -- will cover the vast majority of data centers, I'm assured.</p> <p>Along these lines, Finisar is showing off its SWDM (shortwave WDM) technology that leverages its VCSEL expertise to support multi-wavelength transmission over multimode links. While the TIA is turning its attention to a possible specification for a wideband multimode fiber for multi-wavelength transmission, the SWDM technology works on OM4, according to Finisar's Rafik Ward. The company is showing a four-wavelength 40 Gigabit Ethernet QSFP28 for multimode, as well as a technology demonstration of how they might similarly support 100 Gigabit Ethernet...</p> <p>CFP2-ACO coherent transceivers, QSFP28 100 Gigabit Ethernet optical modules, and CFP4 devices are as popular on the show floor as I expected. But I wasn't ready for the number of SFP28 transceivers I've seen. The device targets 25 Gigabit Ethernet and 32G Fibre Channel applications. <a href="" adhocenable="false">Oclaro</a>, Finisar, and Avago are among several companies showing off the technology. Meanwhile, the IEEE has just started work on its 25GbE specifications. Standards? Who needs standards?...</p> <p>Conspicuous by its absense from the <a href="" adhocenable="false">Consortium for On-Board Optics</a> is Avago, the company that has had so much success with onboard optics in high-performance computing applications. I was told by an Avago source that the company looked at the roster of founding members and dispaired of such a diverse group achieving consensus. The company was also leery of IP sharing provisions within the organization. The souce said the company may reconsider its decision in the future once it sees what progress the coalition makes.</p> <p>There is a lot of <a href="" adhocenable="false">WDM-PON</a> activity among exhibitors, particular approaches that use C-Band and/or L-Band optics. Companies such as Aeponyx, Transmode, ADVA Optical Networking (which has a demo leveraging Oclaro tunable SFPs), and Huawei are all touting their progress. Huawei's Frank Ellenberger, who is about to becom a newly minted OSA Fellow, said it appears that 10G PON may prove to be the next-gen <a href="/content/lw/en/fttx/pon-systems.html" adhocenable="false">PON</a> approach for support of residential services, while WDM-PON may make its mark in applications such as wireless front haul...</p> <p>And here's another anecdote concerning yesterday's research into whether data center operators are really going to throw away their networking gear every three years. I was approached at the very well attended LightCounting evening seminar by a gentleman who said he had a friend who works at one of the larger data center operators. This friend was responsible for disposing of all of the servers and whatnot that the operator was indeed throwing away. Apparently the volume of equipment that needed to be removed presented a signficant problem, I was told. So the score now stands No: 1, Yes: 1.<br> </p> <p>Check out the Reporter's Notebook for <a adhocenable="false" href="">Day 1</a>.<br> </p> <p>&nbsp;</p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2015/03/ofc-2015-reporters-notebook-day-2.html2015-03-26T03:30:00.000Z2015-05-22T15:46:46.886ZOFC 2015 Reporter's Notebook: Day 1noemail@noemail.orgStephen Hardy<p>I <i>hate it</i> when I'm wrong! For months I've been telling anyone who's asked that I thought Finisar would have little or no interest in buying JDSU's CCOP business when <a href="" adhocenable="false">it spins out as Lumentum later this year</a>. (Too much product overlap, I thought, and so I saw such a purchase as a market share buy, which I didn't think the company was interested in pursuing.) So my OFC 2015 got off to a very annoying start this morning when I filmed a video interview with Finisar Executive Chairman Jerry Rawls. I asked Jerry if, were Lumentum on the market, he'd be interested. &quot;Yes,&quot; he said without hesitation.</p> <p>Like the financial analysts that have been speculating along these lines, Jerry said market consolidation would be in Finisar's best interests (not to mention that of the rest of the industry). I happened to film an interview with Lumentum CEO-designate Alan Lowe later in the day and relayed to him Jerry's comment. So if they haven't discussed this already, I have done my part to facilitate a more rational market...</p> <p>Much is being made here about how much the way companies in the suddenly white-hot data center/Web 2.0 space do business differs from the traditional service provider models. One of the more popular tenets, for example, is that the data center folks &quot;throw out&quot; their networking gear every three years. Such pronouncements were met with disbelief by some telco attendees at the OSA Executive Forum on Monday. Supplierss have been equally curious about whether this commonly accepted concept is true, and I talked to one today who has done some investigation. According to this source's research, an employee of one such data center operator admitted they don't literally throw away their networking and related hardware so quickly. However, they do review their network strategies on a regular basis, and may buy different systems from different suppliers every three years. They stop buying the &quot;old&quot; equipment, but it doesn't end up in the recycling bin; these now legacy systems are merely &quot;demoted&quot; to handling less critical requirements...</p> <p>Meanwhile, at a press luncheon panel, Facebook hardware neworking engineer Yuval Bachar said that his networks currently are using 40 Gigabit Ethernet connections, and that these transmission rates could keep pace with requirements only for another year or so. He'd therefore like to start deploying 100 Gigabit Ethernet <a href="" adhocenable="false">optical modules</a> in his network soon. He initially implied that such <a href="" adhocenable="false">optical transceivers</a> were still too expensive for his tastes; however, he later said that there were 6-8 optical transceiver vendors that were getting close enough to the company's $1/gigabit goal that he's at least considering deploying their wares in the foreseeable future...</p> <p>ClariPhy Communications is showing demonstrations of Finisar and Oclaro coherent CFP2-ACO transceivers operating with ClariPhy's merchant <a href="" adhocenable="false">coherent DSP ASICs</a> (along with ClariPhy's newly announced reference design boards). The performance is good but could benefit from optimization, a process that hasn't taken place yet because ClariPhy received the modules less than a week before the show, I was told. They haven't tried to get the two modules to talk to each other yet for the same reason.</p> <p>&nbsp;</p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2015/03/ofc-2015-reporters-notebook-day-1.html2015-03-25T05:30:00.000Z2015-04-01T18:57:02.277ZWhy the JDSU brand will disappearnoemail@noemail.orgStephen Hardy<p>Ever since JDSU announced the names of its two post-split entities (see <a href="" adhocenable="false">&quot;JDSU reveals names for post-split companies&quot;</a>), I've received several comments suggesting that the company is missing a bet by not giving at least one of the new companies the JDSU brand. So I finally did what I should have done originally -- I asked JDSU for an explanation.</p> <p>For those catching up on this matter, JDSU plans to split into two companies by September 30, with one company formed primarily around the Communications and Commercial Optical Products (CCOP) unit -- which JDSU has called &quot;SpinCo&quot; -- and the remaining businesses, principally the network and service enablement units, forming the other. This latter group has been referred to as &quot;RemainCo&quot; (see <a href="" adhocenable="false">&quot;JDSU management details split, M&amp;A possibilities&quot;</a>).</p> <p>That's until now. SpinCo will become Lumentum, while RemainCo will become Viavi Solutions Inc.</p> <p>JDSU has posted the derivations of these names <a target="_blank" href="" adhocenable="false">on a subsection of its website</a> devoted to the split. Polite applause if you quickly determined that &quot;Lumentum&quot; is a conflation of &quot;lumen&quot; (or light) and &quot;momentum.&quot; But you earn a gold star if you figured out on your own that &quot;Viavi&quot; is meant to suggest &quot;through visibility&quot; by combining &quot;via&quot; and &quot;vision.&quot;</p> <p>Which brings us back to the original question on many minds: Why throw away the brand equity JDSU carries? According to Jim Monroe, JDSU's vice president of corporate marketing and communications, while management considered giving the JDSU name to SpinCo (and if I were Alan Lowe, I'd have begged for it), it was determined that because of the company's diversification over its history, the brand name no longer created a common vision among the disparate markets it had grown to serve. So it was determined the best strategy was to rename both entities in a manner that more clearly described their respective activities.</p> <p>I can see the wisdom of avoiding the confusion one post-split company would suffer if its name were given to the other one. On the other hand, I still wonder if there weren't some way to leverage the goodwill the current brand engenders.</p> <p>But, in the end, I doubt, for example, that Oclaro's market performance is at all tied to the fact that it is no longer called either Bookham, Avanex, or Opnext. The success of Lumentum and Viavi will depend on how their management teams execute, not on whether they lack any sort of head start the JDSU name may have given them.<br> </p> <p>&nbsp;</p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2015/03/why-the-jdsu-brand-will-disappear.html2015-03-06T17:30:00.000Z2015-03-11T14:35:57.344ZWatch for silicon photonics on the line sidenoemail@noemail.orgStephen Hardy<p>Most of the hullabaloo around <a href="" adhocenable="false">silicon photonics</a> has centered on its application to data center networks. But presentations and announcements at last month’s ECOC in Cannes, France, indicate the technology will play on the line side as well.<br> <br> For example, Acacia Communications used the show to further publicize its use of silicon photonics within its coherent <a href="" adhocenable="false">optical modules</a>. (The company started talking publicly about this strategy in detail at OFC this past March.) <a href="" adhocenable="false">Acacia’s coherent CFP</a> optical transceiver uses a silicon photonics enabled photonic integrated circuit (PIC) and is now in production. President and CEO Raj Shanmugaraj has told me the use of silicon photonics should prove a key enabler of Acacia’s planned digital coherent CFP2, helping Acacia squeeze the DSP and other necessary electronics alongside the optical functions within the device.<br> <br> We also learned in Cannes that Cisco is exploring the use of silicon photonics for coherent transmission as well. I had asked Bill Gartner, vice president and general manager of Cisco's High End Routing and Optical business unit, about this possibility <a href="" adhocenable="false">when the CPAK module was unveiled</a> at OFC 2013, and he indicated that Cisco was open to the idea. A post-deadline paper delivered at the show revealed some of the results of that openness. In &quot;Experimental Demonstration of Ultra-Low-Power Single Polarization 56 Gb/s QAM-16 Generation without DAC Using Silicon Photonics (PD2.5),&quot; researchers from Cisco and Stanford University reported…well, just what the paper’s title says. The researchers leveraged a silicon photonics-based modulator; the modulator’s design, using a Silicon Insulator Silicon Capacitor (SISCAP) structure, enabled segmentation of the active optical modulation region, meaning it could be driven by digital binary signals using a single IQ modulator without recourse to a digital-to-analog converters (DACs). The necessary optical and electrical devices required an aggregate power of less than 1 W, the researchers assert. In the described demonstration, the optical transmission created via the SISCAP modulator exhibited a signal-to-noise ratio of about 19 dB, which translates to a bit error rate of less than 10<sup>-4</sup>.<br> <br> Demonstrating such a technology in the lab doesn’t necessarily lead to incorporation into available modules, of course, so we’ll have to wait to see if Cisco productizes this advance. But another silicon photonics based effort for the line side does look like it may be ready for the field soon. While Skorpios Technologies says its near-term focus is on QSFP-28 optical modules for the data center, the company’s silicon photonics enabled tunable laser technology showed up twice in discussions of line-side applications. This included a presentation Skorpios delivered (co-authored with Ericsson) on &quot;High Power, Narrow Line-width, Low Noise, Integrated CMOS Tunable Laser for Long Haul Coherent Applications.&quot; The company also co-authored a paper with Coriant, Deutsche Telekom, and BISDN on ultra-dense <a href="" adhocenable="false">WDM-PON</a>.<br> <br> You can bet that these papers describe only a small percentage of the work now underway to apply silicon photonics to line-side applications. We’re still not sure if silicon photonics represents a revolution or just another tool in the kit. But it’s clear that its potential role is wide ranging.<br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2014/10/watch-for-silicon-photonics-on-the-line-side.html2014-10-23T20:04:00.000Z2014-11-03T22:12:25.527ZECOC 2014 Reporter's Notebook: Day 3noemail@noemail.orgStephen Hardy<p>The third and final day of the exhibition has come and gone – except, perhaps, for those literally lost souls who have not yet found their way to an exit. ECOC always has a rather bizarre floor plan, but this year they should have just brought in hedges and a minotaur.<br> <br> Here’s a whirlwind tour of what I ran across today and some of the things I haven’t yet covered from the first two days of the show.<br> <br> Telefonica has partnered with practically every optical systems vendor on the planet as they’ve investigated their <a adhocenable="false" href="">software-defined networking</a> (SDN) options. For example, the results of an exercise in multilayer control and integration with ADVA Optical Networking and Cisco were presented during the poster sessions today. Telefonica’s Juan Pedro Fernandez told attendees of a Market Focus session on IP/optical convergence that most of what the company has seen has been proprietary approaches, rather than the open, multivendor (as in multiple vendors at the optical layer, as opposed to one vendor at Layer 1 and another at Layer 2) approach they’d prefer. The carrier would like to see an approach based on Applications Based Network Orchestrator (ABNO) concepts, which would leverage IETF specified building blocks to create a modular controller. Nevertheless, Telefonica is optimistic enough about further progress that Fernandez expects a series of additional trials next year to lead to deployment of SDN concepts in Telefonica’s network in 2016.<br> <br> Speaking of the poster sessions, a source at Prysmian discussed the capabilities of the company’s new WideCap-OM4 Multimode Fiber in the context of 400 Gigabit Ethernet support. The fiber is designed to accommodate multiple – for example, four – wavelengths, a feature not often associated with <a adhocenable="false" href="">multimode fiber</a>. The Prysmian source said the company plans to present some of the relevant specifications to the TIA in the next couple of weeks with an eye toward promoting what might form the basis of OM5.<br> <br> Infinera has just recently begun to discuss <a adhocenable="false" href="">a terabit photonic integrated circuit</a> (PIC) as a follow on to its current 500G PIC. Researchers at the company have already begun work on what comes after that. Again in the poster sessions, Infinera presented performance results of a 2.25-Tbps transmit PIC. The gentleman fielding questions in front of the poster said he had no idea when such a device might reach commercialization, pointing out that it would have to be paired with a receive PIC.<br> <br> Teraxion touted a prototype silicon-photonics based modulator for PAM4 transmission that was the subject of a conference paper.<br> <br> A round-up of the more salient test and measurement advances can be found in today’s <a adhocenable="false" href="" target="_blank">Lightwave Test &amp; Measurement Direct newsletter</a> – as well as on our homepage.<br> <br> Brandon Collings, CTO within the CCOP group of JDSU that will spin out next year, said that the &quot;silicon photonics partner&quot; <a adhocenable="false" href="">Alan Lowe mentioned</a> as aiding the company in its development of a new datacom product was providing fab services; the design expertise is internal, much as is the case with Finisar’s 50G/40G Ethernet demonstration mentioned in <a adhocenable="false" href="">the Day 2 Notebook</a>. Regarding the ongoing 4x100G versus 8x50G debate for 400 Gigabit Ethernet, Brandon noted that generally, the fewer lasers involved, the cheaper the design. Also, the company’s work on an analog CFP2 continues apace.<br> <br> The JDSU Network and Services Enablement (i.e., test and measurement) staffers with whom I spoke all believe that the JDSU brand will go with the CCOP spin out.<br> <br> Just about everyone with a toe in the space expects the CWDM4 and CLR4 Alliance mid-range 100 Gigabit Ethernet MSAs to merge. There are some sensitivity specification differences as well as differing emphasis on the use of FEC (it’s mandatory in the CWDM4 specs and optional in the CLR4 specs). None of my source thought this would be a major issue. For this reason, companies involved in both MSAs appeared willing to discuss when they’ll have CWDM4 transceivers in the market (at some point next year), but shrugged their shoulders about CLR4.<br> <br> Karen Liu, who recently jumped to the VP of marketing role at Kaiam after many years as an analyst at Ovum, declined to offer any more specifics on the two new strategic investors that <a adhocenable="false" href="">led the company’s recent funding round</a>. But she confirmed that the company’s emphasis on transceiver modules (as well as the PLC business) rather than TOSA/ROSAs was going well enough to require the facilities expansion mentioned in the company’s funding announcement. These include 40 Gigabit Ethernet modules, with 100 Gigabit Ethernet QSFP28 devices now under development. She said it was likely that the company will have interest in the 400 Gigabit Ethernet space, given Kaiam’s expertise in multi-channel transmission.</p> <p><b>UPDATE:</b> Despite the fact that it demonstrated a coherent CFP reference design with ClariPhy at OFC 2013 (see <a adhocenable="false" href="">&quot;ClariPhy, Sumitomo Electric pair to demo 100G coherent CFP technology&quot;</a>), Sumitomo Electric has made more progress on an analog coherent CFP2. While I got the impression at the show (and therefore reported) that a coherent CFP is no longer in the cards, I've subsequently been told that work on such a module remains ongoing, as does the relationship with ClariPhy.<br> </p> <p><b>Jump to <a href="" adhocenable="false">Day 1</a> or <a adhocenable="false" href="">Day 2</a>.<br> </b><br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2014/09/ecoc-reporters-notebook-day-3.html2014-09-24T20:04:00.000Z2014-10-03T21:45:35.051ZECOC 2014 Reporter's Notebook: Day 2noemail@noemail.orgStephen Hardy<p>Now that NEL has finally announced that <a adhocenable="false" href="">its second-generation coherent DSP chip</a> is sampling, the competition with Clariphy Communications for next-gen design slots should become interesting. ClariPhy is first to market with <a adhocenable="false" href="">a second generation device</a> and also supports 200G, which the NEL device as currently described does not. (This discrepancy apparently has caused some consternation among NEL’s sales staff, according to one transceiver vendor.) NEL’s chip might have some power dissipation advantages, based on the fact that the device will use 20-nm technology from Broadcom versus ClariPhy’s 28-nm approach. The company also has an impressive roster of customers that used its first generation device.</p> <p>Early on, it would appear that it will come down to how well Clariphy can translate 200G and first availability into design wins. (Civcom and Oclaro have already aligned themselves with ClariPhy’s chip.) I spoke to Ovum’s Daryl Innis about this in the press room this morning. He doesn’t see 200G as a major factor in the market – it will be deployed, but not ubiquitously, he said. I plan to quiz him on this in more detail tomorrow when I record another installment of Lightwave’s 100G technology podcast series, so keep an eye peeled for that.<br> <br> Meanwhile, specifications for 400 Gigabit Ethernet remain in a very nascent stage, but that hasn’t prevented companies from showing off related technologies. Xilinx has partnered with Sumitomo Electric to show off a Virtex UltraScale FPGA configured as a 400 Gigabit Ethernet MAC/PCS device driving four Sumitomo Electric CFP4 100G optical transceivers. Sources at Xilinx say that the current generation of the FPGA can support up to 32 lanes of 25 Gbps, which makes it applicable in designs based on the anticipated 16x25-Gbps electrical lane layout. A new generation of the FPGA is on its way that will support 50-Gbps lanes, which is another approach expected to be part of the 400 Gigabit Ethernet standards.<br> <br> The current leader as far as a 400 Gigabit Ethernet optical module is concerned is the CDFP. Members of the CDFP MSA this morning announced release the 2.0 version of the specifications. At the OIF’s interoperability demo, two versions of the CDFP (&quot;Style 1&quot; and &quot;Style 2&quot;) were discussed. Style 2 is a bit longer than Style 1. The shorter version may prove more popular in copper-based applications, I was told. Sources within the MSA also said they expect the CDFP to be popular at short to medium distances – but that another form factor likely will be necessary to tackle the requirements of longer-reach optics. Discussions have already started among interested vendors about what this module might look like; creation of another MSA is a distinct possibility, I was told.</p> <p style="text-align: center;">*******</p> <p>Okay, about that Finisar/silicon photonics item I touted in today's newsletter: Here it is. Finisar is investigating the potential wonders of silicon photonics. The company is showing a &quot;technology demonstration&quot; that includes a CFP4 transmitting 40 Gbps over 12 km of fiber using a CW DFB laser at 1310 nm supported by a silicon photonics modulator. On the receive end is a silicon photonics/germanium photodetector. The company is also showing a 50-Gbps transceiver using similar technology.</p> <p>According to VP of Marketing Rafik Ward, the silicon photonics technology is an in-house development; the company did not link with an outside partner for the expertise.</p> <p>Ward says that the company isn't discussing commercialization potential for the technology. The demonstration appears to have two main goals: demonstrate that Finisar is hip to silicon photonics (the company's stock took a major hit last year when an analyst downgraded the company because of the perceived threat silicon photonics savvy competitors posed; see <a adhocenable="false" href="">&quot;Silicon photonics, one year later&quot;</a>) and promote the use of 50-Gbps technology using NRZ modulation. The 50-Gbps technology is part of a major debate raging within the 400 Gigabit Ethernet Task Force regarding the right technology for its various PMDs. Finisar and others (Huawei among them, say sources) are pushing an 8x50-Gbps approach for 400 Gigabit Ethernet in competition with proponents of 4x100G, which would have to leverage some sort of multilevel modulation format such as PAM4.<br> </p> <p><b>Jump to <a href="" adhocenable="false">Day 1</a> or <a href="" adhocenable="false">Day 3</a>.<a href="" adhocenable="false"></a></b><br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2014/09/ecoc-reporters-notebook-day-2.html2014-09-23T13:00:00.000Z2014-10-03T21:45:44.967ZECOC 2014 Reporter’s Notebook: Day 1noemail@noemail.orgStephen Hardy<p>The weather is warm and attendees are eyeing enviously the beachgoers they must pass as they enter the Palais des Festivals in Cannes on the first day of the 2014 European Conference on Optical Communications (ECOC). Show organizers hope that the conference and the activities on the show floor will at least temporarily distract the optical communications community from the sun and sand.<br> <br> As is now customary, I will do my best to provide a running tabulation of some of the more interesting goings on. So please check back regularly<br> <br> For example:<br> <br> News of OneChip Photonics’ demise looks to be premature. The company, which focuses on InP as a method for photonic integration as an alternative to silicon photonics (see, for example, <a adhocenable="false" href="">&quot;OneChip Photonics offers PIC-based 100G components for data center interconnects&quot;</a>), has been in the process of liquidating its assets for the last several months. However, CEO Michael Lebby, the only full-time employee left, revealed at a workshop Sunday that a Canadian investor has purchased OneChip’s IP with the goal of restarting the company. Lebby is now tasked with reassembling a team and figuring out which market to pursue.<br> <br> Sources on the show floor suggest that consensus is building around at least some aspects of the IEEE’s approach to 400 Gigabit Ethernet, at least when it comes to two of the target PMDs. While stressing that nothing is close to being decided yet, the sources suggest there is momentum behind a 4x100-Gbps approach to the 500-m requirement using parallel singlemode fiber – something along the lines of the current PSM4 approach now the subject of a 100G MSA. Meanwhile, for 10 km, 8x50G is gaining popularity, according to the sources. On the electrical side of the equation, both 16x25G and 8x50G have the early look of winning approaches, the sources add.<br> <br> While there’s clearly interest in coming up with a way to transmit 100G in the data center using a single wavelength (for 400G clearly, but certainly 100G applications would benefit as well), Finisar’s Chris Cole told attendees at a Market Watch session that technical hurdles remain to be resolved before such an approach becomes practical. At the current state of the art, power requirements in particular for the necessary DSP, FEC, high-power transmitter, etc., would prevent the support of single-wavelength 100G in the currently popular small form factor optical transceiver packages, he pointed out. For 100G, Cole said that 2x50G likely would be the next hot technology. Perhaps not coincidentally, Finisar is demonstrating 50G transmission in its booth via an approach that leverages silicon photonics technology.<br> <br> <b>Jump to <a href="" adhocenable="false">the Day 2 blog</a> or <a adhocenable="false" href="">Day 3</a></b>.<br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2014/09/ecoc-2014-reporters-notebook-day-1.html2014-09-22T11:30:00.000Z2014-10-03T21:45:54.738ZWe feel prettynoemail@noemail.orgStephen Hardy<p>So, yeah, we’ve redesigned the website. Most of the changes are cosmetic, but there is at least one significant new innovation.<br> <br> The cosmetic touches should be readily apparent. First and foremost you’ll notice that, after varying pronouncements to the contrary in many corners of the fashion, design, and TV worlds, black is indeed the new black. We think the presentation, particularly on the article pages, is pretty clean that way.<br> <br> We’ve also moved more of our article links to the top of the page and decided to highlight our currently most significant story. In a bout of unbridled hubris, I agreed to come up with artwork to accompany this story. I’ve only had to do this once so far, and I’m already wondering what I was thinking. (No, I have no money to hire you to provide artwork for the site – but thanks for the offer. And yes, I’m building a library of those wonderful shots of buildings and signs adorned with company logos.)<br> <br> Those of you into website design are undoubtedly wondering where the content rotator/carousel is. Rotators are those widgets usually located at the top of the page that automatically slide images across the screen. They were the hot design trend for most of last year. However, my research uncovered a huge <a adhocenable="false" href="" target="_blank">backlash</a> against them over the past 8-10 months, with numerous alleged experts asserting they weren’t SEO friendly, didn’t perform well as click generators, and generally hogged a lot of space that could be better devoted to other purposes. Rotators were generally useless, these experts proclaimed – <a adhocenable="false" href="" target="_blank">except, maybe, for media sites</a>. So, bolstered by that definitely don’t/maybe you should advice, we’ll be adding one in the homepage’s middle column in place of the Lightwave Optical Innovation Summit video player later this month. (You are <a target="_blank" href="" adhocenable="false">coming to the Summit</a>, right?) There are also a few smaller ones already for our other video resources.<br> <br> But about the substantive change. The site now enjoys a “responsive design.” That means that in addition to accelerating smartly and handling the tightest of curves, the site will automatically reshape itself to best fit whatever device you’re using to access it. It will arrange itself neatly into a single column if you’re viewing it from a smartphone, provide a larger format if you’re using a tablet, and look like a standard three-column website on your desktop or laptop computer. We’ll implement the same sort of responsive design principles in our newsletters beginning August 1.<br> <br> We have a couple of other things up our sleeves that we’ll roll out in the near future. Until then, I hope you enjoy the new look and feel. Comments, as always, are welcomed.<br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2014/07/we-feel-pretty.html2014-07-02T20:30:00.000Z2014-07-02T22:49:22.394ZNetflix: U-verse stinks and so do interconnect agreementsnoemail@noemail.orgStephen Hardy<p>The folks at Netflix naturally have an interest in the level of broadband services available to their potential customers; after all, the faster their over-the-top (OTT) video content arrives on their subscribers’ screens, the more likely those subscribers will remain loyal. So Netflix tracks the download speeds of the various Internet service providers and <a adhocenable="false" href="" target="_blank">posts the results</a>. And it’s not very impressed with AT&amp;T’s U-verse.<br> <br> “The surprising news is that AT&amp;T fiber-based U-verse has lower performance than many DSL ISPs, such as Frontier, CenturyLink &amp; Windstream,” wrote Netflix CEO Reed Hastings and CFO David Wells in a <a adhocenable="false" href="" target="_blank">letter to shareholders</a> issued in conjunction with the company’s first quarter earnings statement – and shortly after AT&amp;T took Netflix to task for the latter’s support of net neutrality.<br> <br> The executives made it clear that their criticism of U-verse was part of the larger net neutrality conversation. “This reinforces our view that connectivity to the broader Internet is critical to the quality of experience consumers receive,” Hastings and Wells continued. “The 249 customer comments on <a adhocenable="false" href="" target="_blank">AT&amp;T’s anti-Netflix blog</a> post indicate that AT&amp;T customers expect a good quality Netflix experience given how much they pay AT&amp;T for their Internet service. It is free and easy for AT&amp;T to interconnect directly with Netflix and quickly improve their customers’ experience, should AT&amp;T so desire.”<br> <br> While Netflix states that it’s “free and easy” for AT&amp;T to create such direct interconnections, the OTT video service provider agreed last February to pay Comcast for the same sort of links. The deal apparently has produced immediate results; “Comcast is providing a much improved Netflix experience to their broadband subscribers,” Hastings and Wells reported.<br> <br> But that doesn’t mean Netflix is happy with its Comcast arrangement. In the same letter, Hastings and Wells appear to bemoan their new Comcast pact while criticizing the proposed <a adhocenable="false" href="">Comcast/Time Warner Cable merger</a>.<br> <br> “Comcast is already dominant enough to be able to capture unprecedented fees from transit providers and services such as Netflix. The combined company would possess even more anti-competitive leverage to charge arbitrary interconnection tolls for access to their customers. For this reason, Netflix opposes this merger,” the executives wrote.<br> <br> The issue at hand is that Netflix is wrong – the interconnection it seeks from AT&amp;T isn’t free. Netflix just doesn’t want to have to pay for it.<br> <br> And at that’s what will make the net neutrality debate so sticky. For many of those directly affected it’s not about Internet freedom – it’s about who is going to pay. Or, more precisely, who will pass more costs to consumers, the ISP or the OTT service provider. Because it’s the end user who will end up footing much of the bill regardless of what resolution the net neutrality debate reaches.<br> <br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2014/04/netflix-u-verse-stinks-and-so-do-interconnect-agreements.html2014-04-22T20:15:00.000Z2014-04-28T14:47:51.443ZOFC 2014 Reporter's Notebook - Day 3noemail@noemail.orgStephen Hardy<p>All right, so I didn't really write this on the third day of the show. But here's what happened that day.</p> <p>We'll start with what I teased in <a href="" adhocenable="false">the previous blog</a> -- JDSU's stance on pluggable coherent transceivers. As suspected, the company thinks the CFP2 is the right place to jump into the coherent optical module market, according to Brandon Collings, CTO of the CCOP business unit. The fact that the device doesn't contain an onboard DSP (at least in its analog variant) means you can sell it to anyone, including systems vendors who want to continue using their in-house developed DSPs. Since most 100G ports come from such companies, that means a more attractive market opportunity, Collings implied. The CFP2 modules should be available next year, Collings said.</p> <p>The corollary to CFP2 rationale is that JDSU has no interest in making CFP transceivers, which will have the DSP integrated. As was the case with the OIF MSA, JDSU will be happy to supply components to those who are willing to take a chance in this market, Collings said.</p> <p>And while I'm tying up loose ends, Civcom sources say that the DSPs for the line of coherent pluggables I mentioned in the Day 2 notebook are an in-house development. </p> <p>One company that also will supply components to coherent CFP vendors is ClariPhy, which was showing off its recently announced <a href="" adhocenable="false">LightSpeed-II DSP technology</a> at the W Hotel across the street from Moscone. Scott Gardner was particularly proud of the cycle slip mitigation capabilities of the new generation of coherent silicon devices. I asked the engineer overseeing the demo what the most common questions people had asked about. &quot;Performance and power,&quot; he replied. ClariPhy <a adhocenable="false" href="">is discussing performance publicly</a>, but not power.</p> <p>Xtera, which <a href="" adhocenable="false">likely has been using Acacia's OIF MSA transceiver</a> in its current generation of <a href="" adhocenable="false">DP-QPSK</a>-based 100G coherent systems, is readying 16-QAM capabilities for its systems. The company recently conducted field trials with Verizon in which it showed off 400G reach in the ballpark of typical 100G DP-QPSK systems, thanks to Xtera's Raman amplification technology. The company likely will make the 16-QAM technology available for 400G next year -- although could deliver it sooner if the market demanded, Xtera sources said.</p> <p>Huawei, meanwhile, was discussing its work in single-channel 400G transmission, using its &quot;Faster than Nyquist&quot; technology. The approach includes both enhancements to DSP and FEC technology on the receive side as well as digital shaping on the transmit side. The company also highlighted its development of a Flex Wavelength Router, which resurrects the concept of an optical switching fabric based on free-space optics using a liquid crystal on silicon (LCoS) approach. Systems based on the technology could be in field trials by the end of the year, said a company source.<br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2014/03/ofc-2014-reporters-notebook-day-3.html2014-03-17T03:15:00.000Z2014-03-17T17:24:43.099ZOFC 2014 Reporter's Notebook - Day 2noemail@noemail.orgStephen Hardy<p>Buzzing from the sugar rush generated by way too much birthday cake on top of Italian food, I think I'll start off by quashing a rumor that clearly was too juicy (and, let's face it, illogical) to be true. Cisco is not buying Finisar, or at least so says Finisar Executive Chairman Jerry Rawls. You can hear a few choice words on this subject and others from Jerry when we post the video interview I shot with him today. This will happen &quot;soon,&quot; I'm assured.<br> </p> <p>That out of the way, let's look at why it's been a big show for the CFPx module form factor family. On the line side, of course, I've already covered the <a adhocenable="false" href="">coherent CFP2 optical module</a> announcement from Oclaro and the coherent <a adhocenable="false" href="">CFP optical transceiver news from Acacia Communications</a>. Mr. Rawls says that Finisar's work on its coherent CFP2 is well underway and that we should expect to see first samples early next year. This includes the modulator that Finisar was working to develop with u2t (and that's &quot;u-squared-t&quot; the new Finisar employees remind us) before the former purchased the latter.</p> <p>NeoPhotonics announced an integrated coherent transmitter (ICT) for such pluggable optical modules, which gives them all of the main building blocks for a coherent analog CFP2.</p> <p>EMCORE also is displaying an ICT for CFP2 applications, with an eye toward building such a module themselves. The company's focus will be on leveraging its laser technology and likely will partner for the coherent receiver. A beta version of the ICT is due in August, so it doesn't seem likely that an EMCORE coherent CFP2 would be production ready this year.</p> <p>II-VI is showing off an integrated coherent receiver, but doesn't plan to build coherent pluggables, a source at the booth told me.</p> <p>Other companies I haven't had a chance to visit yet also are making noise. Civcom, for example, has announced both 100G and 200G coherent CFPs. The transceivers are based on what the company describes as &quot;an advanced line of DSPs&quot; whose origin I haven't yet discovered. [I thought I solved this on St. Patrick's Day, when a source at Padtec, which owns Civcom, said they're developing the DSPs themselves. However, Padtec CEO Jorge Salomao, in a conversation March 28, said that the modules will use ClariPhy's DSPs, <a adhocenable="false" href="">based on their recently announced agreement</a>.] We know where Sumitomo Electric is getting the DSPs for its coherent CFP -- <a adhocenable="false" href="">from ClariPhy</a>.</p> <p>The wild card here is JDSU. The company decided not to pursue the OIF MSA module market in favor of supplying 100G components. Will the fact that the company won't have to worry about integrating a DSP into an analog CFP2 mean they'll finally get into coherent modules? I'll ask when I see them tomorrow.</p> <p>One next-generation CFPx area we know JDSU is pursuing is the CFP4. The company is part of the CFP4 MSA, along with Avago Technologies, Finisar, Fujitsu Optical Components, Oclaro, and Sumitomo Electric Industries. CFP4s for data center applications are on display across the two show floors (which are either side of Howard Street, making for some spirited sprinting between appointments), including at JDSU.</p> <p>And, since the acronym has the right letters in the right order, I should mention that draft mechanical specs and a drawing have been issued for CDFP, the form factor MSA focused on 400G (check out <a adhocenable="false" target="_blank" href="">the MSA homepage</a>). Representatives for some of the MSA members tell me that yes, it would be nice if the IEEE Working Group for 400 Gigabit Ethernet (expected to be elevated to Task Force status at the upcoming IEEE meeting in Beijing) adopted the form factor as part of its efforts. However, there's plenty of demand for proprietary 400G links (particularly in the form of active optical cables) to make coming up with an MSA now a worthwhile endeavor.</p> <p>Miss the Day 1 blog? <a adhocenable="false" href="">It's here</a>.<br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2014/03/ofc-2014-reporters-notebook-day-2.html2014-03-13T01:30:00.000Z2014-03-28T18:40:55.210ZOFC 2014 Reporter's Notebook - Day 1noemail@noemail.orgStephen Hardy<p>So, is there any better way to celebrate your birthday than attending the first day of exhibits at OFC 2014? (Wait, don't answer that.) I'll once again be touching base here all week with tidbits from my conversations on the show floor, which will distract me from counting how many more gray hairs I have this year than last.</p> <p><b>9:00 AM:</b> There's apparently yet another effort underway to create at least one transceiver MSA based on WDM over singlemode fiber aimed at 1-2 km 100 Gigabit Ethernet applications. Christophe Metivier of Arista Networks discussed the need for such a device yesterday during the OSA's Executive Forum. This &quot;hole&quot; in the IEEE's 100GbE standardization efforts was the subject of the the Google-backed <a adhocenable="false" href="">10x10 MSA</a>, which started with much fanfare but doesn't seem to have made as much of an impact as its organizers had hoped (although NeoPhotonics, which emerged as the leading -- if perhaps not quite only -- provider of such devices will tell you they're selling just fine, thanks for asking). Asked why this new effort might gain more traction -- and optical transceiver vendor support -- than the 10X10 MSA, Metivier surmised that the 10x10G approach might have been ahead of its time, but there seems to be plenty of interest in the newly emerging MSA.</p> <p>Signs are arising that Metivier might be right. ColorChip announced a 100G QSFP28 for 1-2 km applications. Meanwhile, Mellanox and start-up RANOVUS have announced the OpenOptics MSA, which will aim at 2 km and above, so it least touches the upper end of the target reach requirement. The MSA calls for 4x25G WDM, also in a QSFP28 optical transceiver form factor.<br> </p> <p><b>9:55 PM:</b> Meanwhile, Kaiam also is interested in jumping into the 1-2 km CWDM-based transceiver fray. The company unveiled its first QSFP+ LR4 optical transceiver, for 40GbE, but has its sights clearly set on 100G. The company is showing of a 4x28G TOSA for such a transceiver in its booth. Kaiam President and CEO Bardia Pezeshki, who was with Santur when that company developed its 10x10G transceiver before being acquired by NeoPhotonics, says the lack of widespread support among module vendors for the MSA had as much to do with technological limitations -- in short, how to support 10 wavelengths of 10G and meet the MSA specifications -- as anything else. He's very bullish on the market prospects for a CWDM-based approach to the 1-2 km 100G application question.</p> <p>It should be pointed out that there will be alternatives to the impending CWDM-based MSA (or MSAs). An MSA also has sprung up around PSM4 (see <a adhocenable="false" href="">&quot;PSM4 MSA Group targets alternative for data center 100G&quot;</a> and <a adhocenable="false" href="">&quot;Factors driving PSM4 and silicon photonics for data center architectures&quot;</a>). However, this MSA targets reaches of around 500 m and may be limited to applications that lend themselves to active optical cables.</p> <p>Interestingly, Cisco has announced new versions of its CPAK optical transceiver -- one of which is 10x10G.</p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2014/03/ofc-2014-reporters-notebook-day-1.html2014-03-11T16:15:00.000Z2014-03-19T19:57:34.342ZInfinera CEO predicts systems-level consolidationnoemail@noemail.orgStephen Hardy<p>In addition to reviewing the performance of his company in 2013 and his expectations for its performance in 2014, Infinera (NASDAQ:INFN) CEO Thomas Fallon last week during the analyst call to review 4Q13 results (see <a href="" adhocenable="false">“Infinera expects a strong start to 2014”</a>) offered a few opinions about the optical network hardware space, which he believes will begin to get smaller.<br> <br> After suggesting that the momentum of such Chinese heavyweights as Huawei and ZTE has slowed, Fallon suggested that the rest of the <a href="" adhocenable="false">optical transport</a> systems vendor space will likely move toward two dominant players. Naturally, he expects Infinera will be one of them. Coriant likely won’t be the other, he suggested.<br> <br> “I applaud those guys’ courage to go trying to build this into an independent enterprise. I think that they have got a lot of work cut out for them,” Fallon told analysts on the call. “And I am not a believer that private equity companies are going to go into this, with a ‘Let’s lose a lot of money upfront getting a lot of footprint so we could make money later.’”<br> <br> It likely won’t be any of the “multinational companies” either. “I think that some of the large multinationals, you could only lose so much money in this business for so long before you have to take a step back and say what are we going to do differently. And I see that occurring and playing out over the next year.”<br> <br> So who does Fallon think will survive to compete with Infinera? “I think that Ciena. quite frankly, is doing well and I think that they have a good technology,” Fallon said. “I continue to believe that it's going to be more a two vendor world than the eight or nine vendor world it's been. And I think we are well positioned, as is Ciena, to be the leaders in that market.”<br> <br> So Ciena has that going for them, anyway.<br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2014/01/infinera-ceo-predicts-systems-level-consolidation.html2014-02-04T03:45:00.000Z2014-02-18T19:15:45.620ZDid Infinera lose Verizon 100G contract to Alcatel-Lucent?noemail@noemail.orgStephen Hardy<p>George C. Notter, equity analyst at Jeffries LLC, wrote in a company note to investors yesterday that his sources say Verizon (NYSE, NASDAQ:VZ) recently awarded a long-haul 100G contract to Alcatel-Lucent (Euronext Paris and NYSE: ALU) over Infinera (NASDAQ:INFN). If true – and so far no one at any of the three companies involved is talking – it’s a blow to Infinera, but not a mortal one.<br> <br> The award would make Alcatel-Lucent Verizon’s second supplier of long-haul 100G gear, alongside Ciena (NYSE:CIEN; see <a href="" adhocenable="false">“Verizon details 2013 100-Gbps deployments”</a>). Notter wrote that he was surprised at the news, as he had considered Infinera the favorites for the award based on his information that Verizon’s technical staff liked Infinera’s DTN-X. However, sources told him that Alcatel-Lucent CEO Michel Combs swooped in at the last minute to swing the deal in Alcatel-Lucent’s favor. Notter’s sources say that the price tag on the Alcatel-Lucent equipment was “well below market pricing” (quotes his) and that perhaps some sort of package deal with Alcatel-Lucent’s <a href="/content/lw/en/fttx/ftth-b.html" adhocenable="false">FTTH</a> gear, which Verizon is already using, may have been involved.<br> <br> “The anecdote serves as another reminder that the best technology doesn’t always win in the Communications Infrastructure space,” Notter wrote.<br> <br> Leaving aside Mr. Notter’s assessment of the relative merits of the DTN-X and Alcatel-Lucent’s 1830 Photonic Services Switch – or at least his impression of Verizon’s assessment – the carrier’s decision clearly isn’t good news for Infinera. There’s the money, of course (Notter estimates the deal would have been worth about $100 million annually over a two-year period), as well as the negative PR. Verizon bigwig Stuart Elby had spoken at the unveiling of the DTN-X, and Notter believes Infinera had tailored the DTN-X’s capabilities to Verizon’s requirements. If the company still couldn’t land the Verizon account, what does that say about the company’s long-term prospects?<br> <br> Infinera likely would respond that it already has Tier 1 customers for the DTN-X, such as <a href="" adhocenable="false">CenturyLink</a> and <a href="" adhocenable="false">Deutsche Telekom</a>. And while it likely will take a beating on Wall Street – its share price dropped 11.37% in trading yesterday once Notter’s note got around – another big win will likely redeem the company in the eyes of most investors.<br> <br> Meanwhile, a victory at Verizon would be a big step toward building the momentum Combs has been seeking for <a href="" adhocenable="false">The Shift Plan</a>, his strategy to put Alcatel-Lucent back on firm footing. One can certainly wonder whether selling gear at “well below market pricing” is a viable long-term strategy (assuming that’s what happened, of course), but the short-term cost of aggressive pricing would appear to have an immediate payback for the company in terms of its stature and its ability to engender confidence in its technology and future viability.<br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2014/01/did-infinera-lose-verizon-100g-contract-to-alcatel-lucent.html2014-01-22T14:30:00.000Z2014-02-18T19:15:36.429ZNotes from Alcatel-Lucent’s Technology Symposiumnoemail@noemail.orgStephen Hardy<p>Alcatel-Lucent (Euronext Paris and NYSE: ALU) late last week brought analysts and journalists from around the world to Basking Ridge, NJ, for the company’s annual Technology Symposium. The point of the exercise was to convince this motley crew that the company now was headed in the right direction with its new strategy, called The Shift Plan (see <a adhocenable="false" href="">“Alcatel-Lucent shifts and optical transport, FTTx shift with it”</a>), and that it has the technical know-how to reach the light it says it can see at the end of the tunnel.<br> <br> As previously reported, the event opened with a talk from CEO Michel Combes, who explained the plan and why he believes it’s the right way to go (see <a adhocenable="false" href="">“CEO Combes makes case for Alcatel-Lucent's shift”</a>). Two days of presentations, breakout sessions, and interviews followed, many of which pertained to optical communications. Here are a few notes on these elements:<br> </p> <p style="margin-left: 40px;">» Basil Alwan, president, IP routing and transport, told me that the company at one time looked seriously into combining its XRS 7750 core router with elements of the 1830 Photonic Service Switch to create a router with integrated cutting-edge <a adhocenable="false" href="">optical transport</a> capabilities. They ultimately rejected the idea, mainly because they decided that taking up a router slot with a single 100G interface wasn’t economical. This doesn’t mean the router doesn’t have integrated optical transport capabilities because it does – but they’re 10 Gbps. Alwan says the company has adopted an optical integration philosophy that he described as “<i>n</i>-1,” where <i>n</i> is the current state of the art in <a adhocenable="false" href="">optical transmission</a> that likely requires a full slot to support. So at some point 100G will move to inexpensive modules that would make sense to integrate into the router; but you’d still want a standalone optical transport system such as the 1830 PSS to handle 400G via slots that are less expensive than the router’s.</p> <p></p> <p style="margin-left: 40px;">» In general, Alwan isn’t a fan of label switch routers with integrated optical transport, either.</p> <p></p> <p style="margin-left: 40px;">» Software-defined networking (SDN) will be the path to convergence of the IP and optical layers, Alwan believes. And while talk of “white-box switches” in the data center likely will lead to discussions and creation of “white-box optical systems,” he doesn’t think such products will be used ubiquitously. Sam Bucci, vice president and general manager, IP Transport Division, added that he doubted such white-box systems would work well in multivendor environments and at carrier scale.</p> <p></p> <p style="margin-left: 40px;">» Speaking of inexpensive 100G interfaces, Bucci said that as silicon photonics advances, Alcatel-Lucent may be among the first to come up with a module smaller than the OIF MSA for applications such as the metro. However, as the industry debates whether or not coherent CFP2 devices should have the DSP onboard, Bucci said purchasing such integrated CFP2 modules from an outside vendor probably wouldn’t make sense for Alcatel-Lucent. Noting that a digital CFP2 would represent a significant integration challenge, given the amount of supporting electronics the DSP requires, Bucci said Alcatel-Lucent favors the continued use of its in-house DSP, in part to ensure interoperability with fielded Alcatel-Lucent systems.<br> </p> <p></p> <p style="margin-left: 40px;">» Alcatel-Lucent supports direct-detect 100G as part of an encrypted metro transport offering. However, Bucci said that the cost difference between direct detect and coherent is dwindling rapidly and should be negligible by the end of next year.</p> <p></p> <p style="margin-left: 40px;">» In the access space, Alcatel-Lucent sources said they expect to have TWDM-PON systems available in 2015 – right around the same time they commercialize technology for FTTN/FTTC networks. The company touts as the road to 1 Gbps over copper, even though it doesn’t appear such speeds will be covered in the upcoming ITU-T standard, which should be ratified next year. Right now, Alcatel-Lucent has managed 1.3 Gbps over 70 m. However, putting more than one signal on the same cable significantly reduces performance; <a adhocenable="false" href="">in recent tests with Telekom Austria</a>, the company could only manage 500 Mbps over 100 m in such conditions, even with vectoring applied. Nevertheless, Frederico Guillen, president, Fixed Networks Division, pointed out that VDSL2 with vectoring took a while to evolve to its current performance level. He expressed confidence that would meet carriers’ 1-Gbps requirements by the time it’s available and needed.<br> <br> </p> <p></p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2013/11/notes-from-alcatel-lucents-technology-symposium.html2013-11-22T19:00:00.000Z2013-12-04T17:07:03.840ZECOC Reporter's Notebook: Day 3noemail@noemail.orgStephen Hardy<p><b>6:10 PM:</b> Andrew Schmitt isn't going to like this one, either. But my computer is almost out of power, I've been kicked out of the press room because ECOC is over for the day, and I have a newsletter that's two hours overdue. So I'm once again going to merely whet your appetite for the moment with a single entry, then circle back later to fill you in on more doings. So check back later, or check out the <a adhocenable="false" href="">Day 1</a> or <a adhocenable="false" href="">Day 2</a> notebooks, the latter of which has been updated to Mr. Schmitt's satisfaction.</p> <p>Anyway...</p> <p>My last meeting of the show, held here in the food court, was with Raj Shanmugaraj, president and CEO of coherent module vendor Acacia Communications. The company issued what in my opinion was the most interesting announcement at the show, the fact that they've achieved &quot;a mode of interoperability&quot; between their DSP and that of NEL. Interoperability among coherent transmission offerings has been difficult to achieve because each vendor has their own way of implementing coherent DP-QPSK. This hasn't been a significant drawback initially, but as the technology matures and 100G becomes more ubiquitous, interoperability will become more greatly desired -- particularly, I think, in the metro environment and, Raj believes, once people want to ensure that routers from different vendors can talk to each other via 100G coherent links.</p> <p>Raj revealed that what Acacia and NEL have done is create a way for their respective DSPs to interoperate in the ZR mode. This is achieved by the systems vendor via a software command. The devices in question are that inside Acacia's newly announced coherent CFP and the DSP NEL is co-developing with Broadcom (see <a adhocenable="false" href="">&quot;NTT Electronics, Broadcom pair for coherent DSP&quot;</a>). Acacia's DSP will reach the market once its CFP does; it's scheduled to begin sampling. Raj declined to share his opinion on when the NEL/Broadcom DSP will be released...</p> <p>Did Cisco make a big announcement while I was here in London?...</p> <p><b>8:30 PM:</b> Why, as a matter of fact, yes they did. The new Network Convergence System (NCS) line is intende to address the &quot;Internet of Everything&quot; future that Cisco envisions. There are three platforms in the NCS portfolio -- the NCS 6000, NCS 4000, and NCS 2000. The first two systems are large-scale routers with integrated optical transport capabilities; the third system is an optical transport/intelligent ROADM platform. The systems also are noteworthy in that they're the first platforms to ship (at least in the case of the NCS 6000 and NCS 2000; the NCS 4000 doesn't ship until the first half of next year) with Cisco's CPAK modules. I met Cisco's Andrew Carter at the OIF's interop booth, where, among other things, one could see the CPAK interoperating with CFP2 modules. Carter said that the CPAK will be available in LR4 and SR10 options initially; the company also has its eyes on ER4...</p> <p>Speaking of the OIF interop demo, it would appear that the technology derived from the CEI-28G-VSR and CEI-25G-LR specifications has reached the level of maturity necessary for adoption into working designs. As mentioned above, I saw CFP2 modules from the likes of Finisar and Fujitsu interoperating with each other and the aforementioned CPAK, driven by gearbox chips and aided by retimers from the likes of MoSys, AppliedMicro, Semtech, Inphi, and Xilinx (the latter, of course, in the for of an FPGA). Several backplane vendors also demonstrated their compatibility with the CEI-25G-LR specifications. There also was a demonstration of the heat sink issues related to the CDFP 400G module MSA that Molex and TE Connectivity have created. We'll have a podcast covering the demo up later today...</p> <p>Speaking of Inphi, VP of Marketing, High Speed Connectivity Products Siddarth Sheth agreed with the despairing whispers I had heard regarding the prospects for a new 500 m to 2 km PMD for 100GbE coming out of the Next Generation High-Speed Ethernet effort. It probably ain't happening, is Siddarth's understanding, due to the inability to achieve consensus around the question of whether to use a multi-level modulation format or a parallel approach like PMS4. The string of test and measurement vendors who have recently released PAM-4 test gear should not lose heart, however; multilevel modulation will likely be taken up again during the 400GbE effort. Technology used for that effort might end up being applied to 100G in the future, Sheth theorized...</p> <p>&nbsp;</p> <p>&nbsp;</p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2013/09/ecoc-2013-reporters-notebook-day-3.html2013-09-25T17:15:00.000Z2013-10-03T06:08:21.518ZECOC 2013 Reporter's Notebook: Day 2noemail@noemail.orgStephen Hardy<p><b>1:15 PM:</b> Before we get to today, here are a few odds and ends left over from yesterday. (And for more on yesterday, see <a href="" adhocenable="false">yesterday's ECOC 2013 Reporter's Notebook</a>.)</p> <p>The folks at Multiphy are happy that Oplink has announced <a adhocenable="false" href="">a 4x28G module for direct-detect metro 100G</a> applications for more than the obvious reason that the module uses their chipset. With Finisar pretty much the only supplier of 4x28G metro 100G modules to date (the folks at Multiphy say that they previously had been the only, but I haven't confirmed that yet with Oclaro, who also announced such a product), there's now a viable second source and therefore more of a reason for systems houses to deploy the technology. So far, ADVA Optical Networking is the only company to commercially deploy such a system. However, Multiphy&nbsp;CEO Avi Shabtai and VP Sales and Marketing Neal Neslusan told me that they're working with both module and systems vendors, and there should be competition for ADVA right around the corner. A new generation of devices for such direct-detect applications is in the works as well. Meanwhile, work continues at Multiphy <a adhocenable="false" href="">on a coherent ASIC</a>, but that appears to be going at a more leasurely pace...<br> </p> <p><b>11:50 PM:</b> Having been roundly chastized by Andrew Schmitt of Infonetics Research for the paucity of information in this blog, I herewith sacrifice much needed sleep to meet the information needs of my loyal audience -- or, at least, Andrew.</p> <p>As mentioned elsewhere on the site, Finisar is conducting what is the first public demonstraton to my knowledge of a <a adhocenable="false" href="">CFP4 optical transceiver</a>. The device uses Finisar's 25G VCSEL technology, of which VP of Marketing Rafik Ward is quite proud. He expects the device, labeled the FTLC9141, to reach the sampling stage by the end of this year, with production in the second half of 2014. You can hear more about this device, as well as Rafik's thoughts on the market for CFP2 devices and 4x28G metro direct-detect 100G transceivers, <a adhocenable="false" href="">in the podcast we posted earlier today</a>...</p> <p>While NeoPhotonics has been content to supply components only for the coherent market, the advent of the CFP2 -- which means the DSP is someone else's concern -- makes the development and marketing of a coherent transceiver from that company a likely scenario, according to the impression I received upon visiting the company's stand. One question is whether the company will offer a device with the ability to support both 100G and 200G. The rub is not whether they can do it from a technical perspective; it's whether they can do it in a way that meets price expecations. Those expectations will be &quot;at no extra charge,&quot; most likely...<br> </p> <p>The post-deadline papers were, er, posted late yesterday. One that caught my eye is a presentation from Intel and Corning regarding use of silicon photonics to transmit 25 Gbps over 820 m of a newly developed multimode fiber optimized for 1310 nm. Does this mean that we're going to see a new class of fiber specifically designed for data center applications using silicon photonics? Silicon photonics devices traditionally have been associated with singlemode fiber...</p> <p>Stopped by the Oclaro meeting room and spoke with Per Hansen, who could become an employee of II-VI if that company picks up its option to buy Oclaro's amplifier business. We talked briefly about 4x28G metro transceivers -- a niche Oclaro decided to exit after announcing a product at last year's OFC/NFOEC because that niche just wasn't big enough to be worth pursuing, in Oclaro's opinion -- as well as the fact that CDC <a adhocenable="false" href="/content/lw/en/network-design/dwdm-roadm.html">ROADMs</a> were a big deal at last year's ECOC, but not so much of an emphasis this year. Per said that the market had pretty much settled on requirements for these applications, including flexible-grid WSS capabilities. Of course, Oclaro's CEO Greg Dougherty recently revealed that <a adhocenable="false" href="">Oclaro won't produce such WSS modules</a>, which leads to the question of how Oclaro plans to continue playing in the next-gen ROADM space. Collaboration is the answer, Per said; Oclaro has been pursuing this strategy for a while in such applications and will continue along this path going forward. He said that Oclaro doesn't have a preferred partner, mainly because their customers have their own preferences. And those preferences right now are all over the map...</p> <p>Stefan Voll, head of product management, DWDM at Coriant, reviewed with me the field trial of <a href="" adhocenable="false">400G via space-division multiplexing (SDM)</a> the company conducted with Austria's A1. Voll said that the fact that Marlin Equity Partners would support Coriant's research into a technology like SDM that probably won't be ready for commercialization for another 10 years should lay to rest the notion that Marlin is dipping into optical communications (and Coriant specifically) looking to make a quick buck before they dash off to some other market. (For the record, Coriant was working on this technology as part of the MODE-GAP project before Marlin bought them. But certainly Marlin hasn't told them to &quot;cut that out.&quot;) Meanwhile, Voll said that you shouldn't expect to see Coriant come out with a conventional 400G offering based on 16QAM until the enabling technology supports reaches and price points more in line with current 100G systems. A new generation of integrated building blocks -- optical certainly but mainly electrical -- will be needed before 400G can reach these performance levels, he said...</p> <p>And if you want the last word(s) on the subject, please feel free to continue to <a adhocenable="false" href="">Day 3 of the ECOC Reporter's Notebook</a>. </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2013/09/ecoc-2013-reporters-notebook-day-2.html2013-09-24T12:30:00.000Z2013-10-03T06:07:55.088ZECOC 2013 Reporter's Notebook: Day 1noemail@noemail.orgStephen Hardy<p>ECOC 2013 ramped up today after a day of workshops yesterday. London is warm and humid to start the week, and activity within the ExCEL venue is warming up as well (insert your own comment about optical communications and humidity here).</p> <p>Since I plan to jot notes here as the opportunity arises, I'll put time stamps on them so that if you check in here throughout the day, you'll know if anything new has been added.</p> <p><b>2:15 PM:</b> The folks formerly at Kotura are settling in at Mellanox, according to Arlon Martin. The acquisition closed only a few weeks ago, so it's a bit early to say anything about new products, he said...</p> <p>Sumitomo Electric (along with just about every other module vendor on the show floor) is highlighting its 100G CFP2/CFP4 road map. The company says it plans to sample a 100GBase-SR10 CFP by the end of the year. Meanwhile, an ER4 CFP optical transceiver should reach the sampling stage at the end of next year. The company also is touting work on CFP4 devices. A 100GBase-LR4 using a new four-channel TOSA and ROSA with integrated optical mux/demux capabilities and a target power consumptioni of less than 6 W is due for sampling during the second half of next year...</p> <p>Kaiam President and CEO Bardia Pezeshki reports things are going swimmingly at the company, particularly as it focuses on high-speed pluggable devices, such as QSFPs for the data center market. The company has partnered with Oclaro and Sumitomo Electric and is enjoying the benefits of having acquired a fab in Livingstone, Scotland along with other assets from Gemfire (see <a adhocenable="false" href="">&quot;Kaiam to acquire Gemfire assets&quot;</a>). Pezeshki believes that Kaiam's hybrid MEMS-based approach to integration can hold its own with silicon photonics and therefore doesn't see further development of that technology as a threat to his business. In fact, given the opportunity size of the data center market and the company's current level of success, Pezeshki said that an IPO in a few years might become something worth considering...</p> <p>Having unveiled test capabilities for CFP2 modules a while ago (see <a adhocenable="false" href="">''JDSU targets 100 Gbps at ECOC 2012&quot;</a>), Dr. Paul Brooks of <a target="_blank" href="" adhocenable="false">JDSU's</a> Network and Service Enablement business unit (that's the new name for CommTest) says that the complexity of developing such modules (not to mention the enabling technology) has meant CFP2 test capabilities demand has only recently begun to ramp. But that demand has arrived, he reports -- as has a requirement for CFP4 testing capabilities. Since the input/output for both CFP2 and CFP4 are about the same, JDSU is already in position to meet CFP4 test requirements, Brooks reports...</p> <p>Meanwhile, Brandon Collings, CTO within JDSU's CCOP business unit, says that one can expect to see a coherent CFP2 from his company. He wouldn't say when, of course, but the move is significant since JDSU has so far sat out of the coherent module market, preferring to focus on supplying components for such applications. I also cornered Collings for our &quot;100G at ECOC 2013&quot; podcast series, the results of which will be posted soon (<a href="" adhocenable="false">as in now</a>)...</p> <p><b>Rumor Alert:</b> Rumor has it that OneChip Photonics has a new CEO. More on this when I track down the alleged new honcho, who was talking to Verizon's Glenn Wellbrock this morning outside of the exhibit hall.<br> </p> <p><b>7:15 PM:</b> <b>Rumor Confirmed.</b> Michael Lebby, a member of OneChip's board since 2008, confirmed that he has stepped into the role of CEO of the integrated InP components company. Lebby declined to discuss his plans on the record, but one can expect a formal announcement in the near future...</p> <p>Want more? Check out the blogs for <a adhocenable="false" href="">Day 2</a> and <a adhocenable="false" href="">Day 3</a>. </p> <p>&nbsp;</p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2013/09/ecoc-reporters-notebook-day-one.html2013-09-23T13:45:00.000Z2013-10-03T06:07:41.508ZNo national fiber broadband network for U.S.noemail@noemail.orgStephen Hardy<p>The folks at market research and analysis firm IHS (NYSE: IHS) recently made the assertion in a new report from their IHS Screen Digest Operator Multiplay Intelligence Service that Google is unlikely to take its Google Fiber effort nationwide. As was the case the last time someone took up this topic (see <a adhocenable="false" href="">“What is Google Fiber trying to prove?”</a>), I’m somewhat bemused that the idea even comes up. No one else has any designs on building a nationwide <a adhocenable="false" href="">fiber-optic broadband</a> network in the U.S., so why should Google?<br> <br> One thing that has become clear since I wrote the blog I just referenced is that Google does see its <a adhocenable="false" href="/content/lw/en/fttx/ftth-b.html">FTTH</a> efforts as a business, not an experiment; the company’s Milo Medin pointed this out at the OSA’s Executive Forum this past March and <a adhocenable="false" href="" target="_blank">again recently</a> at the Fiber to the Home Council’s Community Toolkit Conference in Kansas City. IHS paints Google as a broadband dilettante in comparison to the Tier 1 carriers and major <a adhocenable="false" href="/content/lw/en/cable-mso.html">cable MSOs</a>. I’m sure Google has no problem with this. When the company does expand beyond its current locales in metro Kansas City, KS/MO, Austin, TX, and Provo, UT, it likely will be to similar markets: places where fiber deployment is easy (or existing networks can be acquired on the cheap) and competition at the 1-Gbps level is minimal, while bandwidth demand will be plentiful. There aren’t a lot of places in the U.S. like that, but they’re out there. So Google will continue to cherry pick.<br> <br> The obvious reason Google won’t go nationwide – and neither will anyone else – is because a fiber deployment on the scale of a country the size of the U.S would be only slightly less expensive than colonizing Mars. (And, to be clear, I’m talking truly nationwide coverage here.) You can also add the fact that while there is competition between the cable MSOs and the Tier 1 telcos (although less of it in areas where Verizon is co-marketing with the local cable operator), there’s little competition within those two groups. Comcast won’t go head-to-head with Cox or Charter; Verizon and AT&amp;T overlap in a few areas, but not many.<br> <br> But cost is the main impediment. That’s why the cable operators will stick to enhancing DOCSIS, and AT&amp;T and CenturyLink will continue to rely mainly on fiber to the node (<a adhocenable="false" href="/content/lw/en/fttx/fttn-c.html">FTTN</a>). Even Verizon, the only U.S. Tier 1 telco to embrace FTTH, decided to curb its fiber roll outs – in part by exiting markets. (One was New Hampshire, where I live. Bitter, me? Of course not.)<br> <br> The fact that no service provider will step up to truly nationwide FTTH leaves those interested in upgrading the nation’s broadband infrastructure with a tough problem: creating a network of networks with uniformly high performance for all citizens, regardless of where they live. And if you want to add “at a reasonable cost and with a choice of service providers,” the problem becomes even more difficult.<br> <br> According to a recent analysis by Arthur D. Little, the current U.S. broadband/regulatory ecosystem likely won’t support nationwide fiber broadband access (see <a adhocenable="false" href="">“Study touts government/private partnership as likely best approach for national fiber plans”</a>). We’ve seen the Obama Administration attempt to follow one of the analysis firm’s favored approaches – graded government support of efforts led by private industry – in the form of the broadband stimulus program and the Connect America Fund. But while one can point to small successes via these efforts, they are unlikely to completely meet the challenge (and, because they’re media agnostic, they don’t mandate fiber use).<br> <br> One step forward Washington could take is to encourage a truly U.S. trait – the willingness to see something that needs to be done, and doing it ourselves. That means clearing the way for municipalities to build their own fiber networks. Of course, if you think incumbent service providers have worked feverishly against such activities at the state level, imagine what would happen if the question arose on Capitol Hill. I don’t foresee this topic coming up for debate anytime soon.<br> <br> All this means that, for the foreseeable future, the U.S. situation will likely stay more or less the same: very good service and choice in some areas, not so good in many others. Where state laws allow and funding becomes available, municipalities and rural telcos will move forward where the major incumbents will not. But one has to wonder how the FCC’s target of 100 Mbps to 100 million households by 2020 will be reached. And, even if it is, how many households will still be left behind.<br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2013/05/No-national-fiber-broadband-network-for-US.html2013-05-31T17:45:00.000Z2013-08-13T17:35:33.595ZHuaweI: We're not quitting U.S.noemail@noemail.orgStephen Hardy<p>Huawei would like you to know that while it has decided that it's unlikely to achieve rampant success in the U.S. carrier network equipment market, it's not pulling up stakes and leaving.</p> <p><a href="" adhocenable="false">As previously reported</a>, Eric Xu, Huawei executive vice president, told attendees at an analyst conference this week that &quot;we are not interested in the U.S. market anymore.” Several press outlets (a quick web search will uncover plenty of examples) took these comments to mean that, in the face of the perception that it is a security threat to the U.S., Huawei plans to take its ball and go home.</p> <p>The company is now telling media outlets such as <a adhocenable="false" href="" target="_blank">Forbes</a> and (for those of you who read Chinese) <a adhocenable="false" href="" target="_blank">The Beijing News</a> that this is not what the company meant to imply. Instead, Huawei points out, Mr. Xu was responding to a question about prospects for growth in network hardware sales among developed countries. There don't appear to be many such prospects here in the U.S., he meant to say.<br> </p> <p>That doesn't mean there aren't opportunities in other niches here in the States, however. In particular, observers believe the company will remain aggressive in the enterprise and smartphone markets here in the U.S.</p> <p>So adjust your expectations accordingly.</p> <p><br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2013/04/Huawei-were-not-quitting-us.html2013-04-25T21:45:00.000Z2013-10-03T06:11:43.518ZThe significance of CPAKnoemail@noemail.orgStephen Hardy<p>No announcement at OFC/NFOEC generated more discussion than Cisco’s (NASDAQ:CSCO) unveiling of its CPAK optical transceiver module (see <a adhocenable="false" href="">“Cisco unveils CPAK 100G silicon photonics-based optical transceiver”</a>). Reactions ranged from “It’s the end of the road for traditional <a adhocenable="false" href="">optical transceiver</a> vendors” to “So what – it’s not much different than a CFP2.” I, of course, have a reaction as well, which falls somewhere between the two extremes.<br> <br> Let’s get a few things off the table right away. The fact that the CPAK transceiver leverages silicon photonics is among the least significant things about it. Yes, silicon photonics holds great potential to reduce footprint, power consumption, and overall costs, and we are starting to see that potential realized. But other avenues toward photonic integration promise (and have started to deliver) similar benefits. Cisco could have taken a different path and arrived at the same destination, if perhaps not in the same timeframe.<br> <br> Surprisingly, the fact that CFP2 modules likely will offer similar performance isn’t as significant as it might appear, either. CPAK optical transceivers won’t have to compete for market share with CFP2 transceivers – except within Cisco. Bill Gartner, vice president and general manager of Cisco's High End Routing and Optical business unit, is convinced that the CPAK is superior to the CFP2. So that market share competition is over.<br> <br> So what’s so significant about CPAK? Principally the ramifications of the fact that Cisco has decided to build its own transceiver:</p> <p>1. Cisco has been able to dictate the direction of the data center transceiver market in the past because it represents such a huge share of volume. That volume is now not available to drive down prices of <a adhocenable="false" href="">CFP transceivers</a> and CFP2 modules – or help module vendors recoup their development investments. Cisco hopes that it will have both a performance and pricing edge over its competitors that use CFP and CFP2 modules. One CFP2 module vendor on the show floor reported that the company had already been approached by at least one of Cisco’s competitors who expressed concern about CFP2 pricing.</p> <p>2. That said, Cisco is taking a risk that these pricing and performance benefits are real. Certain steps toward realizing these benefits remain uncertain. One such step: how to manufacture the modules. Gartner admitted at OFC/NFOEC that the company still hadn’t settled this question, although he said that LightWire had a relationship with a contract manufacturer for transceiver production at the time Cisco acquired it. Gartner said he had been “approached by lots of people from all angles” about the module, so he’ll have plenty of options. He said he wants multiple manufacturing sources.</p> <p>3. Would Cisco sell the transceiver to others or allow someone else to put CPAK on the open market? Only if Cisco wasn’t seeing the cost benefits it had anticipated or the open market modules were inferior to the ones it is using, it seems to me.</p> <p>4. If this foray is a success, what comes next? Gartner said the company probably would develop at least one other module in an unspecified form factor. (He said he’d really like a 100G module in a QSFP form factor – if module suppliers came up with one, he’d happily use it.) He also said it would make sense to look at pairing the Lightwire technology with the coherent expertise gained via the CoreOptics acquisition, which would mean coherent transceivers.<br> </p> <p>The advent of CPAK is the most interesting event of the year so far in the module space. If you have a take on what it means for the market, leave a comment below.<br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2013/04/The-significance-of-CPAK.html2013-04-10T14:45:00.000Z2013-10-03T06:11:29.322ZHow to sell to Facebook (and Google and Microsoft)noemail@noemail.orgStephen Hardy<p>The Optoelectronics Industry Development Association (OIDA) held its third data center workshop March 17. A number of presenters discussed roadmaps and metrics for photonic development, avenues for photonic integration, and what optical technology developers need to do to meet the needs of intra-rack, rack-to-rack, and other data center network needs. But perhaps the most memorable lines came from the first panel, which featured speakers from Facebook, Google, and Microsoft.<br> <br> The fun began when the third speaker on the panel, Facebook data center network engineer Nathan Farrington, promised to tell the audience how to make something Facebook will buy. To this end, the agenda for his talk promised a wish list of technology. Sure enough midway through his presentation, he presented a list of requirements. Well, to be accurate, it was a list of requirement – a repetition of the single phrase “Looking for more bits per second for less capex and opex.”<br> <br> Chuckle, chuckle. But then after truly offering a wish list that included a universal transceiver based on no more than two fibers because “parallel fiber is lame, especially when you don’t even use all of them,” he offered several tips for developing technology for Facebook:<br> </p> <ol> <li>Facebook is now more open to non-standard technology as long as it solves a problem (a sentiment shared by others on the panel, particularly Ryohei Urata, member of the technical staff at Google, who said, “We’re defined by the laws of physics, not by standards.”).</li> <li>But before you sink a lot of NRE into that new technology, you better talk to them first. Facebook’s requirements change quickly, so it’s a good idea to get an idea of where the company thinks its requirements are heading.</li> <li>Along these lines, get a copy of their roadmap (which is available under NDA).</li> <li>Then, send them samples. But don’t send them in locked boxes, because “we like to poke around inside things.” (see #2 above). “If this scares you then go somewhere else,” he said. “Sometimes we have to go in and change something.”<br> </li> </ol> <p>Meanwhile, there wasn’t a lot of love shown for the <a adhocenable="false" href="">10x10G MSA</a> for 100-Gbps technology applications that both Google and Facebook helped establish. Asked whether he was satisfied with the 10x10 MSA transceivers or whether work in the 4x25G arena had caught up to Google’s requirements, Urata replied, “The 10x10 predates me, so it’s not my fault.” A 4x25G approach now seems like a better way to go, he indicated. Tom Issenhuth, optical network architect at Microsoft put the capstone on the appeal of the current generation of 10x10G CFP <a adhocenable="false" href="">optical transceivers</a> by saying, “It’s big, it’s hungry, it doesn’t work in the data center…In a <a adhocenable="false" href="">CFP</a>, it’s of no interest.”<br> <br> All in all, the workshop was a fine way to get your OFC/NFOEC rolling.<br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2013/03/How-to-sell-to-Facebook-and-Google-and-Microsoft.html2013-03-18T03:45:00.000Z2013-10-03T06:11:05.810ZSilicon photonics doesn’t scare Finisarnoemail@noemail.orgStephen Hardy<p>Amid the usual give-and-take during <a href="" adhocenable="false">yesterday’s analyst call</a>, Finisar Executive Chairman of the Board Jerry Rawls was asked for his views on the potential threat silicon photonics poses to his business. <a target="_blank" href="" adhocenable="false">Finisar’s</a> (NASDAQ:FNSR) shares tumbled when Jefferies &amp; Co. analyst James Kisner used the occasion of Intel’s silicon photonics update at the Open Compute Summit in January to downgrade Finisar’s stock.<br> <br> When it comes to silicon photonics, it would appear Mr. Rawls is not impressed. At least not yet.<br> <br> “Now with respect to silicon photonics, it is an interesting technology, but it is one that we haven't used so far in our products because it wasn't economical or it didn't provide competitive performance,” Rawls said in response to an analyst’s question about how much threat, if any, silicon photonics may present.<br> <br> “I don't think there's a technology that's in the market at this time,” he added in response to another question about whether a silicon photonic alternative might blunt the opportunities for the optical engine Finisar has developed for board-level communications. “We've seen a lot of discussions about technology demonstrations that might yield products in the next decade. But not -- most of them aren't in the market at this time, and there's clearly not much productization or commercialization.”<br> <br> One silicon photonics-based effort that appears to be nearing commercialization is Cisco’s CPAK 100-Gbps optical module. But, while he’s not happy a customer is developing its own <a href="" adhocenable="false">optical transceiver</a>, the technology behind it isn’t causing him to lose sleep. That’s because Rawls asserted Finisar’s CFP2 modules will deliver a line card density, power consumption level (less than 7 W), and performance that should be at least equal to the CPAK, based on what he knows about Cisco’s efforts.<br> <br> This is not to say that Finisar will never use silicon photonics technology. If customer demands require, the company will be onboard, Rawls asserted.<br> <br> “If that time comes, we will use one of the multiple foundries that produce and sell custom silicon photonic circuits, and there's at least a half a dozen of them today,” he said. “I don't look at it as a threat. I just look at it is another technology that we will likely eventually use in our product, but just haven't found yet that it was -- that it provided the right solution.”<br> <br> (All quotes courtesy of a transcript provided by <a target="_blank" href="" adhocenable="false">Seeking Alpha</a>.)<br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2013/03/Silicon-photonics-doesnt-scare-Finisar.html2013-03-08T19:30:00.000Z2013-10-03T06:10:51.696ZEU budget cuts to fragment European FTTH landscapenoemail@noemail.orgStephen Hardy<p>As if there weren’t enough items on the agenda, attendees at next week’s FTTH Council Europe FTTH Conference suddenly have something new to talk about: the likelihood that the European Union will not fund broadband infrastructure deployments over the next seven years (see <a href="" adhocenable="false">&quot;European Union slashes broadband funding&quot;</a>).<br> <br> The decision to reduce the items in Neelie Kroes’s toolkit over the next seven years to “a good scolding” not only diminishes her influence, but calls into question the will of Europe to extend <a href="" adhocenable="false">broadband access</a> to those not easily reached economically via current wireline technology. It particularly clouds the future of fiber to the home (<a href="/content/lw/en/fttx/ftth-b.html" adhocenable="false">FTTH</a>) as the medium of choice for super-fast broadband in the region, since <a href="" adhocenable="false">fiber-optic broadband</a> networks represent the most expensive option available.<br> <br> Assuming the Multiannual Financial Framework (MFF) remains as is (it still must be ratified by the European Parliament), the burden for meeting the Digital Agenda’s goals will fall squarely on each member-state. Given the varying levels of fiscal health among these countries, it seems likely that “broadband for all” may not be equally high on each country’s list of things to do.<br> <br> The first consequence will be fewer new initiatives. The subsequent consequence in some cases will be funding shortfalls for existing initiatives.<br> <br> Service providers therefore should expect to have more responsibility for meeting broadband goals thrust upon them. In turn, we can expect these service providers to whine even more loudly about government regulations as a disincentive to investment. Current programs will slow, and new ones will take longer to begin.<br> <br> And with funding likely more difficult to scrape together until the regional economy improves significantly, we’re likely to see more emphasis on “economical” broadband approaches. That means more fiber to the cabinet and less <a href="" adhocenable="false">fiber to the home</a>. It may also mean increased reliance on mobile broadband approaches.<br> <br> The European FTTH community knows what it’s up against. “By reducing the financial envelope down to €1 billion for telecommunications, there will be no room for the support of fibre infrastructure investments,” said FTTH Council Europe President Karin Ahl in response to the MFF.<br> <br> “Now that the EU Member States have failed to realize the importance of the digital part of the Connecting Europe Facility, the role of the FTTH Council Europe on the investment and financing side will be even more important – and we are looking forward to face this challenge,” she concluded<br> <br> The council had better be up for the challenge. The battle for fiber-based broadband in Europe just took a turn for the worse.<br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2013/02/EU-budget-cuts-to-fragment-European-FTTH-landscape.html2013-02-11T21:04:00.000Z2013-02-26T22:19:34.940ZThe secrets I’ve been keepingnoemail@noemail.orgStephen Hardy<p>One of the great things about being the editor of <i>Lightwave </i>is that I often hear about things that aren’t in a press release that I think you’d want to know as well. So while I’m taking written notes, I usually make a mental note to myself to whip together a story about this jewel I’ve uncovered the next time I’m in front of a keyboard. Unfortunately, due no doubt to the many tasks competing for my time (and not the fact that I am now old enough to routinely forget my ATM password and the names of close relatives), those mental notes sometimes get put aside. Which means a year-end review of my notebooks results in frequent “Oh yeah – I was going to write about that” moments.<br> <br> So, in the hope that late is indeed better than never, here are a few things I stumbled across last year that I haven’t yet told you about:<br> </p> <ul> <li>• One <a href="">optical transceiver</a>/transponder vendor to keep an eye on in the 100-Gbps space is Menara Networks. A Menara executive told me back in October at SCTE Cable-Tec that the company has a 4x28-Gbps <a href="">CFP</a> in the works with an integrated OTU-4 framer and Ultra-FEC that should be ready for testing right about now. The company also has plans for a tunable SFP+.</li> <li>• Speaking of Cable-Tec, CommScope filled me in on a new hybrid cable product they call E2O that combines coax with fiber or conduit (or both). Similarly to the hybrid copper/fiber cables available for the enterprise market, E2O is aimed at <a href="/content/lw/en/cable-mso.html">cable MSOs</a> who rely on coax now, but can envision a time when they’ll want to switch to fiber, and don’t want to recable. I’ll have more on this soon.</li> <li>• Last month Google Executive Chairman <a adhocenable="false" href="" target="_blank">Eric Schmidt let slip</a> that the company plans to add more cities to its Google Fiber roster. The statement suggests I should reconsider the position I took in an earlier blog that Google’s <a href="/content/lw/en/fttx/ftth-b.html">FTTH</a> ambitions were relatively modest (see <a href="">“What is Google Fiber trying to prove?”</a>). However, I don’t always take suggestions – and in this case, I’m not going to join those who see Google as a nationwide communications services provider. Schmidt said that Google is examining its options right now. We can assume they’re looking for another city like Kansas City – where the competition is copper-based, the local authorities very cooperative, and the residents are well off enough and bandwidth-hungry enough to pay for a fiber-based connection.</li> <li>• Back at ECOC in September, Liverage showed off a very compact test instrument that combines serial pattern generation, <a href="">BERT</a>, and power meter for SFP+ transceivers called, snappily, the SFP+ Checker. The pocket-size instrument also will read the transceiver’s internal memory EEPROM and interact with the digital diagnostics of DDM-enable modules. It’s controlled via a PC.</li> <li>• There’s a good chance you know this already if you’re active in coherent photonics, but just in case: Teraxion is expanding its product line beyond dispersion compensators to include <a href="">coherent receivers</a>. Beta samples of the device, which a Teraxion source at ECOC described as one-third the size of the OIF specs, were expected to be ready for review this past November. The receiver leverages the company’s work in silicon on insulator technology. Given the fact that coherent-enabled transmission is expected to obviate the need for dispersion compensation, this appears to be a case of “If you can’t beat them, join them.”<br> <br> </li> </ul> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2013/01/The-secrets-Ive-been-keeping.html2013-01-04T19:30:00.000Z2013-02-26T22:36:59.558ZGuess that customer!noemail@noemail.orgStephen Hardy<p>During <a href="" target="_blank">Infinera's</a>&nbsp;(NASDAQ: INFN) third quarter earnings call this afternoon, CEO Tom Fallon was once again quizzed on the company's so far unidentified North American Tier 1 customer for its new <a href="">DTN-X</a>. Fallon -- whose desire to spill the beans was palpable -- said that Infinera would reveal the mystery guest's identity by the end of the year, before hastily adding, &quot;with the customer's permission.&quot;</p> <p>Assuming Mr. Fallon gets his wish, I'm running out of time to play one of the optical industry's favorite parlor games, &quot;Guess That Customer.&quot; In this instance, the rules of the game are that you have to make your selection using only the hints Infinera has dropped and your own reasoning -- since Infinera won't tell you directly (believe me, I've tried).</p> <p>So, what are these aforementioned hints? They are:</p> <ol> <li>It's a long-haul application.</li> <li>Infinera announced it was entering the OSMINE process at the behest of this Tier 1 while the DTN-X was still in the trial stage. The first time the company issued a contract win scorecard was during its 2Q12 earnings announcement, when it said it had 10 customers, three of whom represented new business. (It now has 16 customers, five of whom are new to the company.)</li> <li>That's about it.</li> </ol> <p>So, where do we go from here? Let's start by narrowing &quot;North America&quot; to the United States, thanks to the OSMINE factor and the fact that the customer has been referred to in earnings calls as &quot;domestic&quot; (sorry about that, Canada). So that winnows the candidates quickly to AT&amp;T, CenturyLink, and Verizon.</p> <p>I'm going to dismiss AT&amp;T right off the bat. The company hasn't shown much interest yet in deploying 100-Gbps systems with integrated WDM/ROADM/OTN switching.</p> <p>That leaves CenturyLink and Verizon. Verizon has been a popular selection with several observers. The carrier's Stu Elby spoke at the DTN-X's unveiling, Verizon has an interest in such equipment, and Infinera recently paired with the service provider to <a href="">co-author a paper on PMD mitigation</a> for the <i>Journal of Lightwave Technology</i>.<br> </p> <p>However, I'm thinking it's not Verizon. As hinted, the Tier 1 carrier plans to use the DTN-X for its long-haul network. Verizon does have a major backbone upgrade initiative underway. However, <a href="">as I reported last June</a>, Executive Vice President and CTO Tony Melone in June helpfully listed the major vendors for this effort during a press conference at TIA: Inside the Network -- and Infinera wasn't among them.</p> <p>Of course, that list may not have been complete or plans may have changed. But I'm going to believe my own reporting, just on principle.</p> <p>So that leaves CenturyLink -- which, it just so happens, is using the Infinera DTN in its network thanks to its acquisition of Qwest. (Infinera <a href="" target="_blank">has a blog post</a> about the DTN in CenturyLink's network in Wyoming, if you need further evidence.) And with most of Infinera's DTN-X sales going to existing customers, that increases the likelihood that CenturyLink is the correct choice.</p> <p>So that's what I'm going with. We'll see if I'm right by the end of this year -- with the customer's permission, of course.<br> </p> <p>&nbsp;</p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2012/10/guess-that-customer.html2012-10-25T02:00:00.000Z2013-02-26T22:39:29.994ZECOC 2012 Reporter's Notebook: Day 3noemail@noemail.orgStephen Hardy<p>The final day of the ECOC exhibition had a late fall feeling -- chilly and wet. As is often the case on the last day of an exhibition, my attempts to speak with booth personnel were unimpeded by more interesting visitors, due to a general lack of foot traffic in the aisles.</p> <p>I've already written about the most interesting event of the day -- the discovery that Finisar is completing construction of a <a href="/content/lw/en/articles/2012/09/finisar-builds-flexible-grid-roadm-line-card-for-cisco.html">two-slot flexible-grid ROADM line card for Cisco</a>. As I mention in the story, development work on the card is ongoing. My impression is that at least some of this involves the flexible optical channel monitor (OCM).</p> <p>It appears that there are two &quot;flexible&quot; building blocks necessary to enable flexible-grid ROADMs: a flexible-grid wavelength-selective switch (WSS) and an equally flexible grid OCM. Right now, it seems no one has both -- although several companies are getting close. Finisar has the WSS and is coming in for landing on the OCM. Oclaro <a href="/content/lw/en/articles/2012/09/oclaro-offers-eight-port-flexible-grid-optical-channel-monitor.html">has announced the requisite OCM</a> but is still working on the WSS, they told me later in the day. JDSU is within range on both items, I was told this past Monday, but neither has been announced as ready to go. The WSS will be &quot;released&quot; in November, with the OCM will be &quot;another quarter or two.&quot;</p> <p>Both Finisar and Oclaro also announced CFP2 modules, with schedules for the two companies roughly in synchronization. Sampling this year, production first quarter next year.</p> <p>Catching up on some OFC 2012 announcements, my hosts at both Finisar and Oclaro told me today that their 4x28-Gbps modules for metro 100-Gbps applications are sampling. Neither company wants to&nbsp;offer a date for general availabiity, however.</p> <p>For a wider-ranging look at ECOC 2012, check out <a href="">my overview podcast</a>. And, of course, there are the reporter's notebooks for <a href="">Day 1</a> and <a href="">Day 2</a></p> <p>&nbsp;</p> <p>&nbsp;</p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2012/09/ecoc-2012-reporters-notebook-day-3.html2012-09-19T20:04:00.000Z2013-02-26T22:40:23.045ZECOC 2012 Reporter's Notebook: Day 2noemail@noemail.orgStephen Hardy<p>Yellow umbrellas with the ECOC 2012 logo popped up all around Amsterdam this morning, as it was raining when attendees traveled to the RAI Centre for the opening of today's program. Show organizers presciently included the umbrellas in the attendee registration bags. Unfortunately, I managed to dismantle mine at my Metro stop while trying to shake it out. I'm more or less waterproof; my clothes are not.</p> <p>Fortunately, I dried out enough to prove the point in this blog that there's more being discussed at the show than 100 Gbps and above.</p> <p>Take <a href="/content/lw/en/fttx/ftth-b.html">FTTH</a>, for example. United Technologists Europe Ltd (UTEL) filled me in on the new <a href="/content/lw/en/articles/2012/09/utel-launches-fast-light-ftth-fault-detector.html">Fast Light FTTH field test instrument</a> the company has announced here at the show. The major selling point is that that carriers can install the Fast Light system at the OLT and be able to perform single-ended testing through the splitters to individual ONTs. The system also has an integrated switch in addition to the test capabilities. Fast Light is about to undergo trials with a major European carrier, according to UTEL Managing Director Frank Kaufhold.</p> <p>Meanwhile, NeoPhotonics says its <a href="/content/lw/en/articles/2012/09/neophotonics-unveils-mode-coupling-receiver-for-more-efficient-ftth.html">new Mode Coupling Receiver (MCR) optical line terminal (OLT) transceiver</a> is in qualification with GPON system customers. The NeoPhotonic sources with whom I spoke wouldn't name those customers. However, they did allow that the fact that they've co-authored a paper on the technology for ECOC with British Telecom provides a large hint. The MCR transceiver takes advantage of NeoPhotonics's photonic integrated circuit (PIC) technology, which the sources asserted was best suited for adding functionality to PON equipment rather than removing cost from commodity items. Further PIC-based PON products are in the pipeline, they said.</p> <p>WDM-PON is well represented at the show via startup AEPONyX. Untangling the acroynm that forms the company's name reveals that the company is in the business of providing equipment for Active Ethernet PON in a WDM-PON format. The company is a spin out of Montreal-based communications service provider Tel-Tech, which developed the technology for its own use. Francois Menard, CTO and co-founder of AEPONyX and still an employee of Tel-Tech, is extremely passionate about the potential of WDM-PON to offer streamlined network infrastructures that won't require the complexity of current metro networks.</p> <p>For more on ECOC, read the <a href="/content/lw/en/blogs/lightwave-blog/2012/09/ecoc-2012-reporters-notebook-day-1.html">Day 1</a> and <a href="/content/lw/en/blogs/lightwave-blog/2012/09/ecoc-2012-reporters-notebook-day-3.html">Day 3</a> blogs.</p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2012/09/ecoc-2012-reporters-notebook-day-2.html2012-09-18T20:04:00.000Z2013-02-26T22:40:12.711ZECOC 2012 Reporter's Notebook: Day 1noemail@noemail.orgStephen Hardy<p>After a slate of workshops yesterday, ECOC 2012 got in full swing today with both the conference and exhibition drawing the interest of attendees here in Amsterdam. Just over 1065 people preregistered for the conference, a slightly higher number than last year. Some are more colorful than others.</p> <table cellspacing="0" cellpadding="1" border="0"> <tbody><tr><td><img src="/content/dam/lw/online-articles/2012/09/IMG_0018.jpg"></td> </tr><tr><td><b>Transportation being what it was in the 16th century,<br> some people arrived very late for the flash mob.</b></td> </tr></tbody></table> <p>Not surprisingly, much of the talk so far has centered on 100G and what comes next. In a workshop yesterday entitled &quot;Beyond 100G - Technology Options,&quot; <a target="_blank" href="">JDSU</a> CTO Brandon Collings underscored the fact that when it comes to creating a network that can handle superchannels, flexible-grid <a href="">wavelength-selective switches</a> are only one of the necessary building blocks. I visited with him at the booth today to get further details (and <a href="">in this podcast</a> he'll tell you what he told me). Fortunately, the technology for the non-WSS building blocks are more or less in place; the issue now is getting them commercialized at an appealing cost.</p> <p>Of course, you need to be able to transmit at 400G before any of this flexibiity is necessary. Collings believes that, on the optics side, much of 400G technology will be based on 100G technology, just in units of four. The issue here, again, is cost -- he's thinking the components will have to support speeds of 4X 100G at only 2X the cost.</p> <p>And speaking of 100G, it sounded to me that if you're waiting for JDSU to come out with a 100G coherent transponder module, you'll be waiting a long time. Collings said the company hasn't totally scrubbed the idea, but he noted that the systems houses that are leading the 100G market are relying on in-house designs based on their own ASIC technology. They're going to companies like JDSU for the optical components that complement these ASICs, and JDSU right now is happy with this level of engagement.</p> <p>However, not everyone agrees with this assessment of the 100G coherent module opportunity. Israeli module vendor Civcom <a href="/content/lw/en/articles/2012/09/civcom-touts-100-gbps-coherent-tunable-transponder.html">announced its 100G coherent module last week</a>. According to the company's head of EMEA &amp; APAC, Ehud Bejerano, the Civcom module will consume less than 80 W in an OIF form factor. He expects to be able to differentiate on price and performance as well. The company expects to have engineering samples in their customers hands next month. The company co-developed the module with a lead customer that has committed to it; other potential customer engagements will start at the evaluation stage.</p> <p>Civcom also has its eyes on the 100G metro market. It is looking at both coherent and direct-detect approaches.</p> <p>Meanwhile, back at 400G and beyond, Neda Cvijetic of NEC Labs America described her company's work in optical orthogonal frequency-division multiplexing (OFDM) as an alternative to multi-level signaling and its ever-increasing complexity in yesterday's &quot;Beyond 100G&quot; workshop. She admitted that pulse shaping and the requirement to oversample high-speed analog-to-digital converters (ADCs) remain the primary stumbling blocks to commercialization, but the company is making progress, as a recent demonstration with Verizon highlighted.</p> <p>One technology that could aid optical OFDM or superchannels using comparatively less exotic transmission schemes is optical combs, which can transmit very tightly spaced channels. Irish startup Pilot Photonics is talking up its approach to optical combs at the show. Newly installed CEO Stan Lumish says the company is focusing on R&amp;D test and measurement applications first, and has its first product ready to ship next month. However, because the technology at the heart of the company's approach opens the potential to integration within other systems, Lumish says Pilot sees potential in applications other than test and measurement, potentially including optical transmission. (You can hear Stan discuss this <a href="">in another podcast</a> I recorded earlier today.)</p> <p>For more on ECOC, read the <a href="/content/lw/en/blogs/lightwave-blog/2012/09/ecoc-2012-reporters-notebook-day-2.html">Day 2</a> and <a href="/content/lw/en/blogs/lightwave-blog/2012/09/ecoc-2012-reporters-notebook-day-3.html">Day 3</a> blogs.</p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2012/09/ecoc-2012-reporters-notebook-day-1.html2012-09-17T15:15:00.000Z2013-02-26T22:39:59.905ZWhy 100G demand has hit the acceleratornoemail@noemail.orgStephen Hardy<p>Just two years since its introduction, 100-Gbps technology based on dual-polarization quadrature phase-shift keying (<a href="">DP-QPSK</a>) with coherent detection is taking off. As Daryl Inniss, vice president and practice leader, components, at Ovum plans to discuss in more detail during a Lightwave webcast this Wednesday, September 12, at 1 PM EDT on <a href="">“Getting 100G to Primetime and Illuminating a Path to 400G,”</a> the growth curve for 100G looks like those hockey sticks we all remember from the early part of this century.<br> <br> However, unlike those hockey sticks, the one tracing the demand for 100-Gbps – a demand which, I predict Daryl will say, hasn’t even hit its stride yet – won’t break any time soon (barring a collapsing world economy, of course). There are several reasons for this, with these three coming immediately to mind:<br> </p> <ol> <li>Improved spectral efficiency and lower cost per bit: The ability to cram more bits into more fibers saves money, which is of course essential in the current economic environment.</li> <li>The technology works: Carriers made it clear that coherent-enabled 100G would have to work on existing plant without having to change the location of amplifiers and repeaters. And it does. It also obviates the need for additional dispersion compensation – and, if you believe system manufacturers, the cost and necessity of building dispersion maps.</li> <li>Lower latency: The fact that dispersion compensating fiber is no longer required shortens fiber paths, lowering latency on coherent-enabled links.<br> <br> </li> </ol> <p>These factors have enabled the first generation of 100G systems vendors, who developed their technologies in-house, to lay a solid foundation for market acceptance. Yet even as these vendors increase their feature sets and improve the efficiency of their internally designed ASICs and line cards, we can now expect to see a new type of system come to market next year – a system based on off-the-shelf modules. The fact that Oclaro has announced its module is in full production (and likely isn’t the only vendor to have achieved this milestone) while companies such as <a href="/content/lw/en/articles/2012/09/civcom-touts-100-gbps-coherent-tunable-transponder.html">Civcom have joined the parade</a> means that significant competition – and cost reduction – is about to hit the space. This, of course, will really accelerate 100G deployment even further.<br> <br> Meanwhile, carriers are already coming to the realization that 100-Gbps won’t hold back the bandwidth tide for very long. We’re already seeing discussions about what comes next, particularly 400 Gbps. As Kevin Drury of Alcatel-Lucent <a href="" target="_blank">will touch upon in the webcast</a>, the foundation for this transition is already under construction.<br> <br> It’s clear that the hype that has surrounded 100G is proving to be more fact than fantasy. While 10 Gbps likely will remain the standard bearer for most network data rates well into the future, 100 Gbps already has proven that it will be much more than a niche technology.<br> <br> <br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2012/09/why-100G-demand-has-hit-the-accelerator.html2012-09-10T20:04:00.000Z2013-02-26T22:39:48.409ZWhat is Google Fiber trying to prove?noemail@noemail.orgStephen Hardy<p>The <a href="/content/lw/en/articles/2012/07/google-fiber-open-for-business.html">launch of Google Fiber</a> in Kansas City (Kansas now, Missouri later) is sufficiently advanced to enable pundits to assess what Google has deemed fit to reveal of its business model.<br> <br> For example, we have <a target="_blank" href="">a recent post by Peter Cohan on the <i>Forbes</i> website</a> that raises the question of whether Google is about to blow $28 billion on its <a href="/content/lw/en/fttx/ftth-b.html">FTTH</a> effort. Cohan has concluded Google Fiber is a bad idea, and offers the input of a pair of industry insiders as supporting evidence. One of these is John Cioffi, CEO of DSL software technology vendor ASSIA Inc. (see, for example, <a href="/content/lw/en/articles/2012/06/assia-dsl-expresse-upgrade-aims-at-100-mbps.html">“ASSIA DSL Expresse upgrade aims at 100 Mbps”</a>), who cites FCC data from 2010 that puts the price tag to connect a home with fiber at $2500 (a figure Cioffi believes is an underestimation) and a March/April 2011 article from <i>Rural Telecom</i> magazine that suggests connection costs can be as high as $6000 per home in rural areas. (I'd link to the article, but you need to be an NTCA member to read it.) Cohan quotes Cioffi as stating that, as an alternative to such onerous expenses, Google could have hit its 1-Gbps downstream target by leveraging all the DSL and Wi-Fi channels available in a given “fiberhood.”<br> <br> There are, clearly, problems with this reasoning:<br> </p> <ol> <li>In terms of cost, there’s no reason to believe that Google’s expenses are as high as the average price tag from the FCC’s 2010 report. And, in aggregate, neither Kansas City is exactly rural.</li> <li>Google doesn’t have DSL or Wi-Fi infrastructure in place to leverage. And even if it did, its goal is to offer a consistent symmetrical 1 Gbps that each subscriber can access via one pipe, not by aggregating multiple broadband sources together while hoping his or her neighbors aren’t hogging the area’s shared broadband resources.<br> </li> </ol> <p>Fortunately for Cohan’s argument, he has another insider to lean on, Neal Lachman of Angie Communications. Lachman has been active in various broadband provision endeavors <a href="/content/lw/en/articles/2002/05/ftth-company-to-announce-multinational-broadband-infrastructure-53446857.html">for several years</a>. In a detailed <a href="" target="_blank">post on the website Seeking Alpha</a>, Lachman examines the Google Fiber deployment through the lens of reasoning that has led Angie Communications to pair FTTH with wireless technology as its broadband strategy of choice. <br> <br> The gist of this reasoning is that in many serving areas there will be some homes that are just too expensive to reach via fiber. Lachman suggests that if Google wanted to become a 500-lb broadband gorilla serving 15 million paid subscribers via FTTH alone, it would have to spend anywhere from $19.8 billion to $28 billion (which is where Cohan gets his big scary number). Lachman posits an average cost per connection of between $850 and $1250; he adds these numbers could be slightly higher – although still about half of what Cioffi posits.<br> <br> Lachman’s post is well worth reading. However, his initial assumption also has a serious flaw in the context of Cohan’s argument: Who says Google wants to be a big-time bandwidth provider?<br> <br> In fact, Google’s motives for the Google Fiber project should be determined before we begin to speculate on its potential success or failure. Unfortunately, so far that’s been difficult to do because we’ve heard conflicting goal statements from Google sources. Initially, the company called the project a testbed. In this context, the FTTH initiative seemed a vehicle through which Google could influence telecom policy by showing how consumers would benefit from the myriad services a high-bandwidth, open access (and net neutral) pipe could provide.<br> <br> However, company sources subsequently were quoted as saying they do expect to make money on the project. And the fact that the open access model has been replaced by one where all of the subscriber’s broadband services spend goes directly to Google (for the time being, anyway), as well as its strategy of cherry picking deployment areas based on whether pre-determined take rates would make connections worthwhile, would seem to underscore the company’s desire to at least break even.<br> <br> These two goals aren’t mutually exclusive. In fact, Google’s effort would more effectively influence telecom policy if the company could show that its vision is also a moneymaker. The issue now is that vision is starting to look like those of the mainstream communications services providers with whom it has been battling in Washington.<br> <br> Who knows – maybe in trying to teach those carriers a lesson, Google has learned a thing or two itself.<br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2012/08/What-is-Google-Fiber-trying-to-prove.html2012-08-22T20:04:00.000Z2013-02-26T22:41:00.243ZWired about Wirelessnoemail@noemail.orgStephen Hardy<table cellspacing="0" cellpadding="1" border="0" align="left"> <tbody><tr><td><img src="/content/dam/lw/site-images/stephen_hardy_headshot.jpg"></td> <td>&nbsp;</td> </tr></tbody></table> <p>Infonetics Research’s recent statement that it expects <a href="/content/lw/en/articles/2012/06/infonetics-telecom-capex-will-spike-this-year.html">a spike in carrier capex this year</a> no doubt sparked incredulity in some, given the slow start to telco spending this year. But the firm’s ancillary assertion that much of this spending will go toward wireless networks should not have surprised anyone. Everyone is going mobile, in case you haven’t heard, which is one reason communication carriers continue to see their wireline revenues shrink.<br> <br> Fortunately, while wireless is portrayed as a competitor to fiber optics in some quarters, the emphasis on supporting new mobile services should increase fiber penetration, not retard it.<br> <br> <a href="/content/lw/en/mobile-backhaul.html">Mobile backhaul</a> demands, of course, provide the main catalyst for my confidence. With T1/E1s and other low-speed connections no longer adequate for 3G, never mind 4G requirements, mobile operators will need more fiber connections that support Ethernet-based service provision.<br> <br> Carriers have a number of options for beefing up the fiber quotient in their backhaul networks. As Jon Baldry of Transmode recently discussed during a <i>Lightwave </i>webcast on providing scalable mobile backhaul infrastructure, the right architecture choice depends on a number of factors, including what’s already in place. He provided a series of mini case histories that explored some of these architecture options; they illustrate the flexibility optical communications technology can provide (see <a href="/content/lw/en/webcasts/2012/05/scalability-for-mobile-backhaul-applications.html">“Scalability for Mobile Backhaul Applications”</a>). <br> <br> Against this background, there seems to be uncertainty within the market about the role MPLS should play in mobile backhaul. Some technology suppliers argue for an approach based on native Ethernet, while others suggest schemes based on Optical Transport Network (OTN) provide a better fit. However, with MPLS finding favor in a wider variety of applications, one could wonder if taking MPLS to the cell site would make sense. Telco Systems is a strong proponent of this option; the company’s Nir Halachmi stated this case twice recently, once in the pages of <i>Lightwave</i> (see <a target="_blank" href=";mode=1">“Better Backhaul with MPLS to the Cell Site”</a>) and again <a href="/content/lw/en/webcasts/2012/05/scalability-for-mobile-backhaul-applications.html">in the aforementioned webcast</a>. The statements that MPLS is too immature, insufficiently scalable, and comparatively expensive are more myth than reality, Halachmi asserted. It will be interesting to see how many carriers agree.<br> <br> Regardless of the approach, however, the resulting mobile backhaul network must be both scalable and reliable. Packet-optical transport systems provide a superior option to meet these requirements, asserted Kurt Raaflaub of ADTRAN during the webcast. The key is balancing service-centric attributes such as SLA fulfillment, restoration capabilities, and synchronization support with such opex concerns as power, density, and footprint, Raaflaub said – which means service providers might not want to deploy the same platforms they’re using in the core in their mobile backhaul networks.<br> <br> Meanwhile, we’re seeing not only an increase in fiber to the tower, but up it as well. The evolution toward the use of fiber-fed remote radio head (RRH) architectures offer the best approach for accommodating both increasing bandwidth and the equally increasing number of operators per tower. (You’ll find more details of the RRH architecture and its benefits in <a href="/content/lw/en/articles/print/volume-28/issue-5/applications/fiber-reaches-the-antenna.html">“The X Factor Has Changed in FTTx”</a> from <i>Lightwave </i>magazine.)<br> <br> The fact that fiber-optic projects aren’t main recipients of opex dollars is certainly disappointing, but not a sign that the technology is being relegated to second-tier status. As we well know, wireless networks still need wireline support – support fiber is in the best position to provide.<br> <br> </p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2012/06/wired-about-wireless.html2012-06-20T20:04:00.000Z2013-02-26T22:41:32.134ZOclaro 2Q12 earnings call tidbitsnoemail@noemail.orgStephen Hardy <br>Editorial Director and Associate Publisher, Lightwave<p>Oclaro Chairman and CEO Alain Couder and CFO Jerry Turin discussed the company's results of the company's fiscal second quarter 2012 with analysts January 26 after announcing it had exceeded guidance by $1 million (see <b><a href="">&quot;Oclaro weathers Thailand flooding in fiscal second quarter&quot;</a></b>). Here are some statements of interest I didn't have room to fit in my story:</p> <p>1. Couder says that some of his customers had turned to his competitors to supply products he couldn't as a result of the flooding at Fabrinet's Chokchai facility in Thailand. (They promised to resume purchases from Oclaro when its production came back online, he added.) But Oclaro had also picked up some extra business due to its competitors' flooding problems. Plus and minus ended up a net neutral, Couder said.</p> <p>2. Internally, Oclaro was using a narrow linewidth laser from a flood-affected source in its 40-Gbps coherent module. The experience has led the company to accelerate development of an internally produced replacement, according to Couder.</p> <p>3. Oclaro's high-port-count <a href="/content/lw/en/articles/2010/09/finisar-unveils-flexgrid-wavelength-selective-switches-for-next-gen-roadms-102339379.html">wavelength-selective switch.</a></p> <p>4. (WSS) is sampling. It combines a one-axis MEMS with liquid crystal technology.</p> <p>5. The company is still hoping to sell its Shenzhen production facility. It is now talking to a second contract manufacturer as a potential purchaser.</p> <p>6. While the <a href="/content/lw/en/articles/2011/09/clariphy-oclaro-tout-40-gbps-coherent-130119053.html">40G coherent module's</a> popularity is growing (revenues tripled in the quarter), Couder said he hasn't seen a drop off in sales of 40G DPSK modules. He added that the company's 40G DQPSK module has never been a big seller.</p> <p>7. Couder added that sales are picking up in China. The same can't be said for Europe. He added that it's too soon to make a call on what the North American market will look like in the near term.</p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2012/02/oclaro_2q12_earnings.html2012-02-02T22:10:00.000Z2013-02-26T22:41:47.234ZFTTH boom: Fun while it lasted?noemail@noemail.orgStephen Hardy <br>Editorial Director and Associate Publisher, Lightwave<p>January 16, 2012<br> </p> <p>One tidbit from last month that slipped under the radar (to the point where the source decided to retweet it today) was the release December 13 of details from Infonetics Research's third quarter 2011 PON, FTTH, and DSL Aggregation Equipment and Subscribers report. The report states that revenues in the third quarter declined sequentially.</p> <p>But it is the quote from analyst Jeff Heynen that deserves the most attention. &quot;We expect that 2011 is going to be the peak for the fixed broadband aggregation equipment market,&quot; he is quoted as saying <a target="_blank" href="">in the press release</a>. &quot;The <a href="">PON</a> spending boom in China, coupled with DSL growth in Central and Latin America and VDSL and GPON growth in the EMEA region, created a perfect storm of spending this year, but with peaks come valleys: By 2015, the broadband aggregation market will decline to $5.0 billion, from what we expect to be its high of $7.9 billion this year.&quot;</p> <p>Now $5 billion is a nice little market size, but no one enjoys a slide down the revenue curve. Heynen suggests that carriers will begin to shift spending from DSL to fiber, so FTTH might enjoy a greater share of the total broadband aggregation market, softening the effects of the overall broadband market decline.</p> <p>Still, there are reasons to think the <a href="">FTTH</a> market will still feel a pinch after this year is over:</p> <p>While I hope Heynen is right about carriers moving from DSL to fiber, the introduction of node vectoring, super MIMO, phantom mode, and other DSL-enhancing technologies should slow the pace of such transformation.</p> <p>As in other parts of optical communications, China represents a huge share of the total worldwide FTTH market. The pace of FTTH deployments there will inevitably slow. Any growth we may see in EMEA and Latin America likely won't be enough to offset a decline in China.</p> <p>In the U.S., I'm thinking most of the broadband stimulus spending will occur this year. Other FTTH roll outs will take place subsequently, but not in the numbers the stimulus program created. And I think we can forget about significant deployments from U.S. Tier 1 carriers.</p> <p>The IEEE <a href="/content/lw/en/articles/2011/12/ieee-epon-over-coax-standards-work-kicks-off-135047218.html">EPON over coax</a> standards effort will dampen any urge the cable MSO community may have had to embark on significant FTTH deployments.</p> <p>Now let me emphasize that I'm not suggesting that the FTTH market is dead -- far from it. But as much as I'd like to think that Mr Heynen is being a bit too pessimistic, I find it difficult to argue with his premise that we may be in for at least a slight downturn in the FTTH market.</p> http://localhost:4503/content/lw/en/blogs/lightwave-blog/2012/01/ftth_boom_fun_while.html2012-01-16T21:34:00.000Z2013-02-26T22:42:03.573ZVerizon’s new cableco spectrum deal does not doom FiOSnoemail@noemail.orgStephen Hardy<p>Verizon, Comcast, Time Warner Cable, and Bright House Networks announced a market-changing alliance last Friday. The cable companies agreed to sell 122 Advanced Wireless Services (AWS) spectrum licenses covering 259 million PoPs to Verizon for a whopping $3.6 billion &ndash; or slightly less than what AT&amp;T will have to pay Deutsche Telekom as a breakup fee if it can&rsquo;t convince the U.S. government to approve its planned purchase of T-Mobile. As significant as this deal could be to the wireless industry (assuming this deal passes muster with regulators), those in the optical communication community have wondered about another aspect of the deal. According to the press release issued Friday:</p> <p style="margin-left: 40px;"><br /> &ldquo;The companies also announced that they have entered into several agreements, providing for the sale of various products and services. Through these agreements, the cable companies, on the one hand, and Verizon Wireless, on the other, will become agents to sell one another&rsquo;s products and, over time, the cable companies will have the option of selling Verizon Wireless&rsquo; service on a wholesale basis. Additionally, the cable companies and Verizon Wireless have formed an innovation technology joint venture for the development of technology to better integrate wireline and wireless products and services.&rdquo;</p> <p><br /> While the announcement doesn&rsquo;t provide further details of these agreements, Neil Smit, president of Comcast Cable, <a target="_blank" href="">blogged</a> that four years from now Comcast will be wholesaling Verizon Wireless services, while Verizon Wireless will be selling its products in its stores. Smit says that Time Warner Cable and Bright House Networks have similar deals in place with Verizon Wireless.<br /> <br /> A quote from Time Warner Cable President and COO Rob Marcus in the press release makes it clear that wireline services from its new <a href="/mso-optics">cable MSO</a> partners will be among the offerings Verizon Wireless will eventually sell.<br /> <br /> So where does this leave Verizon&rsquo;s <a href="/fttx/ftth-b">FTTH</a>-enabled FiOS?<br /> <br /> Some have speculated that the deal is a sign that Verizon is surrendering the competitive fight in broadband services. I see this more as a confirmation that Verizon had no desire to expand beyond the FTTH footprint it created during its initial roll out, which only covers about 14% of U.S. households. Through its deals with Frontier and Fairpoint, Verizon shed service areas where it didn&rsquo;t believe it could make money with FTTH. It&rsquo;s also refrained from running fiber into certain urban markets where it believes installation costs are prohibitive. Verizon therefore was already limiting where it planned to play in broadband services going forward. It now does little harm for Verizon Wireless to sell services from cable operators in areas where Verizon will face a competitive disadvantage versus DOCSIS 3.0-enabled offerings that it had no plans to overcome &ndash; or where it&rsquo;s not offering wireline broadband at all.<br /> <br /> As for the 14% of U.S. subscribers Verizon can address with FiOS, I expect the company to continue to market its services just as fiercely as ever. Smit already has told reporters that he still expects to compete with Verizon Communications for these broadband subscribers.<br /> <br /> Details of how Verizon Wireless will sell cable MSO offerings &ndash; and, just as important, at what prices compared to Verizon&rsquo;s own offerings &ndash; remain unresolved. But it seems logical to me that Verizon didn&rsquo;t sink billions of dollars into FTTH networks just to see them rot. And it now appears just as logical to conclude that we won&rsquo;t see another massive FTTH roll out from Verizon any time soon.<br /> <br /> <br /> &nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/12/verizons-new-cableco-spectrum-deal-does-not-doom-fios.html2011-12-05T22:11:00.000Z2013-02-26T22:42:25.598ZOclaro expects Thai flood to affect two quartersnoemail@noemail.orgStephen Hardy<p>While other companies have offered predictions on the impact of the <a href="/business/news/Thai-rains-knock-out-Fabrinet-worry-optical-communications-customers-132694093.html">flooding in Thailand</a> for the rest of this year, Oclaro (NASDAQ:OCLR) Chairman and CEO Alain Couder told financial analysts November 10 that he expects Oclaro to feel the flood&rsquo;s effects for two quarters.<br /> <br /> Couder says he&rsquo;s expecting the flood that has knocked out Fabrinet&rsquo;s Chokchai facility and temporarily closed its Pinehurst campus to cost Oclaro $25 million to $30 million in lost revenue in the current quarter and another $10 million to $20 million in the company&rsquo;s fiscal third quarter, which ends March 31, 2012.<br /> <br /> He is hopeful that things will return to normal in the company&rsquo;s fiscal fourth quarter.<br /> <br /> Oclaro is fortunate in that it relies on Fabrinet for only 30% of its finished goods. The company produces 60% of its products in its facilities in Shenzhen (which will have been put up for sale as part of a cost-reduction plan; see <a href="/business/news/Oclaro-looks-to-downsize-133870218.html">&quot;Oclaro looks to downsize&quot;</a>) with the rest spread out at other facilities.<br /> <br /> However, 60% of its Fabrinet production took place at Chokchai. The company is currently moving as much of that as it can to Pinehurst, which has resumed limited operations this week. It also is attempting to fill in the gap at Shenzhen and elsewhere.<br /> <br /> The flood experience has not soured Oclaro&rsquo;s relationship with Fabrinet, however. The two companies recently reached a new contract for manufacturing services.<br /> <br /> But Oclaro is about to introduce a second source for contract manufacturing services. It expects to sell the Shenzhen facility to a &ldquo;major contract manufacturer&rdquo; in the near future. Fabrinet and this new party will then compete for ongoing business, particularly new products.<br /> <br /> Other companies, such as NeoPhotonics and JDSU, have cited their expectations for the flood&rsquo;s impact on their businesses during the current quarter (see <a href="/business/news/NeoPhotonics-third-quarter-revenue-disappoints-133251358.html">&quot;NeoPhotonics third quarter disappoints&quot;</a> and <a href="/business/news/JDSU-forecasts-down-due-to-Thailand-flooding-133158408.html">&quot;JDSU forecasts down due to Thailand flooding&quot;</a>). Oclaro is the first company to extend its expectations into the coming year.<br /> <br /> &nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/11/oclaro-expect-thai-flood-to-affect-two-quarters.html2011-11-15T13:06:00.000Z2013-02-26T22:42:44.833ZNTT still unsure of 100G supplier for PC-1noemail@noemail.orgStephen Hardy<p>Well, no wonder Infinera declined to comment on NTT's announcement that it would move forward with deployment of 100-Gbps technology on the PC-1 submarine cable network (see <a href="/networking/news/NTT-Com-plans-100G-for-PC-1-transpacific-submarine-cable-network-133356928.html">&quot;NTT Com plans 100G for PC-1 transpacific submarine cable network&quot;</a>). It seems that the trial NTT&nbsp;conducted using Infinera's 100-Gbps systems was enough to convince NTT that 100G was ready for deployment, but not that Infinera was the right company for the job -- at least not yet.</p> <p>Responding today to a query I made on Monday, sources at NTT say the vendor decision is still up in the air. &quot;We have done field trial with Infinera, but have not decided to adopt Infinera as a supplier for the deployment,&quot; reported NTT's Takeshi Kawasaki, who works in the media relatinons department. &quot;We will decide future supplier comparing [Infinera] to equipments of other vendors.&quot;</p> <p>NTT and Infinera transmitted 100-Gbps traffic in both line-side and client-side formats across 9,500 km, Infinera announced last month (see <a href="/networking/news/Infinera-trials-100-Gbps-undersea-with-Pacific-Crossing-131548658.html">&quot;Infinera trials 100 Gbps undersea with Pacific Crossing&quot;</a>). NTT&nbsp;now says that it plans to deploy 100-Gbps technology on the undersea cable network by the middle of 2013.</p> <p>Kawasaki did not indicate when vendor selection would take place.</p> <p><br /> &nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/11/ntt-still-unsure-of-100g-supplier-for-pc-1.html2011-11-09T21:30:00.000Z2013-02-26T22:43:01.983ZJDSU's Waechter: Inventory no longer the problemnoemail@noemail.orgStephen Hardy<p>Talking to analysts November 1 after releasing his company's fiscal first quarter earnings (see <a href="/business/news/JDSU-forecasts-down-due-to-Thailand-flooding-133158408.html">&quot;JDSU forecasts down due to Thailand flooding&quot;</a>), JDSU President and CEO Thomas Waechter said that the inventory issues in optical components <a href="/business/news/JDSU-guides-downward-for-start-of-FY2012-128003483.html">he described during the previous quarter</a> have largely eased. As illustrations, ROADM orders increased almost 25% sequentially and tunable XFP orders jumped 70% versus the prior quarter. (However, revenues in both areas declined, it was later revealed.) In fact, 7 out of 12 optical product lines saw an increase in orders over the prior three months.</p> <p>The issue now is longer capex approval processes at service providers, he lamented. This has led him to conclude that JDSU will not see a seasonal &quot;budget flush&quot; between now and end of the calendar year.</p> <p>Other highlights of the analyst call, with a hat tip to <a href="" target="_blank">Seeking Alpha</a> for the transcription:</p> <ul> <li>JDSU&nbsp;has received indications that Fabrinet may have some capacity at its Pinehurst facility back online in the next week or two. The campus now has power, and water levels in the surrounding area have begun to recede.</li> <li>The company expects to release a 40G coherent module in the second half of its fiscal 2012. Additional client-side 40G products also are on the way.<!--[if gte mso 9]><xml> <w:WordDocument> <w:View>Normal</w:View> <w:Zoom>0</w:Zoom> <w:PunctuationKerning /> <w:ValidateAgainstSchemas /> <w:SaveIfXMLInvalid>false</w:SaveIfXMLInvalid> <w:IgnoreMixedContent>false</w:IgnoreMixedContent> <w:AlwaysShowPlaceholderText>false</w:AlwaysShowPlaceholderText> <w:Compatibility> <w:BreakWrappedTables /> <w:SnapToGridInCell /> <w:WrapTextWithPunct /> <w:UseAsianBreakRules /> <w:DontGrowAutofit /> <w:UseFELayout /> </w:Compatibility> <w:BrowserLevel>MicrosoftInternetExplorer4</w:BrowserLevel> </w:WordDocument> </xml><![endif]--><!--[if gte mso 9]><xml> <w:LatentStyles DefLockedState="false" LatentStyleCount="156"> </w:LatentStyles> </xml><![endif]--><!--[if gte mso 10]> <style> /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Times New Roman"; mso-ansi-language:#0400; mso-fareast-language:#0400; mso-bidi-language:#0400;} </style> <![endif]--><span style="font-size:12.0pt;font-family:&quot;Times New Roman&quot;; mso-fareast-font-family:&quot;MS Mincho&quot;;mso-ansi-language:EN-US;mso-fareast-language: JA;mso-bidi-language:AR-SA"><br /> </span></li> <li>JDSU has 18 design wins for 40G and 100G technology, Waechter says, with an additional 39 &quot;in process&quot; with 20 customers.</li> <li>According to CFO Dave Vellequette, the company had concluded that the September quarter would have represented the bottom of the revenue trough for the CCOP division prior to the flooding at Fabrinet. The company was expecting to guide for a revenue increase in the second quarter for this business before the monsoon hit Thailand.</li> <li>On the Communications Test side, 100G also drove revenue increases, with a 41% jump year-on-year. The company shipped 26 units of 100G test equipment to 15 customers during the quarter.</li> <li>During the Q&amp;A portion of the proceedings, Waechter said he continues to have a healthy appetite for M&amp;A and that he was seeing attractive opportunities in the test and measurement space.</li> </ul>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/11/jdsus-waechter-inventory-no-longer-the-problem.html2011-11-02T18:54:00.000Z2013-02-26T22:43:14.520ZOvum assesses Thai flood impactnoemail@noemail.orgStephen Hardy<p>As previously reported, optical equipment companies who have relied on Fabrinet to manufacture their products have a problem on their hands in the wake of the <a href="/business/news/Thai-rains-knock-out-Fabrinet-worry-optical-communications-customers-132694093.html">flooding in Thailand</a>. Ovum's Daryl Inniss believes that second-sourcing options and other contingency plans will limit the damage to these companies' bottom lines.</p> <p>Inniss, <a href="" target="_blank">in a brief posted on the Ovum website</a>, lists Oclaro, Emcore, Infinera, Opnext, JDSU, and Finisar as among the companies using Fabrinet, all of which can expect to be affected to some degree. Finisar is perhaps in the best shape, since its use of Fabrinet's Chokchai and Pinehurst facilities is not as extensive as the other companies on Inniss's list.</p> <p>While the supply of optical communications components and subsystems will take a hit this quarter as a result of the floods, Inniss believes the impact will prove short-lived. He therefore doesn't believe the flooding offers competitors of the affected companies much of a chance to wrest away market share.</p> <p>&quot;We expect the market to return to capacity by the end of 1Q12,&quot; he writes. &quot;Component vendors will use internal manufacturing capacity where products and processes were developed before being outsourced and will use contract manufacturers where they have long-standing relationships and agreements. They can increase capacity by running multiple shifts, but they will be limited by the availability of test equipment.&quot;</p> <p>Inniss believes larger carriers and network equipment manufacturers to be less likely to feel the effects of a prolonged shortage because of their ability to use &quot;parts in the pipeline&quot; to meet their needs. Companies with smaller inventories may feel the pinch if shortages linger, however.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/11/ovum-assesses-thai-flood-impact.html2011-11-01T14:29:00.000Z2013-02-26T22:43:27.629ZNeoPhotonics honcho explains Santur buynoemail@noemail.orgStephen Hardy<p>Tim Jenks, president, CEO, director, and chairman of the board at optical technology developer NeoPhotonics (NYSE: NPTN), told me in a conversation late last week that several factors made Santur an attractive acquisition target (see <a href="/business/news/NeoPhotonics-agrees-to-acquire-Santur-130949648.html">&ldquo;NeoPhotonics agrees to acquire Santur&rdquo;</a>).<br /> <br /> &ldquo;Santur is a company that has had an important role in developing tunable products for the market over time and has had a strong share position in tunable products,&rdquo; Jenks said. &ldquo;That&rsquo;s both tunable lasers that are used in, for example, 10-Gbps networks but also narrow-linewidth lasers that are used in coherent applications.&rdquo;<br /> <br /> NeoPhotonics, of course, has been active in developing coherent receiver technologies, so it would seem the Santur acquisition should put Jenks and NeoPhotonics in a position to offer coherent transponders.<br /> <br /> &ldquo;They&rsquo;ve also been an innovator in a core technology, which is in laser arrays based on Indium Phosphide. And so the ability to develop products from photonic integrated circuits that involve both actives and passives based on different material systems &ndash; certainly the Indium Phosphide arrays from Santur and/or the silica-on-silicon from NeoPhotonics we think are well matched and bring a pretty broad platform to create solutions for customers,&rdquo; Jenks added.<br /> <br /> Jenks said that while the companies have never worked together, they were well aware of each other&rsquo;s capabilities. When Santur decided earlier this year to explore strategic alternatives, &ldquo;we were already in a discussion with them that made us a natural participant in their process. And ultimately, when the process was done, we were doing the deal.&rdquo;<br /> <br /> Jenks said that he&rsquo;s certain that NeoPhotonics was not the only company bidding for Santur, but declined to speculate on which companies also showed an interest. He also said that it was too early to discuss his integration strategy or which Santur executives, if any, would join NeoPhotonics.<br /> <br /> &ldquo;We do expect the business to continue and the facility to continue as is,&rdquo; Jenks offered. He deferred discussion of the fate of the Santur name until after the transaction closes, a milestone Jenks expects to achieve by the end of the year.<br /> <br /> &ldquo;I will say the NeoPhotonics name will continue,&rdquo; he concluded cheekily.<br /> <br /> &nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/10/neophotonics-honcho-explains-santur-buy.html2011-10-11T13:38:00.000Z2013-02-26T22:44:10.589ZFiber shortage likely to continue for at least another quarternoemail@noemail.orgStephen Hardy<p>One of the more interesting topics of discussion at last week's FTTH Conference in Orlando was the ongoing shortage of fiber-optic cabling. One source put the lead time for delivery at an average of 22 weeks -- which is an entire deployment cycle for some carriers and a real thorn in the side of broadband stimulus winners who hear the clock ticking on their completion deadlines.</p> <p>The shortage has a number of sources, according to cable company executives with whom I spoke. The earthquake in Japan knocked out Sumitomo's fiber plant, which took a fair amount of fiber-making capacity off the market. Broadband stimulus award winners who have had their projects delayed by having to complete environmental impact studies, among other hurdles, are now ordering more fiber in a shorter window to make up time, some said. Also, demand for fiber-optic cable has increased overall, stretching current cable construction capacity to its limits.</p> <p>All the sources with whom I spoke said their companies had boosted production at their current facilities. However, they appear to be leery of creating additional fiber- and cable-making capacity in fear of overshooting demand and creating another glut.</p> <p>Few sources would speculate about when the current shortage would ease. The only one who offered an opinion cited &quot;sometime next year.&quot; </p> <p>So it would appear that the fiber cabling shortage should remain an obstacle to network deployments for several months to come. The question therefore arises whether Congress, the NTIA, and the RUS will extend the broadband stimulus completion deadlines if the shortage remains acute. Sources were split on this scenario. But given the fact that the more conservative members of Congress are already looking for ways to pull broadband stimulus funds back, a deadline extension may be more difficult to achieve than one might wish.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/10/fiber-shortage-likely-to-continue.html2011-10-05T17:16:00.000Z2013-02-26T22:43:59.254ZCoherent receivers in the access?noemail@noemail.orgStephen Hardy<p>Nokia Siemens Networks hasn't been very active in the PON space recently, but that doesn't mean it doesn't have plans for the access niche. In fact, the company has some aggressive technological ideas it hopes to implement -- including coherent receivers.</p> <p>A pair of Nokia Siemens Networks sources at the WDM-PON Forum in Orlando today discussed the company's plans -- which, appropriately enough, involved <a href="/fttx/news/Ericsson-launches-40-Gbps-WDM-PON-platform-115901014.html">WDM-PON</a>. Or, more precisely, ultra-dense WDM-PON, with the ability to support up to 1,000 wavelengths. Nokia Siemens Networks has some advanced optical technology in mind for its Paired Channel Technology, including laser sources capable of transmitting multiple wavelengths that create optical transmission groups. The OLT, of course, transmits in the downstream direction and the ONTs transmit a complementary upstream signal offset from the downstream signal by 1 GHz.</p> <p>But it is the planned use of coherent receivers that raised eyebrows in the room. Coherent technology, of course, is the most widely adopted approach for long-haul 100-Gbps DWDM networks, paired with dual-polarization quadrature phase-shift keying (<a href="/equipment-design/featured-articles/is-dpndashqpsk-the-endgame-for-100-gbitssec-54890687.html">DP-QPSK</a>). Like any emerging technology, it's expensive in its present incarnation. But it won't be too expensive to be used to extend WDM-PON transmission reach up to 100 km, Nokia Siemens Networks believes. The sources pointed out that both the optical and electronic/DSP requirements for 1 Gbps over 100 km won't be nearly as complex as the long-haul technollgy now reaching the field. The relaxed requirements will lead to much lower price points that will be within reason for access networks, the sources said.</p> <p>Nokia Siemens Networks has already demonstrated the full ultra-DWDM-PON technology in its labs. But the technology is probably several years away -- think the 2014/2015 timeframe -- from reaching commercialization.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/09/coherent-receivers-in-the-access.html2011-09-28T02:26:00.000Z2013-02-26T22:45:07.284ZTidbits from Ciena's earnings callnoemail@noemail.orgStephen Hardy<p>Here are some interesting facts and analysis from Ciena honcho Gary Smith and CFO Jim Moylan from yesterday's earnings call (with a tip of the hat to <a href="" target="_blank">Seeking Alpha</a>):</p> <ul> <li>While the macroeconomic environment is making carriers around the world a bit skittish, the problem appears particularly acute in Europe. Ciena derives about 25% of its revenues from that part of the world.</li> <li>However, North American Tier 1 customers don't seem to be slowing down their projects.</li> <li>The company sold its first 100-Gbps 5430 Reconfigurable Switching System during the past quarter. It now has nine customers for the platform.</li> <li>Ciena now has almost 100 customers using coherent technology of some sort. In fact, nearly 50% of its transport WDM revenues come from a combination of 40- and 100-Gbps systems. The 40-Gbps systems represent the largest source of that revenue.</li> </ul> <p>The comment regarding carrier spending scrutiny lengthening sales cycles, particularly in Europe, echoes comments executives at several component vendors have made regarding the headwinds they've seen (see, for example <a href="/business/news/Oclaro-Inventory-correction-over-but-revenue-slump-continues-in-4Q11-126410748.html">&quot;Oclaro: Inventory correction over, but revenue slump continues in 4Q11&quot;</a>). Meanwhile, it would appear that 40-Gbps technology sales are far from dead, even with 100-Gbps sales starting to catch on.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/09/tidbits-from-cienas-earnings-call.html2011-09-02T14:35:00.000Z2013-02-26T22:45:25.592ZWill Motorola Mobility buy influence Google's FTTH plans?noemail@noemail.orgStephen Hardy<p>The underpinnings of this morning's announcement that Google (Nasdaq:GOOG) has agreed to buy Motorola Mobility for $40.00 per share (about $12.5 billion) seem pretty clear. Motorola Mobility has embraced Google's Android mobile phone OS in a big way. It also has a significant patent portfolio, and Google had entered the stalking horse bid that set the Nortel patent auction in motion. But one element that hasn't been discussed much, if at all, is the fact that Motorola Mobility is also where Motorola's fiber to the home (<a href="/fttx/ftth-b">FTTH</a>) and <a href="/mso-optics">cable MSO</a>&nbsp;equipment ended up when the company split in two.</p> <p>So what plans, if any does Google have for those in connection with its FTTH efforts? While the company has been quiet about what kind of technology it plans to use in its Kansas City (or, should I say, Kansas Cities) deployment, it had to have investigated GPON technology (as well as DOCSIS) as it performed its FTTH due diligence. Did the presence of these technologies in Motorola Mobility's portfolio make the deal seem sweeter to Google?</p> <p>My guess is that Google won't be using GPON in its initial deployments. But it now has the technology close at hand if it ever decides to see what PON technology can do in an open access network setting.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/08/will-motorola-mobility-buy-influence-googles-ftth-plans.html2011-08-15T14:05:00.000Z2013-02-26T22:45:47.011ZZombie broadbandnoemail@noemail.orgStephen Hardy<p>Pundits are having fun at the expense of AT&amp;T CEO Randall Stephenson, who apparently referred to the DSL technology his company installed in the 1990s -- and which it is still using to serve its non-U-verse customers -- as &quot;obsolete&quot; at a conference earlier this week. (<a href="" target="_blank">GigaOM has the story here</a>.) Of course, the point of rolling out U-verse is to replace this obsolete technology. U-verse remains reliant on a form of DSL for the final connection to the home. But for the most part that's VDSL and not the ADSL/ADSL2 technology to which Stephenson was apparently referring.</p> <p>As I heretically suggested <a href="/blog/video/58528807.html?player=7225576001&amp;title=1063885839001">in my &quot;First Take&quot; video blog this month</a> (What -- you didn't know I have a video blog? Well, you should <a href="" target="_blank">subscribe to Lightwave</a> then, shouldn't you...), there is more life left in last-mile copper than we proponents of fiber everywhere would like to acknowledge. But I was referring to the life granted by new technologies beginning with VDSL2 and extending to node vectoring and phantom mode -- not to the sputtering technology Stephenson has rightly labeled obsolete.</p> <p>That old technology has no life left in it -- but it will continue to be used, including in AT&amp;T's network, based on what we can glean from the approaching end of AT&amp;T's U-verse roll out. So while those lucky enough to be served by networks that will see enhancement can hope for service upgrades, those outside of the U-verse footprint -- or who reside within the footprint of any carrier that thinks ADSL is &quot;good enough&quot; -- are doomed to remain attached to moribund architecture that their service provider won't let die. Call it &quot;zombie broadband,&quot; for lack of a better term.</p> <p>For those in the zombie zones, perhaps the local cable company&nbsp;provides an alternative -- unless it has allowed its HFC networks to slip into an equally lifeless state. And if that's the case? Well, that's why muni networks are popping up across the country -- no one wants to live in the land of the zombies.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/07/zombie-broadband.html2011-07-20T06:04:00.000Z2013-02-26T22:46:19.128ZDebate: Will 100G kill 40G?noemail@noemail.orgStephen Hardy<p>How much will the emerging generation of 100-Gbps transmission systems dampen the demand for 40-Gbps technology has long been a subject of debate in industry circles. Since it's a question that doesn't seem settled, I've asked Mark Lutkowitz, co-founder and principal at market research and analysis firm <a href="" target="_blank">Telecom Pragmatics</a>, to round up a few knowledgeable people and bring the 40G versus 100G debate to <em>Lightwave</em>. We'll all have the opportunity to see what ensues July 14 at 1 PM&nbsp;EDT when Mark orchestrates what is sure to be a lively discussion and a unique webinar.</p> <p>Mark certainly has found a group of people who should know what they're talking about -- and who should bring a variety of conflicting viewpoints. They include:</p> <ul> <li>Kevin Cackovic, senior strategic marketing manager within the Communications Business Unit at Altera</li> <li>Francesco Caggioni, director of strategic marketing &amp; PLM for transport products and technology at Applied Micro</li> <li>Gilles Garcia, director of marketing, wired communications, Xilinx, Inc.</li> <li>Sam Greenholtz, co-founder and analyst at Telecom Pragmatics</li> <li>William Szeto, CTO, terrestrial applications, at Xtera Communications</li> </ul> <p>I'm expecting a discussion that will range on both the line and client side of the question. I also expect a few assumptions to be challenged as well.</p> <p>For the record, I think there's still going to be plenty of life in 40-Gbps technology for the foreseeable future -- but I'll be interested to see how much support for this position I'll receive. If you're interested as well, I urge you to <a href="">register for this webinar</a>. </p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/07/debate-will-100g-kill-40g.html2011-07-08T14:27:00.000Z2013-02-26T22:46:04.535ZClariPhy not a direct replacement for CoreOptics at Nokia Siemens Networksnoemail@noemail.orgStephen Hardy<p>I just got off the phone with Jim Benson, head of optical networks, North America, Nokia Siemens Networks, who gave me some further insight into his company's recent investment in ClariPhy Communications (see <a href="/business/news/ClariPhys-latest-funding-round-includes-Nokia-Siemens-Networks-124184714.html">&quot;ClariPhy&rsquo;s latest funding round includes Nokia Siemens Networks&quot;</a>). Perhaps the most interesting thing he said was that the idea that ClariPhy is being groomed to replace CoreOptics as Nokia Siemens Networks' supplier of coherent technology for 40-Gbps applications was &quot;a bit too strong&quot; a characterization of what's going on.</p> <p>Benson says that Nokia Siemens Networks wants a multi-source approach for coherent technology -- which, at 40-Gbps at least, means that both ClariPhy and CoreOptics are viable options.</p> <p>While discussions between Nokia Siemens Networks and ClariPhy have uncovered points for collaboration, the two companies plan a series of meetings to more firmly establish how their respective roadmaps will dovetail, Benson says. This clearly means that the partnership likely will not bear fruit in time for the first generation of Nokia Siemens Networks' coherent-enabled 100-Gbps offering. Benson says that he expects customer trials of 100-Gbps to begin by late this summer, and Nokia Siemens Networks plans to have the capability generally available in the second half of this year.</p> <p>Yet Benson sees plenty of potential in the collaboration with ClariPhy. He praised the company's work in 40-nm CMOS 40-Gbps technology and 28-nm CMOS 100-Gbps development. Benson also likes ClariPhy's soft-decision FEC expertise, which he sees as essential to low-latency applications.</p> <p>UPDATE: Just to drive the point home, I asked via email after the interview whether CoreOptics was a viable technology source for NSN's 100-Gbps roadmap as well. This is what I got back:</p> <p style="margin-left: 40px;">&ldquo;NSN has a long-term supplier relationship with CoreOptics, which is not affected by the investment in ClariPhy. We use a multi-supplier sourcing strategy to the benefit of our customers and NSN.&nbsp; NSN will continue to work with all suppliers that offer us the best solutions for key optical system technologies such 40G, 100G, and beyond.&rdquo;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/06/clariphy-not-a-direct-replacement-for-coreoptics-at-nokia-siemens-networks.html2011-06-21T20:26:00.000Z2013-02-26T22:47:01.263ZFinisar factoids from 4Q11 earnings callnoemail@noemail.orgStephen Hardy<p>While you're watching investors and analysts pile on -- or, really, pile off -- optical module stocks, here are a few behind the numbers bon mots from yesterday afternoon's Finisar earnings call:</p> <ul> <li>Finisar CEO Eitan Gertel said that the company's tunable XFP device is in qualification with 15 potential customers, with another 10 waiting their turn. Needless to say, he's very bullish on the prospects for this new product.</li> <li>He also likes the prospects for the company's low-port-count wavelength-selective swithches (WSS). Gertel expects revenues for that product to ramp in the second quarter of fiscal 2012 -- basically, this fall. The FlexGrid high-port-count system for next-gen colorless, correctionless, contentionless, and directionless applications also has experienced satisfactory traction, he added.</li> <li>The WSS-led inventory correction is just that -- and not a sign that Finisar is losing market share, Gertel asserted. If other companies aren't seeing the same correction, it's because they're addressing a different (i.e., not so heavily dependent on Chinese OEMs) customer base, he said.</li> <li>R&amp;D investments for Finisar should increase next quarter, as will expenses. Getting new products up and running is the main reason for the expense increase, different company executives repeated during the call.</li> <li>Asked if he expects the inventory correction to bottom during the second fiscal quarter this fall, Executive Chairman Jerry Rawls replied he was &quot;hopeful&quot; that would be the case.</li> <li>Rawls repeated that he believes the company is experiencing nothing more than an inventory correction, rather than a fundamental change in market dynamics that signals a lessening of interest in optical technology. Bandwidth demands continue to grow, he pointed out. When asked if he thought his company would regain the valuation it has lost within two years, he said he would hope that it wouldn't take that long.</li> </ul> <p>&nbsp;</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/06/finisar-factoids-from-4q11-earnings-call.html2011-06-16T14:21:00.000Z2013-02-26T22:46:50.463ZThe future according to Cisconoemail@noemail.orgStephen Hardy<p>I've just posted <a href="/business/news/Cisco-Video-among-drivers-for-4X-Internet-traffic-increase-by-2015-122974708.html">a brief write-up of Cisco's latest Virtual Networking Index (VNI) Forecast </a>for 2010-2015. As usual, it's full of interesting factoids. Here's a sample just from the press release:</p> <ul> <li>Global IP traffic is expected to reach 80.5 exabytes per month by 2015.</li> <li>Cisco's projected increase in Internet traffic between 2014 and 2015 is 200 exabytes. That's more exabytes than the total traffic for 2010.<!--[if gte mso 9]><xml> <w:WordDocument> <w:View>Normal</w:View> <w:Zoom>0</w:Zoom> <w:PunctuationKerning /> <w:ValidateAgainstSchemas /> <w:SaveIfXMLInvalid>false</w:SaveIfXMLInvalid> <w:IgnoreMixedContent>false</w:IgnoreMixedContent> <w:AlwaysShowPlaceholderText>false</w:AlwaysShowPlaceholderText> <w:Compatibility> <w:BreakWrappedTables /> <w:SnapToGridInCell /> <w:WrapTextWithPunct /> <w:UseAsianBreakRules /> <w:DontGrowAutofit /> </w:Compatibility> <w:BrowserLevel>MicrosoftInternetExplorer4</w:BrowserLevel> </w:WordDocument> </xml><![endif]--><!--[if gte mso 9]><xml> <w:LatentStyles DefLockedState="false" LatentStyleCount="156"> </w:LatentStyles> </xml><![endif]--><!--[if gte mso 10]> <style> /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Times New Roman"; mso-ansi-language:#0400; mso-fareast-language:#0400; mso-bidi-language:#0400;} </style> <![endif]--><span style="font-size:12.0pt;font-family:&quot;Times New Roman&quot;; mso-fareast-font-family:&quot;Times New Roman&quot;;mso-ansi-language:EN-US;mso-fareast-language: EN-US;mso-bidi-language:AR-SA"><br /> </span></li> <li>The 245 terabytes per second transmitted traffic average in 2015 is equivalent to 200 million people streaming an HD movie (1.2 Mbps) simultaneously every day.</li> <li>In 2010, PCs generated 97 percent of consumer Internet traffic. This will fall to 87 percent by 2015 as more connected devices come online.</li> <li>In 2015, 10 percent of global consumer Internet traffic and 18 percent of Internet video traffic will be consumed via TVs.</li> <li>Global mobile Internet data traffic will increase 26X by 2015, to 6.3 exabytes per month (75 exabytes annually).</li> <li>Business IP video conferencing will grow 6X -- more than 2X as fast as overall business IP traffic, at a CAGR of 41 percent from 2010 to 2015. (And, yes, Cisco provides telepresence equipment.)</li> </ul> <p>And you wonder why the outlook for optical communications should remain positive?</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/06/the-future-according-to-cisco.html2011-06-01T21:14:00.000Z2013-02-26T22:46:39.699ZCorning: We can't make fiber fast enoughnoemail@noemail.orgStephen Hardy<p>Those of you who experienced the fiber glut were perhaps heartened to learn of <a href="">the Reuters report</a> that Corning has more orders for optical fiber than it can handle. Interestingly, Verizon is one of the principal reasons for the shortfall. The company has ordered 20% more fiber than Corning has expected -- despite shifting its FiOS deployments from passing homes to customer acquisition.</p> <p>Apparently Verizon is doing even better with FiOS than observers (or, at least, Corning) had predicted. In its last quarterly earnings statement on April 21, 2011, which covered 1Q11 performance, Verizon said it added 207,000 net new FiOS Internet connections and 192,000 net new FiOS TV connections in the first three months of the year. That equates to totals of 4.3 million FiOS Internet and 3.7 million FiOS TV connections by the end of the quarter. Verizon expects to report second quarter performance on July 22, 2011.</p> <p>Meanwhile, Reuters reports that Corning CFO Jim Flaws says the company also is seeing increased activity in Japan and Australia. Earthquake recovery and the National Broadband Network (NBN), respectively, are clear catalysts in these instances.</p> <p>Corning says that it plans to expand its manufacturing capacity to meet the surge in demand. But Flaws's statement may also mean good news for OFS, which also has <a href="/fttx/news/Verizon-selects-OFSs-EZ-Bend-cable-to-support-MDU-FIOS-deployments-60087957.html">supplied Verizon with fiber for MDU applications</a>.</p> <p>It also means good news for the optical communications industry in general and the <a href="/fttx/ftth-b">FTTH</a> segment in particular. Just because the initial rollout of FiOS may be over doesn't mean the optical opportunity ended with it.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/05/corning-we-cant-make-fiber-fast-enough.html2011-05-25T15:44:00.000Z2013-02-26T22:47:31.628ZMetro Ethernet Forum reaching for the cloudsnoemail@noemail.orgStephen Hardy<p>The Metro Ethernet Forum (MEF) has recognized the growing importance of cloud services to its constituents and plans to address Carrier Ethernet&rsquo;s role in cloud services provisioning.<br /> <br /> MEF board member Ralph Santitoro, director of Carrier Ethernet market development at Fujitsu Network Communications, tells me the organization is in the requirements development stage, sorting through what areas other standards organizations, such as the National Institute of Standards and Technology in the U.S., are doing and what gaps the MEF might fill.<br /> <br /> Santitoro says that two areas currently look worth close examination. The first is enterprise access to private network cloud services via Carrier Ethernet. The second is cloud-to-cloud connectivity.<br /> <br /> The first effort could include definitions of such a service and what attributes would be necessary. These attributes could include quality of service, class of service, latency, packet loss, and availability, among others. The second effort would focus on the ability to maintain quality as a service passes from one cloud to the other &ndash; similar to the ability of a carrier to ensure Ethernet service quality off net, Santitoro says.<br /> <br /> The MEF likely will need three to six months before it has a firm grip on the project&rsquo;s scope, Santitoro estimates.<br /> &nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/05/metro-ethernet-forum-reaching-for-the-clouds.html2011-05-04T20:20:00.000Z2013-02-26T22:47:20.997ZAll sales final?noemail@noemail.orgStephen Hardy<p>M/A-Com Technology Solutions is currently sorting through its options in the wake of the lawsuit GigOptix slapped on Optomai the day before M/A-Com Tech announced its acquisition of the two-year old company founded by former GigOptix employees (see <a href="/business/news/GigOptix-sues-ex-employees-Optomai-for-alleged-agreement-breaches-120698794.html">&quot;GigOptix sues ex-employees, Optomai for alleged agreement breaches&quot;</a>).</p> <p>For the moment, M/A-Com Tech is distancing itself from the suit to the best of its ability. &quot;Optomai is aware of the GigOptix press release but has not yet reviewed the complaint in detail,&quot; M/A-Com Tech said in a statement released to <em>Lightwave </em>via M/A-Com Tech's PR firm. &quot;Optomai intends to defend itself vigorously against the press release allegations and believes that the allegations are entirely without merit.&quot;</p> <p>While the statement portrays Optomai as an independent entity, GigOptix see things differently. The company announced that it has amended its suit to include M/A-Com Tech as a defendant. &quot;[B]y acquiring Optomai and by selling Optomai's products, M/A-Com has improperly accessed and used GigOptix's proprietary information,&quot; GigOptix stated.</p> <p>The more I dig into this, the more I get the impression the lawsuit caught M/A-Com Tech by surprise. The company isn't saying how much it paid for Optomai, but I wonder whether the cost might be too high.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/04/all-sales-final.html2011-04-27T05:09:00.000Z2013-02-26T22:48:03.376ZGive cable operators their duenoemail@noemail.orgStephen Hardy<p>When considering triple-play service provision from an optical perspective, it may be tempting to discount <a href="/mso-optics">cable MSOs</a> as the group that are doing something other than <a href="/fttx/ftth-b">FTTH</a> -- as the optical competition. Accepting this view ignores several realities.</p> <p>First, of course, is the fact that cable MSOs have plenty of fiber in their networks -- hybrid fiber/coax networks by definition have fiber as a foundation. And if you're going to discount HFC because of the coaxial element, you'd logically have to do the same for telcos who won't go any farther down the <a href="/fttx">FTTX</a> path than fiber to the node (<a href="/fttx/fttn-c">FTTN</a>) -- like, for the most part, AT&amp;T.</p> <p>Second, cable MSOs are beefing up the fiber content in their network diets. The deployment of DOCSIS 3.0 often is accompanied by an extension of the &quot;F&quot; part of HFC. Meanwhile, as cable operators increase their focus on servicing business customers or supporting wireless backhaul, fiber plays a large role. The recently announced DOCSIS Provisioning of EPON (<a href="/fttx/news/Sumitomo-Electric-Networks-completes-first-DPoE-10-interoperability-test-116633913.html">DPoE</a>) specifications testifies to the increasing number of DOCSIS-friendly fiber weapons now in the cable MSO arsenal.</p> <p>If you're attending the National Association of Broadcasters (NAB) conference next week, you can learn more about this at the Telecom Access Networks Conference. There I'll moderate a session entitled &quot;Cable Operators Respond to Telcos' Fiber Strategies.&quot; The session will feature a pair of speakers:</p> <p>Gary Schultz, <span id="ctl00_MainContent_lblContactTitle">president/principal Analyst</span> at <span id="ctl00_MainContent_lblContactCompany">Multimedia Research Group, Inc.</span>, will describe DOCSIS 3.0 and how quickly cable MSOs plan to roll it out.</p> <p>Shridhar Kulkarni, product manager, access network solutions, at Aurora Networks will then highlight the fiber options at the cable operators' disposal. These include fiber deep, DPoE, and RF over glass (RFoG).</p> <p>You can <a href="" target="_blank">find out more about the event on the NAB website</a>.</p> <p>It's a mistake to discount the level of investment cable operators have made in fiber. They haven't fully embraced FTTH -- but the same can be said for telcos as well.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/04/give-cable-operators-their-due.html2011-04-07T13:51:00.000Z2013-02-26T22:47:51.939ZCompete in the field, not in the legislaturenoemail@noemail.orgStephen Hardy<p>Incumbent carriers have used the opportunity presented by the wave of recently elected pro-business/anti-government politicians to once again attack municipal fiber to the home (<a href="/fttx/ftth-b">FTTH</a>) initiatives at the state level. The most successful example of this trend so far can be found in North Carolina, where a measure to &quot;level the playing field&quot; between commercially and publicly financed network operators recently passed the State House of Representatives and will now be considered by the State Senate (see <a href="">this story and related links</a> at the <em>TechJournal South</em> website.)</p> <p>As an alumnus of a North Carolina university (UNC Greensboro -- go Spartans!), I'm chagrined that a state with which I'm affiliated might limit the broadband opportunities of its citizens. The bill addresses what critics of muni broadband initiatives have commonly called &quot;unfair advantages&quot; that municipal broadband efforts might enjoy. The bill forbids municipalities from borrowing money for capital costs without voter approval (giving the incumbents a clear shot at scuttling any muni-based competition), from offering Internet services below cost, and from using funds from other city utilities to build their networks.</p> <p>Not that incumbent carriers would slash their own prices to levels well below those in neighboring markets or subsidize low-cost service provision with funds from their more lucrative businesses to prevent a newly rolled out municipal service from succeeding -- no, no, no.</p> <p>More often than not, if a municipality does have an advantage, it's the willingness to offer high-speed broadband services where the incumbent won't. That willingness must be strong, given the risks involved in launching any new broadband network, regardless of where its funding originates.</p> <p>Given the risk-averse nature of most politicians, the best way for incumbents to stave off competition from municipal network initiatives is to upgrade their infrastructures and provide the kind of high-speed broadband services envisioned in the National Broadband Plan. Such a strategy likely will nip any notions of a municipally backed broadband network in the bud.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/03/compete-in-the-field-not-in-the-legislature.html2011-03-29T19:50:00.000Z2013-02-26T22:49:00.393ZOFC/NFOEC Reporter's Notebook, Day 3noemail@noemail.orgStephen Hardy<p>The exhibition portion of OFC/NFOEC came to an end today with the announcement that attendance exceeded 11,000. That means both attendance and exhibitors numbers rose versus last year.<br /> <br /> Before we highlight today&rsquo;s conversations, we turn to our Corrections Department for two brief messages:</p> <ul> <li>Oclaro&rsquo;s president and general manager, optical networks solutions, is Terry Unter, not Unger. Terry Unger is Felix Unger&rsquo;s younger brother.</li> <li>Finisar&rsquo;s 100GBase-LR4 CFP will be available even sooner than I was told yesterday. The module will sample in the second half of this year and reach general availability in the first half of 2012, the company&rsquo;s senior manager of corporate communications, Victoria McDonald, informed me via email this afternoon.</li> </ul> <p>And now we resume our regularly scheduled programming:</p> <ul> <li>Paul Voois, CEO of ClariPhy Communications, calmly told me that Mr. Unter (that&rsquo;s Unter with a &lsquo;t&rsquo;) will have the ASIC that helps drive Oclaro&rsquo;s upcoming 40G coherent transponder in time to meet the announced sampling date of next month. ClariPhy continues to plan to make such a device available to the open market next year. 100G coherent is next on the docket, with Oclaro assured of being a partner. I got the impression that there may other partners as well.</li> <li>In a conversation that will form the main interview in our upcoming May/June issue of Lightwave, Huawei Technologies USA VP of Network Marketing and Product Management Reg Wilcox acknowledged that the company won&rsquo;t make much headway in the North American (and particularly the US) market until it can overcome the perception that it is a &ldquo;fast follower&rdquo; rather than a technology leader. To that end, the company showcased a prototype Optical Burst Transport Network (OBTN) system in the truck it parked on the show floor. The smarts in the Huawei system are in the receive end of the transmission. The company also unveiled the OSN8800 OTN switch.</li> <li>Opnext Chairman (and acting president and CEO) Harry Bosco reports that his company isn't suffering from the same market headwinds reported by Finisar and Oclaro -- primarily because the company isn't in the ROADM space and has enough diversity among its 40G customers to smooth over any issues with a particular customer. So not every optical subsystems supplier is in for a few quarters of inventory correction.</li> <li>Ciena's Joe Berthold is quite bullish on coherent technology in the metro, thinking that his company has the technology to make the economics work. Ciena has already has demonstrated 400 Gbps and 1 Tbps transmission with its in-house derived chip technology, he adds.</li> <li>u<sup>2</sup>t is sampling the second generation of its fully integrated 40G/100G coherent receivers. The devices include a beam splitting function in the same package size as the first generation device, which is fully qualified and in deliveries. A u2t device could be seen in the 100G coherent muxponder on display in Nokia Siemens Networks' truck.</li> <li>Discovery Semiconductor was demonstrating a 25-Gbaud photoreceiver for 850-nm VCSELs over multimode fiber.</li> <li>GigOptix highlighted a demonstration of its <a href="/equipment-design/news/GigOptix-40G-DPSK-optical-modulator-completes-Telcordia-high-temp-lifetime-tests-117398863.html">polymer-based 40G DPSK modulator</a>. Samples of the device, based on the company's Thin Film Polymer on Silicon (TFPS) technology, are now in the hands of potential customers. The module is significantly smaller than standard LiNbO3 devices. Dr. Avi Katz, chairman, president, and CEO of the company, also was pleased with the company's agreement to acquire RF technology company Endwave. In addition to adding RF/microwave components to the company's portfolio, GigOptix will get a large cadre of engineers, a new facilitiy, and Endwave's $20 million in cash. The ability to leverage the new engineering experise to work on coherent communications projects also holds appeal, Katz said.</li> </ul> <p>For more on OFC/NFOEC, see the <a href="/blog/OFCNFOEC-Reporters-Notebook-Day-1.html">Day 1</a> and <a href="/blog/OFCNFOEC-Reporters-Notebook-Day-2.html">Day 2 Reporter's Notebooks</a>.</p> <p>&nbsp;</p> <p><br /> &nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/03/ofcnfoec-reporters-notebook-day-3.html2011-03-11T04:59:00.000Z2013-02-26T22:48:47.894ZOFC/NFOEC Reporter's Notebook, Day 1noemail@noemail.orgStephen Hardy<p>Greetings from sunny LA, where financial analysts have rediscovered OFC/NFOEC -- although they all might be gone tomorrow after reviewing Finisar's Q4 guidance.</p> <p>The show is quite lively this year, the vibe energized by the comparatively large number of system vendors either exhibiting or walking the floor in search of attention. It seems that many of these vendors have chosen OFC/NFOEC as the best replacement for Supercomm. I wonder what the TIA: Inside the Network show is going to look like...</p> <p>Before I offer some highlights of my discussions today, here's a leftover tidbit from yesterday's OSA Executive Forum. Much has been made about the evolution from legacy circuit-switched, TDM based networks and services to packet. One naturally might wonder how long such a transition might take. Here's a clue from Scott Mountford, principal-network planning engineer at AT&amp;T Services. Discussing the need to support legacy TDM services, he mused, &quot;You know, I can't remember the last time we retired a service.&quot;</p> <p>So, yes, there is going to be long-term demand for OTN-enabled network platforms.</p> <p>And now, the news:</p> <ul> <li>The <a href="/equipment-design/news/10X10-MSA-ratifies-100-Gbps-module-specification-117401653.html">10x10 MSA</a> -- and what Google's success in getting it bootstrapped in such a short amount of time portends for who is driving the data market -- continues to be a hot topic of conversation. Founding MSA member Santur Corp. hosted an event early in the morning to discuss current milestones and future plans. It seems that there will be three module suppliers initially -- Santur (which is the only company to have shipped such modules so far), JDSU, and Oplink. And three just might be the magic number when it comes to adequately serving the market, said MSA Chairman Scott Kipp, senior technologist at Brocade. Such an arrangement is typical in the Fibre Channel space, he told me. The question will be when JDSU and Oplink bring 10x10 CFPs to market. Sources at JDSU acknowledged that they don't have such a module yet, but declined to say when that situation would be rectified. Expect to see more on this topic later.</li> <li>Juniper Networks (which has received some of Santur's 10x10G CFPs) is showing off its new <a href="/networking/news/PTX-Series-Packet-Transport-Switch-starts-Juniper-Networks-down-packet-optical-transport-path-117317128.html">PTX Series Packet Transport Switch</a>. VP Product Markeing Luc Ceuppens confirmed that it has a long-term arrangement with ADVA Optical Networking to supply the separate ROADM/amplifier/other optics shelf. Juniper will be happy to sell the whole package for greenfield applications where ROADM and related technology is not in place. Juniper's integrated optics effort won't obviate the need for separate optical transport gear, he allowed, but will certainly change the features that optical transport piece will require. Cueppens said that Juniper is taking a collaborative approach to the integrated optics, although it is leveraging its newly beefed up in-house capabilities as much as possible. The integrated optics should be ready for demonstration in the second half of this year, with GA about six months later.</li> <li>Stefan Ekman, CEO at dispersion compensation company Proximion, thinks the advent of coherent technology will actually help his business because he's betting DCMs won't go away, regardless of what the system houses might say. His customer checks also indicate that 40G will &quot;boom&quot; this year.</li> <li>Emcore is one of several vendors stepping up to give JDSU competition in the tunable XFP space. They've developed both zero and negative chirp XFPs to address both metro and long-haul applications. Devices are undergoing qualification with customers. Jaimie Reloj, VP of business development, also asserts the company is cleaning up with its narrow-linewidth external cavity light sources for 40G and 100G coherent applications. Right now the company is quite happy to stay out of the coherent module business.</li> <li>OFS/Furukawa Electric also has unveiled a tunable narrow-linewidth laser for coherent applications. Engineering samples will be available shortly, with production slated for this October. The company also unveiled a coherent mixer with built-in polarization multiplexer/demultiplexer for receiver applications, as well as a Raman amplifier with automatic gain control. It's typical for such amplifiers to be paired with EDFAs in 100G applications, which led me to ask whether the company plans to offer a hybrid unit. The source at the booth said that such a combination would be possible, but would likely be a custom development for a particular user. </li> <li>Red-C Networks already has an integrated hybrid Raman/EDFA on the market with enough intelligence to automatically sense when changes in gain are required and acti accordingly. Needless to say, a high degree of synchronization and software smarts is required, company sources acknowledged proudly. Red-C debuted the product last year at OFC/NFOEC. It's now being deployed, souces revealed. </li> <li>Ireland's Intune Networks is very close to first office carrier applications (one in the US, another in Europe) of its novel all optical packet switch. The company is also in the midst of assembling about $40 million in Series C funding -- the last funding round it will need, CTO John Dunne told me.</li> </ul> <p>I'll have more show highlights tomorrow, so stay tuned.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/03/ofcnfoec-reporters-notebook-day-1.html2011-03-09T02:09:00.000Z2013-02-26T22:48:36.976ZGoogle's optical shopping listnoemail@noemail.orgStephen Hardy<p>Google Senior Network Architect Bikash Koley appeared on a pair of panels at the Optical Society of America's Executive Forum March 7. This gave him plenty of opportunity to suggest areas of technology development to systems houses and component developers. Herewith a list of things on the Google optical wish list:</p> <ul> <li>A cost-effective label switched router with integrated 100G DWDM optics. (Juniper Networks was undoubtedly happy to hear this, given <a href="/networking/news/PTX-Series-Packet-Transport-Switch-starts-Juniper-Networks-down-packet-optical-transport-path-117317128.html">its recent PTX Series announcement.</a>) The LSR needs to have a capacity of 4 Tbps now, scalable to 8 Tbps next year.</li> <li>Intra-data center DWDM. As Google expands its data centers and moves from multimode to singlemode fiber, getting enough fiber in and out of their fabrics is a problem. So decreasing fiber requirements via DWDM would be a real help.<!--[if gte mso 9]><xml> <w:WordDocument> <w:View>Normal</w:View> <w:Zoom>0</w:Zoom> <w:PunctuationKerning /> <w:ValidateAgainstSchemas /> <w:SaveIfXMLInvalid>false</w:SaveIfXMLInvalid> <w:IgnoreMixedContent>false</w:IgnoreMixedContent> <w:AlwaysShowPlaceholderText>false</w:AlwaysShowPlaceholderText> <w:Compatibility> <w:BreakWrappedTables /> <w:SnapToGridInCell /> <w:WrapTextWithPunct /> <w:UseAsianBreakRules /> <w:DontGrowAutofit /> <w:UseFELayout /> </w:Compatibility> <w:BrowserLevel>MicrosoftInternetExplorer4</w:BrowserLevel> </w:WordDocument> </xml><![endif]--><!--[if gte mso 9]><xml> <w:LatentStyles DefLockedState="false" LatentStyleCount="156"> </w:LatentStyles> </xml><![endif]--><!--[if gte mso 10]> <style> /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Times New Roman"; mso-ansi-language:#0400; mso-fareast-language:#0400; mso-bidi-language:#0400;} </style> <![endif]--></li> <li>Low-cost, large port count (1000x1000) optical crossconnects cheaper than the crossconnects developed for telco long-haul applications.</li> <li>Pluggable, tunable DWDM optics with &gt;0.2 bps/Hz spectral efficiency for metro reach</li> <li>Variable bit rate 100G+ coherent optics for long-haul applications.</li> </ul> <p>He probably also wants more transceiver vendors to <a href="/equipment-design/news/10X10-MSA-ratifies-100-Gbps-module-specification-117401653.html">join the 10x10 MSA</a>, but he didn't mention that specifically. [UPDATE: Perhaps that's because he doesn't need more. Santur -- which has already shipped more than 1000 modules -- JDSU, and Oplink look to be in line to supply the devices. And that might be just enough.] A note here: While I've consistently mentioned lower cost as a reason for Google's push for such an MSA, Koley repeatedly cited the fact that a 10x10G module likely could be developed faster than a 4x25G module as the rationale behind the MSA effort. He praised the 4x25G approach as the way to go -- five years from now.</p> <p>Balancing these desires are a few things he doesn't want. These include:</p> <ul> <li>An OTN fabric or additional control plane in the optical layer for his long-haul links. Google prefers to have most of the network intelligence at the service layer; having intelligence at the optical layer gets in the way of this approach.</li> <li>40G</li> <li>Technology developed for telco applications that are over-engineered for most of his requirements and therefore cost more than he wants to pay.</li> <li>Module MSAs developed without user input. He cited the CFP MSA as an example.</li> </ul> <p>You now have your marching orders -- now get to it.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/03/googles-optical-shopping-list.html2011-03-08T02:00:00.000Z2013-02-26T22:48:26.484ZAT&T Labs questions contentionless ROADMsnoemail@noemail.orgStephen Hardy<p>After <a href="/blog/Verizons-Wellbrock-on-ROADMs.html">speaking to Glenn Wellbrock of Verizon</a> for my upcoming article on next-gen <a href="/about-us/lightwave-issue-archives/issue/ensuring-profitability-with-a-3g-roadm-system-53428682.html">ROADMs</a>, I had a chance to talk to Mark Feuer and Sheri Woodward, principal members of technical staff at AT&amp;T Labs, to get another viewpoint on what's ahead for ROADM requirements and deployments. </p> <p>While they were careful to say that they wouldn't speak about AT&amp;T's ROADM deployment plans, they had plenty of opinions about the current and future state of the art. Salient among these were their doubts about the usefulness of contentionless capabilities.</p> <p>It seems AT&amp;T has done studies that show that if you put a switch on the client side between the ROADM transponders and whatever is feeding traffic into the ROADM, you can solve virtually all contention problems within the ROADM itself. In addition, you also get such benefits as the ability to do bridge and rolls for traffic grooming and 1x<em>N</em> transponder protection.</p> <p>&ldquo;So I&rsquo;m not convinced we&rsquo;ll ever end up with an architecture where the ROADM part of the node is strictly contention-free, because if you have this switching on the client part of the node, the small bit of contention within the ROADM part might become insignificant,&rdquo; Woodward told me. &ldquo;And so the jury is still out on whether there will ever be a contention-free architecture that pays for itself.&rdquo;</p> <p>You can read the full article in the upcoming March/April 2011 issue of <em>Lightwave</em>.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/02/att-labs-questions-contentionless-roadms.html2011-02-22T22:10:00.000Z2013-02-26T22:49:36.150ZVerizon's Wellbrock on ROADMsnoemail@noemail.orgStephen Hardy<p>I just finished transcribing an interview with Glenn Wellbrock, director optical transport network -- architecture and design at Verizon, that I conducted for a cover story on next-gen <a href="/about-us/lightwave-issue-archives/issue/ensuring-profitability-with-a-3g-roadm-system-53428682.html">ROADMs</a> for our upcoming March/April 2011 issue. He had some very interesting things to say.</p> <p>For example, all the talk about colorless, directionless, and contentionless ROADMs is indeed just talk, he says -- at least when it comes to applying these features to the 100G-ready long-haul applications where Verizon would like to employ them first. Wellbrock says he expects to see early prototypes of such capabilities this year. Actual products -- which Verizon would like to see incorporated within <a href="/about-us/lightwave-issue-archives/issue/packet-optical-transport-systems-the-new-pots-54894827.html">packet-optical transport</a> systems -- will appear next year, he believes.</p> <p>He described these first-generation colorless/directionless/contentionless ROADMs as &quot;rudimentary&quot; in nature. &quot;Most [developers] are brute forcing it,&quot; Wellbrock says of initial designs. &quot;They&rsquo;re just using piece parts that can be purchased today and building a node out of them. There are more optimal designs that I think will show up in the 2013, even 2014 timeframe that will provide another layer of integration and lower cost, lower power.&quot;</p> <p>Interestingly, he says ROADM developers still haven't told him how much they plan to charge for these new features.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/02/verizons-wellbrock-on-roadms.html2011-02-15T15:04:00.000Z2013-02-26T22:49:24.526ZCable operators as an investment opportunitynoemail@noemail.orgStephen Hardy<p>Sure, we're seeing private equity jump into the regional carrier market and bankrolling much of the consolidation now underway in that space. But Gillis Cashman, a general partner at private equity firm M/C Venture Partners (which has a stake in the Zayo Group, among other network consolidators), thinks that cable operators also may be an attractive play.</p> <p>Well, some cable operators, anyway.</p> <p><!--[if gte mso 9]><xml> <w:WordDocument> <w:View>Normal</w:View> <w:Zoom>0</w:Zoom> <w:PunctuationKerning /> <w:ValidateAgainstSchemas /> <w:SaveIfXMLInvalid>false</w:SaveIfXMLInvalid> <w:IgnoreMixedContent>false</w:IgnoreMixedContent> <w:AlwaysShowPlaceholderText>false</w:AlwaysShowPlaceholderText> <w:Compatibility> <w:BreakWrappedTables /> <w:SnapToGridInCell /> <w:WrapTextWithPunct /> <w:UseAsianBreakRules /> <w:DontGrowAutofit /> <w:UseFELayout /> </w:Compatibility> <w:BrowserLevel>MicrosoftInternetExplorer4</w:BrowserLevel> </w:WordDocument> </xml><![endif]--><!--[if gte mso 9]><xml> <w:LatentStyles DefLockedState="false" LatentStyleCount="156"> </w:LatentStyles> </xml><![endif]--><!--[if gte mso 10]> <style> /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Times New Roman"; mso-fareast-font-family:"Times New Roman"; mso-ansi-language:#0400; mso-fareast-language:#0400; mso-bidi-language:#0400;} </style> <![endif]--></p> <p class="MsoNormal">&quot;I think cable is going to be a very interesting space over the next couple of years,&quot; he told me in an interview you'll find recapped in the latest issue of Lightwave. &quot;I don&rsquo;t think it becomes distressed. But it&rsquo;s one where you&rsquo;ve got to really dig in and figure the ease of the transition [from being video-focused to a full broadband services provider] with respect to the profile of the company. And where it is [located]. If you&rsquo;re investing in cable, going head-to-head with Verizon FiOS is not a great position to be in. But if you&rsquo;re more in the suburban or rural areas of the country and you own the best and highest capacity network into the home, you&rsquo;re going to be in a great position long term.&quot;</p> <p class="MsoNormal">You can <a target="_blank" href=";mode=2">read the full interview online</a> (no registration required).</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/01/cable-operators-as-an-investment-opportunity.html2011-01-28T15:21:00.000Z2013-02-26T22:50:29.860ZThere's still light in Light Peaknoemail@noemail.orgStephen Hardy<p>By now you may have heard that Intel will offer a copper version of Light Peak, its 10-Gbps interconnect technology for consumer applications (see <a href="/about-us/lightwave-current-issue/Intel-plots-Light-Peak-interconnect-revolution.html">&quot;Intel plots Light Peak interconnect revolution&quot;</a>). While the news made me wonder whether the project has hit a snag, one of the Light Peak component suppliers tells me that the decision to pursue the copper version was customer driven.</p> <p>Ensphere Solutions is providing optical transceiver chips for Light Peak (see <a href="/equipment-design/products/Ensphere-Solutions-joins-Light-Peak-pursuit-with-transceiver-IC--63806127.html">&quot;Ensphere Solutions joins Light Peak pursuit with transceiver IC&quot;</a>). According to Al Gharakhanian, VP of marketing at Ensphere, there currently are two major PC manufacturers who have decided to adopt the technology. However, one of them has decided they would prefer to start with a copper version, then evolve toward optical.</p> <p>The other customer remains committed to the optical version, Gharakhanian asserts. In fact, he says his company expects to increase Light Peak transceiver IC shipments significantly in the first quarter of this year.</p> <p>Gharakhanian didn't say who the customers were. However, Apple has been touted by reporters in the computer space as one of the drivers for the technology's development. And Intel has quoted positive comments from Sony in its promotion of Light Peak technology.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/01/theres-still-light-in-light-peak.html2011-01-20T20:40:00.000Z2013-02-26T22:50:17.165ZYou know nothing of loyaltynoemail@noemail.orgStephen Hardy<p>Or at least you display very little of it if you're a typical buyer of optical transceivers and transponders. That's at least one of the conclusions we've reached through our <em>2010 Transceiver/Transponder Buyer Survey.</em></p> <p>As was the case with <a href="/about-us/lightwave-issue-archives/issue/readers-reveal-transceivertransponder-purchase-factors-53440722.html">similar transceiver/transponder buyer surveys in 2007</a> and <a href="/about-us/lightwave-issue-archives/issue/readers-renew-transceivertransponder-purchase-discussions-54889197.html?c=no">2008</a>, we polled those members of our audience database who purchase optical modules and asked them a series of questions about what they're buying, how they're buying, and who their favorite suppliers are.</p> <p>One thing we asked is how likely buyers are to change transceiver/transponder suppliers during the lifetime of a given application. A mere 27.5% said they either definitely wouldn't or probably wouldn't change their vendor.</p> <p>Even given the fact that the remaining 72.5% includes those who responded &quot;I'm not sure,&quot; that's not a heck of a lot of commitment -- or loyalty.</p> <p>The report also includes some interesting data on brand recognition (or lack thereof), how quickly 40G and 100G devices will approach mainstream status, and which suppliers are most likely to have a good year in 2011, among other things.</p> <p>You can <a href="/education/research/">get your hands on a copy of the report this way</a>...</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/01/you-know-nothing-of-loyalty.html2011-01-13T20:28:00.000Z2013-02-26T22:50:06.653ZThe right choice for Opnextnoemail@noemail.orgStephen Hardy<p>As <a href="/business/news/Opnext-to-launch-search-this-month-for-Bouchard-replacement-113008139.html">the story I posted today</a> indicates, Opnext plans this month to warm to the task of replacing Gilles Bouchard as president and CEO. While his voice won&rsquo;t be the lone one in the room during this month&rsquo;s board meeting, interim President and CEO Harry Bosco listed knowledge of the optical industry, operations expertise, and &ldquo;international flare&rdquo; (which I took to mean the ability to work with Hitachi executives in Japan) as his criteria for the position.<br /> <br /> Those who want to speculate about who will land the job might combine the above attributes with a look at Bouchard&rsquo;s background, since the fact the job is open pertains to personal, not performance issues.<br /> <br /> Bouchard joined Opnext as chief operating officer in November 2007 from HP, where he had spent 17 years. His last position at HP was executive vice president of global operations. Combined with a few years in the Opnext COO spot to hone his skills at dealing with his Japanese counterparts, Bouchard certainly would seem to have had both the operational expertise and international flair attributes covered by the time it was <a href="/general/opnext-completes-stratalight-acquisition-names-new-management-54891962.html">announced that he would step into Bosco&rsquo;s shoes</a> in 2009.<br /> <br /> Bosco pointed out that the management team Bouchard left behind has a considerable amount of expertise, in his opinion. &ldquo;Most of the senior management has been with me for eight to ten years,&rdquo; he said. &ldquo;They pretty well understand the industry.&rdquo;<br /> <br /> So <a href="" target="_blank">a look at internal candidates</a> is undoubtedly in order. The company currently doesn&rsquo;t have a COO, but it has four people who are president of some aspect of Opnext&rsquo;s business as well as a vice president of operations (Scott Clark) with more than 20 years of experience, including 10 working in Japan.<br /> <br /> Meanwhile, the search committee will also look outside of the company. The recent spate of consolidation means that a few people with management experience at an optical module/subsystem company might be getting itchy (non-compete agreements notwithstanding).<br /> <br /> However, I wouldn&rsquo;t be surprised if the committee scouts for someone with experience at an optical systems house. There&rsquo;s nothing better for maintaining customer focus than having been one.<br /> <br /> I&rsquo;ve just started thinking about who might be in the running. If you want to make some suggestions, feel free to do so in the comments section below.<br /> &nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2011/01/the-right-choice-for-opnext.html2011-01-06T20:35:00.000Z2013-02-26T22:49:55.728ZHints about my holiday presentsnoemail@noemail.orgStephen Hardy<p>The holiday season is always a busy time -- usually so busy that projects you should be working on get pushed to the side. This is particularly true for your humble narrator; nevermind getting out Xmas cards (I haven't) -- I'm in the midst of trying to write three articles (on OTN testing, DOCSIS Provisioning of EPON, and OFC/NFOEC 2011) and a department for Lightwave's January/February 2011 issue in four days. Therefore I'm having a bit of trouble doing anything with a couple of choice interviews as well as some contributed articles that will now serve as holiday presents (because that's about when I'll get a chance to work on them).</p> <p>Here are a few hints:</p> <p>Yesterday I interviewed Harry Bosco, past and now current president and CEO of Opnext. He insisted that his successor/predecessor <a href="/blog/Bouchard-out-Bosco-back-in-at-Opnext.html">Gilles Bouchard resigned</a> for personal reasons, and that Bouchard's replacement will need to be well versed in the optical communications industry as well as have the expected operational and technical expertise. Bosco also said reaching profitability ASAP was the company's number one goal -- and talked about how the company will achieve it.</p> <p>I also spoke recently to ADVA Optical Networking CEO Brian Protiva, who offered his predictions for 2011. These include an acceleration in the evolution from legacy to packet-based networks, driven by growing bandwidth demands from residential users (via video) and business users (with cloud computing and low latency among the catalysts) against a backdrop of increasing mobility. Needless to say, he believes his company is well positioned to take advantage of these trends. He also had interesting comments on <a href="/about-us/lightwave-issue-archives/issue/packet-optical-transport-systems-the-new-pots-54894827.html">packet optical transport</a> and why he prefers the market in India to that in China.</p> <p>Speaking of low latency, BTI Systems has written an article you'll see soon that answers in the affirmative the question &quot;Does anybody not involved in algorithmic trading care about <a href="/networking/news/Timing-everything-when-selling-low-latency-network-services-94110129.html">low latency networks</a>?&quot;&nbsp; The examples are compelling. </p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/12/previews-of-coming-attractions.html2010-12-21T15:42:00.000ZBouchard out, Bosco back in at Opnextnoemail@noemail.orgStephen Hardy<p>Gilles Bouchard has resigned from his position as CEO and president of optical module vendor Opnext. Non-executive Chairman of the Board Harry Bosco, the man Bouchard replaced in April 2009, will slip on his old shoes on an interim basis while the company's board of directors searches for a permanent replacement.</p> <p>I conducted video interviews with Bouchard <a href="/equipment-design/video/51932427.html?player=31834881001&amp;title=18160942001">shortly before</a> he assumed his now former position and <a href="/equipment-design/video/51932427.html?player=31834881001&amp;title=76657828001">one year into his tenure</a>.</p> <p>Bosco currently is in Japan (Opnext is an offshoot of HItachi). I'm working on getting an interview with him now.</p> <p>Meanwhile, the question lingers whether Bouchard's resignation was entirely his own idea. The company touted quarter-on-quarter revenue and gross margin improvements November 4 when it announced its second quarter fiscal results. &quot;At $86.4 million, revenue in the September quarter set a new high for Opnext, largely based on strong 40-Gbps and above module sales,&quot; Bouchard was quoted as saying in the results announcement. Opnext offered revenue guidance of $87million to $92.0 million in the following quarter, which ends this month.</p> <p>However, the results still represented a loss. Meanwhile, Finisar (which <a href="/business/news/Finisar-confirms-record-revenues-and-profitability-111189489.html">also reported record revenues</a>) and JDSU have each achieved profitability.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/12/bouchard-out-bosco-back-in-at-opnext.html2010-12-13T20:26:00.000ZNew details emerge on Verizon 100GbE deploymentnoemail@noemail.orgStephen Hardy<p>Thanks to a conversation late last week with Glenn Wellbrock, director optical transport network -- architecture and design at Verizon, I have a few more details about the 100 Gigabit Ethernet deployment announced November 11 (see <a href="/networking/news/Verizon-to-deploy-100-Gigabit-Ethernet-on-Paris-to-Frankfurt-route-107244563.html">&quot;Verizon to deploy 100 Gigabit Ethernet on Paris-to-Frankfurt route&quot;</a>).</p> <p>Glenn confirmed that the new 100 Gigabit Ethernet deployment will use the same Ciena OME 6500 platform as the existing 100-Gbps link announced in December 2010 (see <a href="/networking/news/Nortel-unveils-commercially-available-100G-optical-system-79220847.html">&quot;Nortel unveils commercially available 100G optical system&quot;</a>). While that link used a 10x10 muxponder on the client side, this one will use a direct 100 Gigabit Ethernet connection via CFP modules. While there currently are no details regarding the module vendor, Finisar provided modules for a 100 Gigabit Ethernet field trial involving Juniper Networks and Verizon last March (see <a href="/networking/news/Verizon-Juniper-NEC-Finisar-conduct-100G-field-trial-86874052.html">&quot;Verizon, Juniper, NEC, Finisar conduct 100G field trial&quot;</a>). Finisar has not yet responded to an email regarding the company's potential role in the upcoming deployment.</p> <p>The new 100 Gigabit Ethernet stream will also share the same fiber as the existing 100-Gbps link, as well as an undisclosed number of 10-Gbps wavelengths.</p> <p>Wellbrock declined to speculate on future 100G deployment plans, either in Europe or North America.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/11/new-details-emerge-on-verizon-100gbe-deployment.html2010-11-15T15:32:00.000ZOutsourced IP demand strong at 40G and 100Gnoemail@noemail.orgStephen Hardy<p>While doing research for the webcast on outsourcing intellectual property (IP) for 40G and 100G technology development applications, I discovered that while some may doubt the long-term demand for 40-Gbps technology, you wouldn't know it by how much IP outsourcing vendors are selling for it.</p> <p>This is not to say that 100G isn't more popular, just that 40G isn't dead. For example, sources at the webcast sponsor, Avalon Microelectronics, said that 44% of the company's business last year came from 40G applications. Most of their business will be 100G this year (around 70%, the sources predicted), but 40G will still command 18%, the second-largest share of total revenue.</p> <p>Meanwhile, Jim Keszenheimer, business development manager at ViaSat -- which is delivering DSP and FEC IP for coherent applications -- reports that IP demand for 40G applications remains very strong.</p> <p>You can still catch the webcast, &quot;40G/100G Design Choices: Make or Buy IP,&quot; <a href="/webcasts/40G100G-Design-Choices-Make-or-Buy-Standard-Based-IP.html">from free our archives</a>.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/11/outsourced-ip-demand-strong-at-40g-and-100g.html2010-11-11T15:49:00.000ZSCTE Cable-Tec Notebook, Day 2noemail@noemail.orgStephen Hardy<p>We interrupt our regularly scheduled cableco optical access programming to bring you this special presentation on...100G?</p> <p>Yes, even at SCTE Cable-Tec, 100G cannot be avoided. That's because all that video programming (and burgeoning business services, etc.) requires a robust backbone. So, several vendors figure, why not see whether the major MSOs are interested in 100G?</p> <p>Alcatel-Lucent has a corner of its booth devoted to the topic. But perhaps the most aggressive systems vendor at the show when it comes to 100G is Fujitsu Network Communications, which is offering a 100G technology demo in the booth. Tom McDermott, distinguished strategic planner at Fujitsu, said that his company fully plans to follow the OIF's lead when it comes to modulation format, which means <a href="/about-us/lightwave-issue-archives/issue/is-dpndashqpsk-the-endgame-for-100-gbitssec-54890687.html">DP-QPSK</a> with coherent detection. Fujitsu naturally will leverage in-house expertise, which means optical components, such as a modulator and receiver hybrid from Fujitsu Optical Components (several of which were also on display) as well as electronics from Fujitsu Semiconductor Europe that will enable soft-decision FEC. Despite the demonstration, McDermott admits Fujitsu is still working out some of the integration aspects of its final product, which he says won't be generally available until Q4 of next year. Fujitsu plans to have both a transponder card and a 10x10 muxponder. The company hasn't yet done any carrier trials -- at least not any they're willing to talk about.</p> <p>How much interest do MSOs have in 100G? Well, &quot;the booth was pretty full yesterday,&quot; McDermott said early Thursday morning.</p> <p>Meanwhile, Fujitsu is moving full-speed ahead with its 40G coherent offering, announced earlier this year. The company has been quite about technical details of its 40G coherent approach -- McDermott would only say that it's different than what Fujitsu is doing for 100G -- although it's believed Fujitsu is using the CP-QPSK technology developed by CoreOptics (<a href="/business/news/Cisco-wraps-up-CoreOptics-acquisition-98602669.html">now part of Cisco</a>). McDermott says that Fujitsu does not have plans to replace its current 40G coherent approach with DP-QPSK/CD.</p> <p>Menara Networks, makers of the &quot;transponder in a pluggable module,&quot; also has its sites on 100G, up from the current 10G offering. VP, Product Line Management Adam Hotchkiss wouldn't provide many details of Menara's technical approach when I spoke with him on Wednesday -- but it won't be DP-QPSK/CD. The OIF-favored modulation format is fine for long-haul applications, Hotchkiss allowed, but probably is too expensive for the metro/regional space Menara Networks has in mind, he said. He expects to have a 100G product in CFP format in about 18 months. Menara <a href="/mso-optics/news/Menara-Networks-announces-North-American-MSO-selection-of-its-OTN-system-in-an-XFP-70307947.html">has had success with its 10G line in the cableco space</a>, supplying two of the top three MSOs for business service overlays, Hotchkiss asserted.</p> <p>However, not everyone is gung-ho on the 100G market. Scott Wilkinson at the Hitachi Communications Technologies America booth said his company certainly will have a 100G offering (courtesy in large part to the company's close personal friends at Opnext). However, he's not sure there's going to be a large market for it in the near term. He notes that only a small handful of 100G routes have been fielded so far; given how quickly 10G prices are going down, it seems likely to him that carriers with sufficient fiber will find it more economical to continue to add 10G wavelengths to meet growing bandwidth demands.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/10/scte-cable-tec-notebook-day-2.html2010-10-22T00:37:00.000ZSCTE Cable-Tec Notebook, Day 1noemail@noemail.orgStephen Hardy<p>SCTE Cable-Tec, one of the major conferences on the cableco schedule in the U.S., is off to a good start in New Orleans. Yesterday's opening session was free of the &quot;We have fiber too you know&quot; defensiveness that marked last year's opening. However, it was acknowledged that the cable companies overall had seen a decline in subscribers, blame for which was placed with both the economy and competition.</p> <p>Some of that competition comes from over the top (OTT) sources. In a morning panel discussion, Motorola CTO Office Dr. Robert Howald asserted that the attributes that make OTT popular are convenience, flexibility, and choice -- with price not as much of a factor as one might believe. Dr. Howald asserted that cablecos can compete with OTT on these terms, thanks to their networks.</p> <p>Speaking of networks, an interesting exchanged occurred in a subsequent technology panel when Michael Hayashi, EVP, architecture, development, and engineering at Time Warner Cable, was put on the spot about his company's DOCSIS 3.0 deployments -- or, more specifically, why they lagged those of fellow panelists from Comcast and Cox. Hayashi in essence said that TWC's DOCSIS 3.0 rollouts are being driven mainly by competitive forces; where those forces don't exist, they're not in a hurry to switch.</p> <p>Meanwhile, out on the show floor, JDSU's Jim Walsh reported that cablecos deploying DOCSIS 3.0 were reporting a greater incidence of laser clipping than anticipated. The higher power needed to support the wider, higher order modulation carriers DOCSIS 3.0 requires are the cause. The return path is a particular area of potential trouble, Walsh reported.</p> <p>Elsewhere, RF over glass (<a href="/mso-optics/featured-articles/RFoG-plus-PON--Enabling-cables-all-IP-future-64581082.html">RFoG</a>) seems to be growing in popularity -- at least among vendors. Titan Photonics showed off a complete line of RFoG gear; high performance is the line's calling card, according to Director of Product Marketing Charlie Chen.</p> <p>In one of the more unusual displays on the floor, 4Cable TV earned a place in the Green Pavilion with an EDFA compatible with RFoG powered by a solar cell array. The EDFA's small power draw enables the eco- (and cost-) friendly arrangement, says President/CEO Steven Richey. The company also has leveraged ROPA technology from the undersea world to create a passive EDFA. Richey says the company has already seen business from broadband stimulus winners.</p> <p>Aurora Networks, which earlier in the year <a href="/mso-optics/news/Is-RFoG-a-rural-only-technology-87738577.html">expressed its believe that RFoG was perhaps a niche technology</a> for rural applications, nevertheless filled out its RFoG offering with new diplexer/return receivers, a PON filter module, two high-powered optical amplifiers and an MDU CPE device.<br /> &nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/10/scte-cable-tec-notebook-day-1.html2010-10-21T14:00:00.000ZReport from Chinanoemail@noemail.orgSusan Smith<p>Last month, I attended <a href="" target="_blank">Lightwave China&rsquo;s</a> <a href="" target="_blank">FTTH China 2010 Conference</a> in Shenzhen, which was collocated with the 12th China International Optoelectronic Expo (<a href="" target="_blank">CIOE</a>). As a first-time visitor, I found Shenzhen to be a modern city undergoing a fairly orderly transition--not what I had envisioned from the past decade of negative press about China&rsquo;s rapid growth.<br /> <br /> Until recently, Shenzhen was a small village bordering Hong Kong. Just 30 years later, thanks in large part to its designation as China&rsquo;s first Special Economic Zone, Shenzhen has a population of more than 14.5 million that puts it on par with cities the size of Los Angeles, Beijing, and Moscow.<br /> <br /> Many Shenzhen residents live and work in high-density apartment and office buildings served with FTTB, but those that I spoke with long for more green space and single-family homes. Given those sentiments, it&rsquo;s not hard to understand why the consistent forecast from speakers at the FTTH China Conference was for <a href="/fttx/ftth-b">FTTH</a> growth to quickly outpace China&rsquo;s FTTB demand. With so many people in cities like Shenzhen accustomed to the benefits of FTTB, why wouldn&rsquo;t they demand the same once they can buy a house?</p> <p>Outside of the conference, a visit to <a href="" target="_blank">Huawei&rsquo;s</a> lush, green campus and product demo area was a flashback to my visits to the Cisco and 3Com campuses in San Jose more than 10 years ago -- impressive, modern and PR-aware. The Huawei representatives I met, including Director of Access Product Line Marketing Jim Wan, described the company&rsquo;s patient strategy for expanding its systems business in North America. Just as it has done in Europe, Huawei plans to increase its North American market share by slowly winning over customers with products that reflect a deep understanding of their specific needs. This long-term approach may be one of the greatest differentiators between China&rsquo;s business culture and that of the U.S. in recent years.</p> <table cellspacing="1" cellpadding="1" border="0" align="center" width="300"> <tbody> <tr> <td><img height="225" width="300" alt="Monitors in Huawei's lobby" src="" /></td> </tr> <tr> <td><strong>In a scene reminiscent of Silicon Valley, monitors dotted the lobby of Huawei's Shenzhen campus.</strong></td> </tr> </tbody> </table> <p>&nbsp;</p> <p>I also visited Shenzhen <a href="" target="_blank">Gigalight Technology&rsquo;s</a> headquarters in the city&rsquo;s bustling downtown technology park area. Li Zhendong, the company&rsquo;s chairman and a Lightwave reader for many years, ties Gigalight&rsquo;s value proposition to innovation. While he was complimentary of Lightwave&rsquo;s recent innovations, Mr. Li challenged us to continue improving. I found this competitive and ambitious spirit in many of the people I met in China.</p> <table cellspacing="1" cellpadding="1" border="0" align="center" width="300"> <tbody> <tr> <td><img height="171" width="300" alt="Lightwave's Susan Smith, Gigalight's Lee Zhendong, Lightwave China's Adonis Mak" class="enlargeable" title="Click to Enlarge" original="/content/dam/etc/medialib/new-lib/lw/blogs/2010/10/lw_gigalight.jpg" src="/content/dam/etc/medialib/new-lib/lw/blogs/2010/10/lw_gigalight.jpg/_jcr_content/renditions/pennwell.web.300.171.jpg" /></td> </tr> <tr> <td><strong>Gigalight CEO Li Zhendong (center), with Lightwave Publisher Susan Smith (left) and Lightwave China Publisher Adonis Mak (right)</strong></td> </tr> </tbody> </table> <p><br /> <br /> Despite Shenzhen&rsquo;s many successes, the city faces its share of challenges. Many employees there struggle with a rapidly rising cost of living. And there is speculation about how well the city will function as the government withdraws the support that has helped it grow so quickly. How the next few years play out in Shenzhen is not only important for China, but could have tremendous repercussions for those involved in optical networking in other parts of the world.<br /> &nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/10/report-from-china.html2010-10-13T06:54:00.000ZLightwave ONE, Day 2noemail@noemail.orgStephen Hardy<p>You can't talk about fiber in the enterprise without discussing copper, it seems. As one of the focal points of Day 2 of Lightwave's Optical Networks for Enteprises (ONE) Conference was cabling the data center, that meant a look at where copper might fall short when it comes to high data rates. &quot;Might&quot; is the operative word, as it seems twisted pair has more lives than your average movie monster.</p> <p>For example, it had been confidently predicted that the advent of 40 and 100 Gigabit Ethernet would finally establish the upper boundary of the current generation of twisted pair. While there are copper-based PMDs within IEEE 802.3ba, surely Cat6A's days are numbered, right?</p> <p>Not so fast, reported JDSU's Assaji Aluwihare at the end of the first day of the conference. Studies are already underway of whether Cat6A can transmit 40GbE for a useful distance -- and preliminary reports are encouraging, at least when measured at 1600 MHz bandwidth, he says.</p> <p>However, Doug Coleman of Corning Cable Systems told attendees on the second day that Cat6A -- not to mention proposed Cat7A -- will have a hard time getting standards acceptance. Coleman, who attends Ethernet Alliance meetings on Corning's behalf, reports that organization has already shot down a proposal to work on specs for Cat7A.</p> <p>Still, it's clear that copper isn't going away any time soon -- which even members of the TIA Fiber Optic LAN Section admitted during their panel session that followed Coleman's presentation. At 10G, 10GBase-T will enable copper-heads with shorter runs to still use their preferred medium. Offline, one panelist predicted that power over Ethernet (PoE) may be another factor that will maintain copper's place in enterprise networks, even as wireless encroaches.</p> <p>But fiber has a role to play in PoE applications as well, according to a presentation from Ty Estes of Omnitron Systems. The combination of fiber for long runs and PoE-enabled media converters for copper runs over the last few feet can extend the reach of PoE applications. IP camera networks will be one area where fiber-enabled PoE will have an impact, he predicted.</p> <p>Turning to security, Zeev Draer of MRV Communications lauded the security provided by hardware-based encryption at Layer 1. He said that government networks are using a highly secure Layer 1 quantum encryption mechanism that should find its way into the commercial arena in the not too distant future.</p> <p>&nbsp;</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/10/lightwave-one-day-2.html2010-10-06T12:00:00.000ZLightwave ONE, Day 1noemail@noemail.orgStephen Hardy<p>The first day of the Lightwave Optical Networks for Enterprises (ONE) Conference kicked off yesterday to a room full of enterprise and service provider managers as well as others interested in the use of fiber optics to meet enterprise requirements. Highlights included:</p> <p>In separate presentations, Cisco's Kyle Hollasch and Alcatel-Lucent's Andres Belloni asserted that ROADMs have now become essential -- and economical -- for enteprise WAN private optical networks.</p> <p>Belloni also predicted that 400-Gbps optical systems will be a reality in two years.</p> <p>Several speakers also said that 100-Gbps technology is a near-term requirement. For example, NoaNet's Michael Henson reported that his companies have customers looking for 100G links now -- only a year or two after these same customers asserted that they couldn't imagine needing it. The lack of an IEEE 40GbE serial transport standards was one factor in going straight from 10G to 100G, he said.</p> <p>Shawn Morris of NTT&nbsp;America acknowledges that financial institutions are taking advantage of NTT America's low-latency capabilities. But the customers who were most vociferous about latency issues are video service providers.</p> <p>VMware's Svein Karevik shared the hurdles his company faced getting connectivity between data centers.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/10/lightwave-one-day-1.html2010-10-05T11:50:00.000ZA final look at ECOCnoemail@noemail.orgStephen Hardy<p>Having dodged a national strike in France last Thursday, I'm back in the office and reviewing my notes from ECOC in Turin. While I don't have official numbers from the show organizers yet, my impression is that the show topped last year's in terms of attendance. Certainly the companies on the floor with whom I spoke expressed approval of the traffic levels and quality.</p> <p>As for the conference, the only complaints I heard concerned the conference rooms themselves. Most were temporary structures, and you could occasionally hear what was going on in the room next to you.</p> <p>But overall, the bar has been set for Geneva next year, September 18-21. Meanwhile, here are a few salient tidbits that didn't make my earlier notebooks (see <a href="/blog/ECOC-Reporters-Notebook-Day-1.html">the Day 1 notebook</a> and that for <a href="/blog/ECOC-Reporters-Notebook-Day-2.html">Days 2 and 3</a>):</p> <ul> <li>With all the talk about 40G/100G from the technology suppliers, the test equipment developers stood ready to assist. Companies such as JDSU (with its ONT 100G), Agilent (with its N4391A optical modulation analyzer), and EXFO (with its PSO-200 Optical Modulation Analyzer) showed off established products, the latter two primarily aimed at coherent modulation applications. Optametra also is active in this field, and the company's VP of engineering, Daniel van der Weide, was showing off its OM4106 Coherent Lightwave Signal Analyzer in the LeCroy booth, paired with LeCroy's WaveMaster 830 Zi oscilloscope. van der Weide reported that, judging by the level of expertise visitors to the booth displayed, technology developers were now up to speed on coherent technology overall, and were asking questions about how to perform specific measurements. This trend illustrates the growing maturity of the technology, as well as the fact that carriers will soon deploy coherent-enabled systems in more than ones and twos -- which probably will mean field versions of the constellation analyzers Optametra and its competitors currently supply for the lab. That idea has occurred to van der Weide as well, and he's already started thinking about how to pull off such a transition. No word on when that might happen, however.</li> <li>LeCroy, meanwhile, touted an upgrade to the WaveMaster 830 Zi that will up its capabilities from 4x30 GHz to 4x45 GHz by the first month of 2011. van der Weide was excited about the kinds of measurements one could do with that kind of horsepower. The scope has more high-speed uses than as a platform for coherent analysis, however, including 40/100 Gigabit Ethernet technology testing. LeCroy also showed off a soon-to-be announced S-parameter analyzer, called the SPARQ (for &quot;S-parameter Quickly,&quot; according to a wag at the booth).</li> <li>Anritsu highlighted its <a href="/test-and-measurement/news/Anritsu-intros-MD1260A-40100G-Ethernet-Analyzer-for-lab-manufacturing-and-field-testing-102911909.html">newly released MD1260A 40/100G Ethernet Analyzer</a>.</li> <li>However, not everyone needs all the bells and whistles -- and the expense -- of the fancy BERT and communication analyzer test sets, according to Picosecond Pulse Labs. For these users, PPL offers the new Model 12060 20-GHz pattern generator. It offers programmable data rate, amplitude, offset, and crossing point, among other features. It supports single bit error injection; a built-in jitter injection option is coming soon.</li> <li>Luceo Technologies of Germany displayed the L-6001-PHS-1 Phase Shifter Module for the XBERT and ParalleX chassis. The module splits the REF-CLK into six different REF-CLK outputs; each of these outputs can be delayed by up to +/- 200 ps. The feature enables users to align the output patterns of multiple Luceo E-BERT modules or introduce well-defined delays. This would help with crosstalk measurements in multichannel applications, Luceo says.</li> </ul> <p>Meanwhile, what's good for coherent technology is good for Raman amplifiers and Raman/EDFA hybrids. <a href="/equipment-design/news/RED-C-intros-Hybrid-Raman-EDFA-module-for-coherent-networks-102837689.html">RED-C Optical Networks unveiled one</a>, Furukawa Electric displayed one, and JDSU has one too. Raman also is good for low-latency networks -- an application whose 15 minutes of fame are just about up, sources agreed, unless someone can find a demand for it beyond financial networks.</p> <p>Finally, Oclaro will finally give JDSU some competition in the tunable XFP space, thanks to <a href="/equipment-design/news/Oclaro-unveils-full-band-tunable-XFP-transceivers-103438009.html">its newly announced module</a>. However, JDSU has taken full advantage of being first to market, and has signed up 32 customers for its device. Meanwhile, a source at Opnext labeled his company's failure to have a tunable XFP on the market &quot;a disappointment.&quot;</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/09/a-final-look-at-ecoc.html2010-09-27T13:12:00.000ZECOC Reporters' Notebook, Days 2 & 3noemail@noemail.orgStephen Hardy<p>After a day so busy I didn't have time to post anything, I'm back with more from Turin. My schedule on Day 2 mirrored activity on the show floor, which was quite busy.</p> <ul> <li>Someone at the Oclaro booth remarked that the sudden interest in 100G at last year''s ECOC caught him and others by surprise. Companies clearly came prepared this year, as&nbsp;40G and 100G coherent technology dominates the show floor: One of the more&nbsp;interesting companies discussing&nbsp;their role in&nbsp;coherent detection development is ViaSat. As its name implies, the company cut its teeth in&nbsp;the satellite realm -- which, representatives at the booth said, makes optical transmission a&nbsp;piece of cake by comparison.&nbsp;The company is offering FEC and DSP IP, which they'll supply in ASIC form if required. Not surprisingly, given the fact&nbsp;that these have been considered two of the more&nbsp;difficult&nbsp;pieces to fit properly within the coherent receiver puzzle, they say they&nbsp;have customers for their IP&nbsp;already. They also say they have QAM and OFDM expertise&nbsp;ready for what comes next.&nbsp;</li> <li>The show continues to buzz about a debate during a&nbsp;Market Focus presentation on Monday between Bikash Koley of Google and Chris Cole (sp?) of Finisar regarding the merits of 10x10G versus 4x25G for 100 Gigabit Ethernet. The gentleman from Google asserted that 10x10G would be superior, less complex, and less costly; the gentleman from Finisar respectfully disagreed. The debate highlighted a potential question about who will drive the 100GbE market. Cisco is said to favor 4x25G, and therefore many module vendors are following that lead. However, can large customers such as Google (potentially with the assistance of other large data center customers such as Facebook and Amazon) turn the tide toward 10x10G? A Santur emitter array for 10x10G is on display within the JDSU booth; a JDSU source says this signals an arrangement designed to last longer than four days in Turin.</li> </ul> <p><strong>1:20 PM:</strong> While most of the talk here at the show is about 40/100G, there's also a fair amount going on regarding gridless ROADM technology.&nbsp;Carriers' desire to be ready for what comes after 100G -- and the likelihood that it&nbsp;won't fit neatly on the ITU grid. The problem, of course, is that no one is really sure what comes after 100G (although there seems to be a consensus&nbsp;building around&nbsp;the idea of&nbsp;aiming for 400 Gbps before tackling 1&nbsp;Tbps) and what its spectral requirements will be. But&nbsp;such uncertainty has not stopped carriers from assuming that their suppliers will figure something out -- and they're right:</p> <ul> <li><a href="/equipment-design/news/Finisar-demonstrates-ROADM-transmission-advances-at-ECOC-103495149.html">Finisar is showing off its FlexGrid technology</a>, which&nbsp;offers 12.5-GHz granularity from 25 GHz to 200 GHz. The demonstration&nbsp;uses the company's WaveShaper platform, but&nbsp;of course would eventually be applied to wavelength-selective switches (WSSs).&nbsp;The software-based application can change both the spectral bandwidth and the center channel. Finisar's Rafik Ward says they'll see the technology to their system vendor customers to enable them to provide the capability as a feature of their equipment, rather than to carriers directly,</li> <li>Oclaro, meanwhile, is touting <a href="/equipment-design/news/Oclaro-delivers-dynamic-spectrum-capabilities-on-2x1-wavelength-selective-switch-102984669.html">dynamic spectrum capabilities for its 2x1 WSS</a>. The new feature supports granularity of 25 GHz, which Oclaro's Scott Parker says should prove adequate to the occasion. Oclaro should have the enhanced MEMS-based WSS generally available by the second quarter of next year. Application to WSSs with higher port counts will follow shortly thereafter. Parker and Krishna Bala say that the demand for this capability is not universal; some carriers see a tradeoff between such flexibility and high port counts, which they also want. So it would not appear that Oclaro will make this capability a standard feature on its WSSs, at least not initially.</li> <li>So when do carriers really want this capability? Between sources at Finisar and Oclaro, plus other companies in the ROADM space or who supply to it, timelines range from &quot;next year&quot; to &quot;2012 to 2013 -- far enough away for something to happen that might provide direction, but near enough that vendors have to work on it.&quot;</li> </ul> <p>&nbsp;<a href="/blog/ECOC-Reporters-Notebook-Day-1.html">Read the Day 1 Notebook</a> as well as <a href="/blog/A-final-look-at-ECOC.html">my final look at ECOC</a>.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/09/ecoc-reporters-notebook-day-2.html2010-09-22T09:25:00.000ZTransceiver survey returnsnoemail@noemail.orgStephen Hardy<p>Those of you who read Lightwave may recall our annual transceiver purchasing surveys. Once a year, we'd poll our readers who either buy transceivers for use in their networks (in the case of service providers and enterprise network managers) or in their equipment designs to ask them what they're buying and why.</p> <p>You can find an example of the results by reading <a href="/about-us/lightwave-issue-archives/issue/readers-rate-transceiver-transponder-vendors-54889922.html">&quot;Readers rate transceiver/transponder vendors&quot;</a> from one of our past issues.</p> <p>As you'll see, the insights we gained from these surveys were sometimes surprising and always interesting.</p> <p>So, by popular demand, we're doing it again. We've sent questionnaires to our e-zine circulation database -- but, of course, our subscription list no longer entirely encompasses our audience. So I'm blogging about this for two reasons:</p> <ol> <li>For our subscribers who have received the questionnaire but have not yet completed it, now is the time -- we're wrapping up the survey this week.</li> <li>For those of you who read us only via this website, I'd like to invite you to participate as well. If you buy transceivers for your network or design equipment that uses transceivers, we'd like to hear from you. (Notice I&nbsp;didn't include &quot;If you make or design transceivers.&quot; This survey isn't for you if you're a transceiver supplier.) If you'd like to participate, email me at <a href=""></a>, tell me how you or your company uses transceivers, and I'll send you a link to the survey.</li> </ol> <p>But time is of the essence -- as I said, we're about to close the survey. (We're taking advantage of the fact that I'm running around ECOC and not in position to write up the survey results to gather a few more responses.) So take advantage of the opportunity to tell us -- and your suppliers -- what's important to you when it comes to selecting and using transceivers/transponders.</p> <p>I promise you'll find the results fascinating.&nbsp;</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/09/transceiver-survey-returns.html2010-09-21T06:50:00.000ZECOC Reporter's Notebook: Day 1noemail@noemail.orgStephen Hardy<p><strong>12:30 PM:</strong> Greetings from the press room in Turin, where ECOC 2010 is off to a somewhat sedate start. Not surprisingly, much of what's on display pertains to 40 and 100 Gbps. For example:</p> <ul> <li>Among other things, GigOptix is showing off modulator drivers for 100G, in particular the quad GX62455 and the dual-channel GX62255 for DP-QPSK applications. In addition to meeting requirements for smaller size, combining drivers in multi-channel packages became necessary when single-channel devices exhibited too much cross talk problems, says Padraig O'Mathuna. These single-channel units likely will be targeted now at 40G applications, he says. Speaking of 40G, O'Mathuna says the company still sees strong demand for DQPSK components. -- but that the demand for 100G technology is even stronger.</li> <li>Opnext President and CEO Gilles Bouchard, meanwhile, says that his company's 40G business is doing just fine, thanks for asking. While the market &quot;went dry&quot; during the beginning of the economic downturn last year, Bouchard says he is seeing strong module demand for&nbsp;both DPSK (particularly for deployment in Brazil and China) and DQPSK (for China, Japan, submarine, and even some European applications). The company has 40G coherent on the roadmap, although its main coherent focus is on 100G. Despite the fact that Tier 1 suppliers such as Alcatel-Lucent, Ciena, and others have developed the first generation of 100G in-house, Bouchard believes his company will be able to address at least half of the total market for 100G when its offering becomes available next year. Bouchard and David Sykes (ex-StrataLight), meanwhile, believe that even the current&nbsp;100G do-it-yourselfers will look at modules for second-generation systems, particularly those targeted at metro and regional applications.</li> <li>Cube Optics&nbsp;is nearing beta sampling of a 40G&nbsp;LR4 TOSA/ROSA module.&nbsp;The company also is readying a 100G ROSA, which it hopes to have in production by the end of the year.</li> </ul> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/09/ecoc-reporters-notebook-day-1.html2010-09-20T15:30:00.000ZLightwave ONE: A service provider conference?noemail@noemail.orgStephen Hardy<p>Well, actually, yes: The Lightwave's Optical Networks for Enterprises (ONE) Conference is an event for service providers.</p> <p>It's not just for service providers, of course -- there will be plenty of of sessions that will appeal to enterprise managers. We'll offer presentations on cabling the enterprise, choosing the right equipment, and developing an optical networking strategy that will keep the average network manager on his/her boss's good side.</p> <p>But enterprise managers frequently can't meet their requirements alone. That's where service providers come in, with managed offerings, dark fiber, low-latency connections, and other products -- all of which require an understanding of the optical networking options as well. So the conference also will have sessions that will meet the information needs of service providers as well. For example:<span><font size="2"><font size="3"><br /> </font></font></span></p> <ul> <li>The Role of Optical Networks in Cloud Computing</li> <li>Low Latency Network Strategies</li> <li>Powering Desk to Desk Dynamic Ethernet Lightpaths</li> <li>Evolved Packet Transport and Enterprise Services</li> </ul> <p>...and more. You can <a href="" target="_blank">find out more about the conference program</a> on the Lightwave ONE website.</p> <p>Don't let the name fool you. Lightwave ONE&nbsp;will appeal to service providers as much as it will to enterprise network managers.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/09/lightwave-one-a-service-provider-conference.html2010-09-14T22:42:00.000ZAustralia NBN back on -- by one seatnoemail@noemail.orgStephen Hardy<p>Australia's National Broadband Network (NBN), placed in limbo when national elections failed to create a clear majority government, is back on after the incumbent Labor Party won the backing of enough Green and independent members of Parliament to stay in power.</p> <p>Labor and the independents who back the party now control 76 of the 150 seats in the Australian Parliament.</p> <p>NBN&nbsp;Co., the company created to manage the NBN's construction, naturally welcomed the news. &quot;The team at NBN Co has been working on business as usual in the post-election period, while limiting discretionary expenditure, extending the deadline for some tenders and putting the award of several tenders on hold,&quot; the company said in a press statement. &quot;We will now work to restore deferred processes, including the recruitment of staff.&quot;</p> <p>Alcatel-Lucent, <a href="/fttx/news/Alcatel-Lucent-to-supply-GPON-gear-for-Australias-National-Broadband-Network-97066139.html">which won a contact to supply GPON FTTH gear for the NBN</a>, is undoubtedly happy as well.</p> <p><a href=",nbn-co-back-in-business.aspx">According to ITNews</a> of Australia, at least two of the critical members of Parliament cited the NBN as a factor in backing Labor. A deal between Labor and the independents covering uniform national wholesale prices for the NBN and priority infrastructure rollout in regional areas was crucial to gaining the independents' support, ITNews says.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/09/australia-nbn-back-on----by-one-seat.html2010-09-07T14:05:00.000ZAn update on Lightwave ONE speakersnoemail@noemail.orgStephen Hardy<p>Yeah, we've got a lot of big names from the vendor community on the speaker roster for Lightwave's Optical Networks for Enterprises Conference (Lightwave ONE). They include Alcatel-Lucent, Ciena, Cisco, Ericsson, JDSU, etc., etc. (<a href="" target="_blank">see the full roster</a> on the Lightwave ONE website). But lest you think that vendors are the only people I&nbsp;know, I will point out that:</p> <ol> <li>NTT America has just signed up to discuss how to build <a href="/networking/news/Timing-everything-when-selling-low-latency-network-services-94110129.html">low-latency networks</a></li> <li>Yahoo and NoaNet (<a href="" target="_blank">Northwest Open Access Network</a>) will join VMware and ADVA Optical Networking to discuss data center interconnection.</li> </ol> <p>You can find out more about Lightwave ONE -- which will run October 4-6, 2010, co-located with <a href="" target="_blank">ITEXPO</a> in Los Angeles -- a<a href="" target="_blank">t the conference website</a>. And here's a quick link to registration information.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/08/an-update-on-lightwave-one-speakers.html2010-08-31T14:42:00.000ZAustralia's NBN on holdnoemail@noemail.orgStephen Hardy<p><a href="">The Sydney Morning Herald</a> and Dow Jones Newswire reports that NBN Co., the organization administering the Australian Government's National Broadband Network, has put the project on hold after national elections in Australia failed to produce a clear majority in Parliament.</p> <p>The incumbent Labor Government had launched the NBN at a projected cost of AUS$43 billion. However, the opposition Conservative Party offered a plan of its own, reduced both in scale and cost ($6.3 billion).</p> <p>According to the two news organizations, NBN Co. issued a statement saying that it would not award any &quot;significant&quot; new contracts or issue &quot;significant&quot; new tenders. Existing operations will continue.</p> <p>A copy of the announcement had not yet been posted on <a href="" target="_blank">the NBN site</a> at the time of this writing.</p> <p>Alcatel-Lucent recently <a href="/fttx/news/Alcatel-Lucent-to-supply-GPON-gear-for-Australias-National-Broadband-Network-97066139.html">won a contract from NBN Co.</a> worth at least AUS$70 million for <a href="/featured-articles/is-gpon-the-access-technology-for-the-next-decade-54886752.html">GPON</a> gear. Needless to say, the total value of that deal now becomes uncertain.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/08/australias-nbn-on-hold.html2010-08-24T14:09:00.000Z40GBase-T, anyone?noemail@noemail.orgStephen Hardy<p>So, you didn't think the copper cabling industry was just going to capitulate in the face of 40- and 100-Gigabit Ethernet, did you?</p> <p>Recognizing that a path towards tomorrow's speeds is essential to selling product today, copper cabling makers are laying plans to propose work on 40GBase-T. The current fly in the ointment is that tranmitting 40 Gbps an appreciable distance over copper would require use of Category 7a twisted pair -- a cabling category not yet recognized by U.S. standards bodies. Sources tell me that this initial step will probably delay the 40GBase-T drum beats until next year.</p> <p>The hope is that 40GBase-T will support 40-Gigabit Ethernet transmission over 100 m. The advent of such a specification (once products based on it reach affordability, anyway) would certainly reduce demand for fiber-based 40-Gigabit Ethernet products. However, not all the news would be bad for the fiber industry, say the sources -- aggregating 40GBase-T traffic would require fiber plant.</p> <p>And, no, it doesn't appear that 100GBase-T would be just around the corner.</p> <p>If you're interested in the role of 40 Gigabit Ethernet in optical networks, I have two pieces of advice for you:</p> <ol> <li>Read the article on this subject by Meghan Fuller Hanna that will appear in our upcoming September/October issue.</li> <li>Attend the Lightwave Optical Networks for Ethernet Conference this October 4-6 (<a href="" target="_blank">find more info here</a>).</li> </ol>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/08/40gbase-t-anyone.html2010-08-12T14:25:00.000ZFiber’s broadband stimulus winning streak snappednoemail@noemail.orgStephen Hardy<p>AUGUST 5, 2010 By Stephen Hardy -- Agriculture Secretary Tom Vilsack yesterday announced the funding of 126 new infrastructure projects that will be funded via the second round of the Broadband Incentives Program (BIP). Unlike previous funding rounds, fiber-based projects did not dominate.<br /> <br /> Which is not to say that fiber did poorly. By my count, just under 50 projects involve fiber in some way. (You can count yourself by viewing <a href="" target="_blank">the full list of new projects</a>.) But wireless, xDSL, and HFC networks finally received more than a handful of awards.<br /> <br /> The projects announced yesterday will receive $1.2 billion in stimulus funding, to be matched by $117 million in private investment. <br /> <br /> The USDA has issued $2.65 billion in loans and grants to construct 231 broadband projects in 45 states and one territory so far in the two funding rounds. It still has an additional $1 billion in loans and grants it is mandated to release by September 30, 2010. <br /> <br /> <br /> &nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/08/fibers-broadband-stimulus-winning-streak-snapped.html2010-08-05T14:37:00.000ZOclaro execs explain Mintera dealnoemail@noemail.orgStephen Hardy<p>Late yesterday afternoon, I had a chance to talk with two executives from Oclaro -- CFO Jerry Turin and Executive VP Sales/Marcom Scott Parker -- about the company's acquisition of Mintera (see <a href="/business/news/Oclaro-acquires-Mintera-with-40-100-Gbps-in-mind-98954969.html">&quot;Oclaro acquires Mintera with 40, 100 Gbps in mind&quot;</a>). Here are some highlights:</p> <ul> <li>The acqusition's boost to Oclaro's 100G efforts was at least as important, and perhaps more important, than its effect on the company's 40G portfolio, Turin said. It probably will have more of an impact when the 100G market reaches the module stage, however. The execs repeated Oclaro President and CEO Alain Couder's assertion that the initial play in the 100G space is at the component level (see <a href="/business/news/Oclaro-ClariPhy-deal-boosts-100G-component-and-module-play-95017574.html">&quot;Oclaro: ClariPhy deal boosts 100G component and module play&quot;</a>).</li> <li>Parker conceded that the 40G space has been soft for a few quarters, after a certain &quot;major deployer&quot; (that would be AT&amp;T) slowed its rollouts. However, he said that demand for 40G has started to pick up again, particularly for <a href="/equipment-design/news/Oclaro-announces-qualified-40G-RZ-DQPSK-transponder-88247002.html">DQPSK</a> technology.</li> <li>Parker foresees DQPSK&nbsp;ruling metro applications and <a href="/equipment-design/transport/featured-articles/Engineering-DPSK-spectral-properties-enables-superior-performance-through-multiple-cascaded-optical-wavelength-selective-switches-68645527.html">DPSK</a> and coherent 40G duking it out in long-haul applications. Of the three modulation formats, the market opportunity for 40G coherent technology will be most affected by the advent of 100G coherent. Both gentlemen foresee a healthy market for 40G in general over the next two years at least (after which 100G is expected to begin rolling out in earnest). Support for these deployments should make 40G a continuing source of revenue after cost-effective 100G arrives as well.</li> <li>Terry Unter's tenure with Oclaro -- beyond leading the new 40G/100G modules and subsystems division &quot;for a transitional period&quot; -- remains up in the air because it was decided not to hold up the acquisition until that could be defined, the executives said. There isn't a clock on the transitional period, they added.</li> </ul>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/07/oclaro-execs-explain-mintera-deal.html2010-07-22T13:36:00.000ZLightCounting's Kozlov on consolidationnoemail@noemail.orgStephen Hardy<p>Readers of Lightwave are aware that I interviewed Dr. Vladimir Kozlov, head of transceiver market research and analytics firm LightCounting LCC, for our July/August issue. (Those of you who have not yet subscribed to Lightwave <a href=";mode=2" target="_blank">can read the interview here</a>. Then you can <a target="_blank" href="">correct your serious lapse in judgment</a>.) One discussion topic I didn't have room to include in the article was Kozlov's thoughts on industry consolidation.</p> <p>In the interview, Kozlov noted that the module space has seen gradual consolidation. &quot;We&rsquo;re getting close to the situation where we will have four or five giants dominating the industry,&quot; he said. &quot;But there will certainly be plenty of room for new companies or specialists.&quot;</p> <p>Specialists and newcomers can thrive by addressing niches too small to interest the major players or by being nimble enough to serve short-term but lucrative opportunities, Kozlov believes.</p> <p>The trick is to define which end of the spectrum is home. &quot;You can run a very good business by being a specialist. And you can run a very good business by being a generalist,&quot; he said. &quot;The danger is getting stuck in between, where you&rsquo;re not quite diversified enough to be a good generalist and you&rsquo;re not quite specialized enough to be a specialist.&quot;</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/07/lightcountings-kozlov-on-consolidation.html2010-07-16T15:34:00.000ZFTTH quandary for Tier 2/3 carriersnoemail@noemail.orgStephen Hardy<p>The broadband stimulus program has certainly stimulated the bottom line of <a href="/fttx/ftth-b/">FTTH</a> gear supplier Calix. The company recently announced that it had topped the $100 million mark in terms of broadband stimulus programs in which it's involved (see <a href="/fttx/news/Calix-touts-100M-in-broadband-stimulus-customers-with-XIT-Rural-Telephone-Cooperative-contract-98022194.html">&quot;Calix touts $100M in broadband stimulus customers with XIT Rural Telephone Cooperative contract&quot;</a>). But that hasn't made stimulus-enabled projects its entire focus. Geoff Burke, senior director of corporate marketing at Calix, says the size of his company's opportunities with Tier 2 and 3 carriers who haven't received stimulus funds depends on how decision makers answer three questions.</p> <p>With AT&amp;T and Verizon shifting into a lower gear when it comes to FTTH/<a href="/fttx/fttn-c/">FTTN</a> deployments, the Tier 2 and 3 market becomes the main opportunity for vendors in the space. So far, these smaller players have dominated the broadband stimulus awards. But with the stimulus funding scheduled to be apportioned by this September, the question becomes what happens to the market next. Therefore, the behavior of carriers who either failed to acquire stimulus money or chose not to pursue it will have a significant effect on the demand for FTTH technology over the next several years.</p> <p>Burke says that executives at such Tier 2 and 3 players are weighing three main issues:</p> <ol> <li>Will government subsidies ever get any higher than they are now?</li> <li>Will low-interest debt become any more readily available than it is now?</li> <li>Will their competitive position ever be better than it is now?</li> </ol> <p>Burke says that some executives don't believe they have enough visibility to answer these questions, and are therefore delaying buildouts -- or making building decisions at all. However, he also indicated that several projects are moving forward. While he declined to put any numbers on the size of the opportunity or how successful Calix has been in taking advantage of it, my take was that a significant amount of FTTH activity is taking place outside of the stimulus halo. </p> <p>&nbsp;</p> <p>&nbsp;</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/07/ftth-quandary-for-tier-23-carriers.html2010-07-12T14:01:00.000ZUNH IOL looking to start 40GbE/100GbE test consortiumnoemail@noemail.orgStephen Hardy<p>Jeff Lapak, senior engineer attached to the 10 Gigabit Ethernet Consortium at the University of New Hampshire's Interoperability Lab (UNH-IOL), told me yesterday that the lab is preparing to offer test services for <a href="/education/news/IEEE-launches-next-generation-of-high-rate-Ethernet-with-new-IEEE-8023ba-standard--96862989.html">IEEE 802.3ba</a> 40 Gigabit Ethernet and 100 Gigabit Ethernet.</p> <p>UNH-IOL is probably the best-known communications interoperability test facility in the United States. It offers private test services and has hosted a variety of plug-fests for standards bodies and industry groups such as the Ethernet Alliance and the Broadband Forum (for example, see <a href="/education/news/FSAN-Broadband-Forum-to-collaborate-on-GPON-interoperability-91751644.html">&quot;FSAN, Broadband Forum to collaborate on GPON interoperability&quot;</a>). Many of UNH-IOL's test activities revolve around consortia -- groups of vendors who fund test activities associated with a particular technology or application.</p> <p>And the welcome mat has been placed in front of the lab for companies interested in being part of a 40 Gigabit Ethernet and 100 Gigabit Ethernet test consortium. Preliminary test suites are nearly complete, Jeff says, and discussions about an interoperability event have already begun. If you want more information or are interested in participating in the consortium, you can email Jeff at <a href="javascript:location.href='mailto:'+String.fromCharCode(106,114,108,97,112,97,107,64,105,111,108,46,117,110,104,46,101,100,117)+'?subject=40%2F100%20Gigabit%20Ethernet%20Consortium'"></a>.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/07/unh-iol-looking-to-start-40gbe100gbe-test-consortium.html2010-07-08T14:12:00.000ZIs broadband a basic communications service?noemail@noemail.orgStephen Hardy<p>FICORA, the communications regulatory agency of Finland, thinks so. As of today, consumers and businesses &quot;are entitled to a 1-Mbps broadband subscription&quot; as part of the country's definition of universal service, the agency announced.</p> <p>The imperative is now part of the Communications Market Act. FICORA also has defined 26 telecommunications operators as unversal service providers, each of whom FICORA says &quot;have an obligation&quot; to provide the 1-Mbps services at a price and delivery time the agency determines is &quot;reasonable.&quot; While what might be reasonable as far as delivery time would vary from case to case, FICORA says, the price ought to be between 30 and 40 euros a month.</p> <p>The move underscores the increasing importance of broadband service in the eyes of regulators, not only in Scandinavia but around the world. I would expect the decree isn't the last of its kind we'll see. While FICORA says the choice of communications medium is up to the individual operator, one could expect such directives to be beneficial to the optical communications industry.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/07/is-broadband-a-basic-communications-service.html2010-07-01T14:06:00.000ZFiber rules first round of RUS BIPnoemail@noemail.orgStephen Hardy<p>The Department of Agriculture's Rural Utilities Service (RUS) has confirmed the obvious: fiber projects dominated the first round of Broadband Initiatives Program (BIP) awards.</p> <p>In a report entitled &quot;Connecting Rural America,&quot; the RUS says it awarded $1.068 billion to 68 recipients in 31 states and one U.S. territory through BIP. The grants included:</p> <p>13 last-mile remote projects that received $161 million</p> <p>49 last-mile non-remote projects that received $739 million</p> <p>6 middle-mile project that garnered $167 million.</p> <p>As I've noted previously, projects that leverage fiber earned the most attention within BIP -- 48, to be precise, the report says. Fixed terrestrial wireless came in second with 23 projects, followed by xDSL (14), terrestrial wireless mobile (5), hybrid wireless (3), coax (2) and satellite (1).</p> <p>For-profit entities received the most awards (37), but cooperative or &quot;mutual&quot; organizations received the most money (approximately $500 million).</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/06/fiber-rules-first-round-of-rus-bip.html2010-06-21T14:23:00.000ZStimulus Round 2: DSL's revenge?noemail@noemail.orgStephen Hardy<p>Round 1 of the broadband stimulus awards confirmed fiber as the most direct path to 100 Mbps to residences and 1 Gbps to anchor institutions in the United States. However, at least one observer expects DSL technology to edge into the spotlight in Round 2.</p> <p>Occam Networks' Director, Solutions Marketing and Strategy Juan Vela, in discussing his company's role as supplier to BIP winner Rural Telephone (see <a href="/fttx/news/Rural-TelephoneNex-Tech-selects-Occam-Networks-for-101M-broadband-stimulus-project--95858299.html">&quot;Rural Telephone/Nex-Tech selects Occam Networks for $101M broadband stimulus project&quot;</a>), believes there will be pressure to grant awards to Qwest/CenturyLink or other high-profile applicants who plan to leverage DSL technology (although Qwest's application abstract also mentions fiber). Vela says that these carriers will offer channel bonding (perhaps such as <a href="/fttx/news/Alcatel-Lucent-Bell-Labs-phantom-mode-pumps-300-Mbps-over-two-DSL-lines-91701684.html">the 300-Mbps technology Alcatel-Lucent has touted?</a>), at least in more urban areas, as a pathway toward the 100-Mbps to the home 10-year goal the FCC has set in its National Broadband Plan.</p> <p>Meanwhile, Vela says that smaller carriers, municipalities, utilities, etc., will continue to focus on fiber in their Round 2 applications.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/06/stimulus-round-2-vdsls-revenge.html2010-06-08T14:20:00.000ZAnalysts offer Cisco/CoreOptics perspectivesnoemail@noemail.orgStephen Hardy<p>Having finally succeeded in getting the June issue out the door (we&rsquo;ve redesigned the mag -- I think you&rsquo;re going to like it), I finally have a chance to catch up on a few things. This includes more perspective on <a href="/business/news/Cisco-to-acquire-CoreOptics-94447114.html">Cisco&rsquo;s plans to purchase CoreOptics</a>.<br /> <br /> The best source of such info, of course, is Cisco itself. In the email <a href="/blog/Cisco-Well-suppot-CoreOptics-customers.html">I mentioned in my previous blog</a>, Cisco PR Manager, Corporate Communications Robyn Jenkins-Blum wrote that &ldquo;CoreOptics&rsquo;s technology could be used on any Cisco product with 10/40/100G interfaces. Cisco is also investigating other potential applications of the technology.&rdquo;<br /> <br /> To some analysts, those &ldquo;other potential applications&rdquo; include <a href="/about-us/lightwave-issue-archives/issue/packet-optical-transport-systems-the-new-pots-54894827.html">packet optical transport</a> platforms. When I asked her via email whether she agreed with this sentiment, Ovum optical networking VP and practice leader Dana Cooperson stopped short of leaping to the same conclusion with both feet. &ldquo;It certainly signals more understanding and awareness on Cisco&rsquo;s side of the importance of optical expertise and optical performance as packet and optical increasingly converge. (Specifically, they are signaling the importance of the ability to control costs, supply continuity, and differentiation and the willingness to invest.),&rdquo; she responded.<br /> <br /> Similarly Karen Liu, VP, components at Ovum, wrote in a note published shortly after Cisco&rsquo;s announcement, &ldquo;Cisco&rsquo;s CoreOptics acquisition -- coupled with other <a href="/networking">optical network</a> business unit roadmap plans discussed at its customer conference in Dallas&hellip; indicates a renewed understanding of just how important the optical part of &lsquo;converged packet optical&rsquo; will be to tomorrow&rsquo;s network performance. When Cisco is on record as embracing OTN (Optical Transport Network) you know things are getting interesting.&rdquo;<br /> <br /> Liu sees a strategic shift at 100G among systems houses away from outside technology suppliers. &ldquo;Cisco is one of the proponents of MSA modules for optics; it is widely credited (blamed?) for commoditizing the datacom optical modules business. For it -- even more for traditional telecom OEMs -- to buy an optical modules company outright highlights the new need for OEMs to control supply of not only the immediate parts but the technology going forward,&rdquo; she wrote. &ldquo;Formerly, there was a view that cost reduction requires highly competitive supply for the optics. At 100G and beyond, differentiated performance in terms of reach becomes the key enabler to lower cost.&rdquo;<br /> <br /> This would spell bad news for module vendors hoping to supply such OEMs&rsquo; 100G needs. However, Dr. Vladimir G. Kozlov, LightCounting&rsquo;s founder and chief analyst, has a rosier view. &ldquo;Cisco's move is certainly good news for component and transceiver vendors that invested heavily in the development of 40- and 100-Gbps products,&rdquo; he wrote in a note released today. &ldquo;In light of Cisco's clear commitment to delivering this technology to end users, other system vendors will have to follow suit, boosting demand for these higher margin products.&rdquo;<br /> <br /> <br /> <br /> &nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/06/analysts-offer-ciscocoreoptics-perspectives.html2010-06-01T20:46:00.000ZCisco: We'll support CoreOptics' customersnoemail@noemail.orgStephen Hardy<p>Well, here's something of a surprise. Contrary to expectations, it appears that Cisco plans to continue to support CoreOptics' existing 10G and 40G customers once the acquisition has completed.</p> <p>Cisco cited its expectation that CoreOptics can help the company add 100-Gbps capabilities to its platforms <a href="/business/news/Cisco-to-acquire-CoreOptics-94447114.html">in announcing the planned acquisition</a>. However, CoreOptics initially has made a name for itself by providing extended-reach 10-Gbps modules based on its version of Maximum Likelihood Sequence Estimation (MLSE) electronic dispersion compensation to systems vendors such as Siemens (now Nokia Siemens Networks). Most recently, <a href="/equipment-design/news/CoreOptics-announces-40G-coherent-MSA-transponder-module-88889677.html">it announced a 40-Gbps module</a> based on polarization-multiplexed <a href="/about-us/lightwave-issue-archives/issue/is-dpndashqpsk-the-endgame-for-100-gbitssec-54890687.html">QPSK</a> with coherent detection, with NSN and Cisco as announced customers and Fujitsu Network Communications and Ericsson as suspected customers.</p> <p>Cisco, obviously, should see no interruption in its supply of 40G modules. Other CoreOptics customers were assumed to be not quite as fortunate.</p> <p>However, Cisco spokeswoman Robyn Jenkins Blum emailed Lightwave in response to queries about Cisco's plans that &quot;Cisco is committed to ensuring the continued supply and availability of CoreOptics&rsquo; current 10G and 40G products for existing customers. Cisco and CoreOptics will continue to operate as independent entities until the transaction closes.&quot; Asked if that commitment extended only for the period during which the companies would remain separate, Blum repeated, &quot;Cisco is committed to ensuring the continued supply and availability of CoreOptics&rsquo; current 10G and 40G products for existing customers.&quot;</p> <p><br /> <br /> &nbsp;</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/05/cisco-well-suppot-coreoptics-customers.html2010-05-21T02:44:00.000ZHFC isn’t a fiber networknoemail@noemail.orgStephen Hardy<p>When is a fiber-optic <a href="/networking">network</a> not a fiber-optic network? For advertising purposes, when it&rsquo;s a hybrid fiber/coax (HFC) network.<br /> <br /> So ruled the Better Business Bureau Council's National Advertising Division (NAD), as the result of complaints Verizon lodged against Time Warner Cable and Cox Communications in the ongoing war for the hearts, minds, and wallets of residential broadband subscribers.<br /> <br /> Verizon, of course, has positioned its <a href="/fttx/ftth-b/">FTTH</a>-based FiOS service as providing a superior alternative to the HFC-based networks <a href="/mso-optics/">cable MSOs</a> use most frequently. In response, some MSOs -- Time Warner Cable and Cox among them &ndash; have positioned their services as being delivered via &ldquo;fiber-optic networks&rdquo; as well.<br /> <br /> Seeking to maintain its competitive differentiation, Verizon complained to the BBB that such advertising gives consumers the impression that the MSO networks use fiber all the way to the home in the same way Verizon&rsquo;s does. The NAD agreed, and recommended to both Time Warner Cable and Cox that they change their messages.<br /> <br /> Time Warner Cable, at least, isn&rsquo;t willing to roll over. The company plans to appeal the division&rsquo;s ruling to the National Advertising Review Board.<br /> <br /> &ldquo;Despite the undisputed fact that TWC&rsquo;s network is more than 90% fiber optic, NAD&rsquo;s decision seeks to stop TWC from making the true statement that it has a fiber optic network,&rdquo; the company said, apparently believing that more than 90% is close enough to 100% not to be worth quibbling about.<br /> <br /> Cox, on the other hand, while stating that it still believes it is correct to tell consumers that it operates a fiber-optic network, has agreed to take the NAD&rsquo;s recommendations into consideration in its marketing going forward.<br /> <br /> Comcast, meanwhile, refused last November to participate in the NAD&rsquo;s adjudication process of its similar claims. The NAD referred the issue to the Federal Trade Commission.<br /> <br /> All of which illustrates that the benefits of fiber technology must be resonating strongly with consumers -- whether you&rsquo;re bringing it to the side of their home or not. <br /> <br /> &nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/05/hfc-isnt-a-fiber-network.html2010-05-06T21:23:00.000ZOIDA's Lebby becomes Translucentnoemail@noemail.orgStephen Hardy<p>Michael Lebby -- recently ex-president and CEO of the Optoelectronics Industry Development Association (OIDA), member of the Lightwave Editorial Advisory Board, and eyewear savant -- reports that he is joining Translucent Inc. in Palo Alto as general manager and chief technology officer. The company originally spun out of Stanford to make erbium-doped waveguide amplifiers using erbium-doped silicon waveguides about a decade ago but has since changed its focus. Backed by Australian-based investor Silex, the company now hopes to supply what Lebby calls &quot;virtual substrates&quot; based on rare earth oxides.</p> <p>&quot;Work is progressing on SOI technology as well as Ge-on-silicon technology for solar and very high speed detectors,&quot; he writes in an email.&nbsp; &quot;Translucent has published in the past light emitting silicon using rare earth oxides.&quot;&nbsp;</p> <p>The virtual substrates would be SOI, SiGe, or Ge epi-ready for fabrication, Lebby adds.</p> <p>The company currently has a landing page up at <a target="_blank" href=""></a>.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/04/oidas-lebby-becomes-translucent.html2010-04-27T02:33:00.000ZFTTH Council numbers show continued growthnoemail@noemail.orgStephen Hardy<p>When the FTTH Council <a href="/education/news/FTTH-Council-survey-shows-widespread-plans-for-FTTH-networks-90830044.html">announced earlier this week</a> its semi-annual review of North American <a href="/fttx/ftth-b/">FTTH</a> penetration figures (garned with the help of RVA LLC), the council highlighted the fact that 1.4 million homes within the continent now enjoy the benefits of optical access technology.</p> <p>But the survey also provides a deeper look at the current state of FTTH deployment in North America. For example, the survey also contains the following data points:</p> <ul> <li>As of March 30, 2010, there are approximately 18.2 million homes-passed in North America. About 99% of this activity has been in the United States to date. This compares to 17.2 million in September 2009.</li> <li>Homes marketed stand at 16.992 million, up from 16.048 million six month ago.</li> <li>Homes connected was slightly more than 5.8 million, compared to just under 5.3 million last September.</li> <li>FTTH has now reached nearly 16% penetration of U.S. households in terms of homes passed and 5% in terms of homes connected.</li> <li>Overall take rates hit 34.1%, up from 32.5% in September 2009. Take rates for carriers other than Verizon were just over 52%. While this is impressive on the face of it, this figure represents a modest decline from the 53.1% figure of last fall. Verion&rsquo;s take rate was estimated to be 29.5%, up from 28% six months previously.</li> </ul> <p><br /> All of this illustrates the growing the progress FTTH continues to make. But there's still a lot of work to do before subscribers in North America enjoy the level of service Asian carriers supply to their customers. As if to underscore this fact, Hong Kong Broadband Network Ltd. announced April 13 the immediate availability of a symmetrical 1-Gbps service for approximately $26 a month.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/04/ftth-council-numbers-show-continued-growth.html2010-04-15T13:32:00.000ZOFC/NFOEC Reporter's Notebook, Day 4noemail@noemail.orgStephen Hardy<p>Just a few quick notes today from the airport, thanks to wireless snafus at my hotel...</p> <p>There was no escape from 40G/100G at the show this year, so after a look at other areas yesterday, it's back to high-speed technology. With the arrival of 40G coherent technology on the market, the question of where the various modulation form factors will find success arises. As a supplier of receiver technology for DPSK, DQPSK, and DP-QPSK (100G now, 40G-specific in the future), u2t Photonics is in a unique position to comment Andreas Umbach at the u2t booth said he believes DPSK will triumph in the long haul, with coherent and DQPSK battling it out for shorter links. In addition to the company's receiver technology, Andreas was proud of new modulators based on GaAs. The technology enables small package sizes...</p> <p>GigOptix, meanwhile, showed off its new modulator technology that leverages polymer. Those package sizes are even smaller. The company also displayed technology that could be applied to whatever will come after 100G coherent...</p> <p>CoreOptics basked in the glow of the success of its coherent 40G modules, based on technology it calls CP-QPSK. Saeid Aramideh, between sips of champagne served in honor of his birthday Thursday, expressed confidence that coherent will dominated future 40G applications. In the CoreOptics booth, a CP-QPSK in a Nokia Siemens Networks platform was part of a working demo. (The recently announced <a href="/networking/news/Columbus-Networks-expands-40G-DWDM-network-adds-capacity-to-handle-demand-89092692.html">Columbus Networks 40G deployment</a> did not use this technology, NSN said.) A static demo showed the same module in a Cisco ONS 15454 MSTP M6 platform. The company has a Japanese customer Saied says he was not allowed to name. (Hmmm...did a Japanese company <a href="/networking/news/Fujitsu-adds-coherent-40G-to-FLASHWAVE-7500-ROADM-88390472.html">recently make a 40G coherent announcement</a>?)</p> <p>After the show, Clariphy, which has made a name for itself in 10G EDC, talked about its work with Mintera on the latter's upcoming 40G module. As noted, the trick with coherent is the high-speed receiver electronics, and ClariPhy and Mintera found the former's mixed signal and DSP expertise a good match for Mintera's algorithm IP and related skills. Mintera will have initial exclusive rights to the resultant chip (Mintera plans to have the 40G coherent product available by the end of this year, says company head Terry Unter), but ClariPhy honcho Paul Voois says the chip will be available on the open market at some point next year.</p> <p>Needless to say, there was a lot more going on (you have to test all this technology, for example), so you can expect more about the show next week. Meanwhile, it's time for a trip to Starbucks before my flight leaves.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/03/ofcnfoec-reporters-notebook-day-4.html2010-03-26T15:16:00.000ZOFC/NFOEC 2010 Reporter's Notebook, Day 3noemail@noemail.orgStephen Hardy<p>Despite appearances, 40G/100G is not the only area of innovation evident here in San Diego. So in honor of these potentially overshadowed developments, today's blog will be 40G/100G free. You're welcome.</p> <p>At a breakfast roundtable hosted by Nokia Siemens Networks to discuss the technology I just said I'm not mentioning here, Alan Sardella, senior product marketing manager, high end systems at Juniper Networks, reported that the Menara Networks <a href="/general/menara-networks-to-debut-otn-in-a-transceiver-at-ofcnfoec-54884837.html">&quot;OTN in an XFP&quot; product</a> for which the router vendor was the first announced customer is currently in lab trials. He described it as &quot;less flexible than LR4 optics&quot; but useful for metro and other medium-distance links &quot;not needing many wavelengths.&quot; The product has a niche, Sardells said...</p> <p>Companies interested in tunable XFPs will soon have two choices, as Finisar says it will step up to challenge JDSU. The company is showing of a device VP of Marketing Rafik Ward says will offer an extinction ratio of greater than 13 dB. Finisar has both PIN and APD versions planned and is looking at an APD and variable optical attenuator pairing as well. The devices should begin sampling by the end of June, with general availability in the fourth quarter of this year. Performance, power dissipation, and flexibility will be the key differentiating points, Ward predicted...</p> <p>That said, Tom Fawcett of JDSU indicated his company's tunable XFP, which is already in production, won't take a back seat to anyone's when it comes to performance. JDSU is displaying several other new developments in its booth as well. These include a module version of the optical building blocks that make up its <a href="/about-us/lightwave-issue-archives/issue/several-factors-shape-the-new-optics-landscape-54891217.html">AON Super Transport Blade</a>. The company discovered that some customers were interested in the building blocks but didn't want to partner with JDSU to come up with a blade. So JDSU now offers do-it-yourselfers an integrated package that would fit in a two-slot-wide design. Meanwhile, the company also announced a 1x23 wavelength-selective switch (WSS). The idea is not to support 23-degree ROADMs, but rather 4D colorless directionless add/drop applications in place of a pair of 1x9s in add/drop banks. The company isn't commenting yet on a release date...</p> <p>Bookham also announced a 1x23 WSS, thus leveraging the technology is acquired when it bought Xtellus. Krishna Bala, who is now exeutive vice president and division manager for WSS at Bookham, commented that the &quot;gridless&quot; <a href="/about-us/lightwave-issue-archives/issue/ensuring-profitability-with-a-3g-roadm-system-53428682.html">ROADMs</a> Verizon is said to be calling for (just in case whatever transmission technology is required to support the data rates that come after those I'm not discussing here must operate off the ITU grid) would be just a small part of the overall rethinking of the network that gridless capabilities would require -- which I take to mean he's not going to begin work on a gridless WSS tomorrow...</p> <p>One company that won't have a 1x23 WSS -- but now could afford to build one -- is Nistica. CEO Ashish Vengsarkar said he was very pleased to add NTT Electronics as an investor as part of the company's <a href="/business/news/Nistica-announces-65-million-Series-C-financing-88889092.html">recent Series C round</a>. He also noted the loss of Finisar, which had previously been an investor, but now has WSS technology of its own through its acquisition of Optium. Vengsarkar says he'll use the money to bring the company's edge ROADM subsystems to full production in Vietnam. The company also has a new 1x2, 25-GHz subystem, which Vengsarkar says is indicative of the company's direction, which doesn't include 1x23 and similar large-port-count offerings. He says Nistica has the technology to play in a gridless environment...</p> <p>Back at 10G, that data rate is new in the <a href="/fttx/pon-systems/">PON</a> world. OneChip Photonics is licking its chops at the chance to roll out 10G PON devices by the end of this year. The first will likely be 10G EPON, thanks to the fact that the relevant standards are already in place and Asian vendors are eager to take advantage of them. The company thinks that its PIC-based technology, which VP of Marketing Communications Steve Bauer and VP, Product Line Management Andy Weirich say that the 20-25% cost advantage their technology enables for current PON systems could grow to as high as 50% at 10G. The company has interest in producing its own version of an &quot;ONU in an SFP&quot; (keep an eye peeled for an article on such devices in the April issue of Lightwave) and is also looking at other markets for its technology. (<a href="/about-us/lightwave-current-issue/PIC-based-transceivers-enable-cost-effective-1G-to-10G-PON-migration.html">See the related article</a> in our March issue for more.)...</p> <p>More electronics to enable 10G EPON are starting to come on line as well. For example, Gennum has introduced the GN7350 single-chip transceiver, which includes both CDR and laser driver capabilities. The device is one of six new products on display in the company's meeting room on the show floor. Most of these focus on 10G, but it also has a 6G device primarily for wireless backhaul applications in China...</p> <p>Avago Technologies is touting an embedded <a href="/equipment-design/news/Avago-Technologies-collaborates-with-IBM-on-high-bandwidth-optical-interconnect-breakthrough-for-supercomputers-88880152.html">mid-board fiber-optic interconnect technolog</a>y it developed with IBM for use in the latter's supercomputers. The tiny 12x10G parallel optic modules, which leverage PRIZM connector technology developed by USConec, could ship in extremely high volumes, Avago expects.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/03/ofcnfoec-2010-reporters-notebook-day-3.html2010-03-25T01:38:00.000ZOFC/NFOEC 2010 Reporter's Notebook, Day 2noemail@noemail.orgStephen Hardy<p>It's a 40G/100G world, we just attend OFC/NFOEC in it....On the line side, Opnext demonstrated the 100G technology that <a href="/networking/news/ATT-Opnext-Cisco-demonstrate-127-Gbps-transmission-with-live-traffic-87115457.html">AT&amp;T used in its recent 900-km trial.</a> While I couldn't get closer than a barrier 6 feet from the demo because customers had arrived before me, the three racks of equipment were meant to simulate a 240-km submarine festoon network. Among the prime uses of the demonstrator, a source said, was to aid in risk reduction for the DSP algorithm and related coherent detection elements Opnext has developed and, in some instances, <a href="/equipment-design/products/Opnext-delivers-first-ultra-high-speed-SMT-multiplexer-IC-for-100G-coherent-transponders--82823242.html">publicized</a>. A 100G product -- which will be housed in an OIF-generated MSA module, and example of which was on display elsewhere in the booth -- will be released by the end of the year. Meanwhile, a 40G coherent transponder product will be &quot;not too far behind&quot; the 100G module. On the client side, the company's 100GBase-LR4 module is sampling to customers, which are expected to include makers of core routers, high-end switches, and optical transport platforms. At 40G, the company demonstrated interoperabiity between its 40GbE LR4 CFP module with a similar product from Sumitomo...</p> <p>Finisar has plans in the coherent world, but right now is focusing on the 40GbE and 100GbE market. The company showed of the FTLC1181 100GBase-LR4 CFP that has been used in a variety of demos (see <a href="/networking/news/Verizon-Juniper-NEC-Finisar-conduct-100G-field-trial-86874052.html">&quot;Verizon, Juniper, NEC, Finisar conduct 100G field trial&quot;</a>), including one with Ixia and EXFO test gear in Finisar's booth. While the IEEE specs call for 10-km reach, Finisar says its module will support at least 20 km. The module is sampling today, and Finisar expects production to follow by the end of this summer. Since potential customers were &quot;actively involved&quot; in the module's development, Finisar is confident the product will find use. A 40GbE product will follow as well...</p> <p>One company that doesn't have to wait for such products, at least on the line side, is Ciena, which has a trailer truck on the show floor that shows the interaction of its pre-Nortel MEN&nbsp;products with their new siblings (see <a href="/networking/news/Ciena-offers-Nortel-MEN-integration-insight-88793747.html">&quot;Ciena offers Nortel MEN integration insight&quot;</a>). The ability of the 4200 FlexSelect to carry OME 6500 generated 40G traffic as a alien wavelength were among the demonstrations. However, the really interesting demos were housed behind closed doors in the company's booth next door. Besides standard 40G and 100G transmission, Ciena showed off the ability of the adaptive optical engine technology it just acquired to support coherent-based colorless transmission at 40G. The capability will have particular impact on ROADMs, enabling a significant increase in WSS functional capacity,. The capability is expected to be announced as a product next year...</p> <p>Meanwhile, the folks at ADVA Optical Networking are still not sure how much demand there will be for coherent-based technology in the near term, particularly for carriers and other customers smaller than Tier 1. The company made a splash last year with its announcement that it would take a different direction than the OIF-sanctioned DP-QPSK with coherent detection for 100G. The company began to publicize a modulation format called <a href="/general/adva-demos-serial-100g-transmission-for-metroregional-applications-54893407.html">DPSK-3ASK</a>, which it asserted would be less expensive than coherent DP-QPSK and still support&nbsp; 600 km, which would be adequate reach for such applications. ADVA's Jim Theodoras revealed today that customer feedback has led ADVA to conclude that even DPSK-3ASK will not reduce costs enough. The company is now working on what it calls &quot;Enterprise 100 Gigabit Ethernet&quot; that will support 200-300 km and better meet customers' costs targets. The company isn't saying much about what technology this will entail, other than it will be a 4x28G approach that leverages 40GbE gearbox electronics. That said, the company isn't abandoning DPSK-3SK, the development of which is still be funded by European grants. But when and if the technology will reach the product stage is now unclear....</p> <p>Meanwhile, companies that will support 40G/100G transponder development or system deployment are basking in the reflective glow. For example, Raman amplification will take a leading position in such networks, and both Red-C Networks and Fitel are touting hybrid EDFA/Raman amplifiers on the show floor. Meanwhile Inphi and u2t Photonics are showing a demonstration of the latter's 100G coherent receiver that is enabled by the former's electronic devices.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/03/ofcnfoec-2010-reporters-notebook-day-2.html2010-03-24T02:10:00.000ZOFC/NFOEC 2010 Reporter's Notebook, Day 1noemail@noemail.orgStephen Hardy<p><strong>3:20 PM PST:</strong> In a break at the OSA/Lightwave Executive Forum. I'll have an article later on the 40G/100G components and subsystems panel I moderated this morning. Meanwhile, here are a few select snippets from other panels and speeches: Google's Vinton Cerf admitted that he didn't envision Twitter when he &quot;co-invented&quot; the Internet. He says he asked the head of Twitter whether his title was &quot;chief twit&quot;...Cerf said that while fiber makes the most sense for the majority of Google's <a href="/fttx/ftth-b/">FTTH</a> trials, the search engine giant is pondering the use of microwave at some edges of its networks, principally where laying fiber would be too expensive...Ovum's Dana Cooperson said that the WAN DWDM 40/100G CAGR between 2008-2014 will be approximately 79%. Meanwhile, the CAGR for 40GbE/100GbE transceivers will be 200% 2009-2015. There will be more than 2.5 million 40/100GbE ports shipped in 2015. Meanwhile, she thinks it will take 8-10 years to get to 100G to 6x cost for 10x the bandwidth compared to 10G...Bikash Koley, network architecture at Google, reports that there's no incentive to install 40GbE or 100GbE until the 20x power penalty versus 10GbE typical for bleeding edge technology approaches parity. He also said that, since Google already is running 800 Gbps per fiber, the choice of 100GbE versus 40GbE is meaningless when the standards only assume one interface per fiber. He'd like to see a rate adaptive version of 100GbE, similar to what's available in certain flavors of DSL (albeit at lower bit rates)...Hans-Juergen Schmidtke of Nokia Siemens Networks, says that 40G/100G optics costs will come down once &ldquo;overlooked manufacturing innovations (automation) become effective.&quot; So stop overlooking automation, you components and subsystems vendors! Meanwhile, he also offered that there's plenty of opportunity to innovate and differentiate at the component/subsystem level.</p> <p>Gotta go -- the next session is starting...</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/03/ofcnfoec-2010-reporters-notebook-day-1.html2010-03-22T22:22:00.000ZBye bye Supercomm?noemail@noemail.orgStephen Hardy<p>There won't be a Supercomm this year. The question now is when we'll see the next one.</p> <p>In a terse statement released this morning, Reed Expo's EXPOCOMM Events LLC, which had taken over management of the show last year and moved it from June to October, said that it would not renew its contract to produce Supercomm in 2010. </p> <p>&quot;The SUPERCOMM co-owning associations have also decided against producing a SUPERCOMM event in 2010,&quot; added the release.</p> <p>Those &quot;co-owning associations&quot; would be TIA and USTelecom, which partnered on the show for 18 years, <a href="/about-us/lightwave-issue-archives/issue/split-custody-53914372.html">split to do their own events</a> after the 2005 Supercomm, then <a href="/general/tia-and-ustelecom-combine-shows-again-53432817.html">reconciled in late 2006</a> to produce NXTcomm in 2007 when their individual shows failed to meet expectations. NXTcomm was <a href="/general/supercomm-returns-54889442.html">renamed Supercomm</a> for 2009 in an attempt to regain the old mojo. Apparently, it didn't work.</p> <p>We'll provide more details as they become available.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/02/bye-bye-supercomm.html2010-02-08T16:28:00.000ZYou think this is easy?noemail@noemail.orgStephen Hardy<p>Yeah? Then you should try it yourself. No, seriously -- I mean it. You should try blogging for Lightwave.<br /> <br /> Not that I&rsquo;m going anywhere -- I&rsquo;m not. But it occurs to me that, with Meghan still too busy with young Tripp (who&rsquo;s apparently starting to motor around upright and getting into all the expected kinds of trouble; thank you very much for asking) to write something here, we could probably use another voice or two on the site.<br /> <br /> So I&rsquo;m holding open auditions for bloggers. And, yes, that could mean you.<br /> <br /> How do you take advantage of this opportunity for potential fame? (Notice that I didn&rsquo;t add &ldquo;fortune.&rdquo; This is not a paying gig. Although certainly getting exposed to the industry by blogging on this site could convince other people to pay you to do things.) First, you have to get by me.<br /> <br /> <strong>Things you need to do to get your blog idea past me</strong>:</p> <ol> <li>Convince me you know what you&rsquo;re talking about.</li> <li>Talk about something germane to optical communications</li> <li>Don&rsquo;t use this space to try to sell something</li> <li>Know and practice the difference between &ldquo;opinionated&rdquo; and &ldquo;libelous&rdquo;</li> </ol> <p><br /> Of course, once you get by me, you also have to impress our audience. Generate comments, page views, and other evidence of a burgeoning following, and I&rsquo;ll let you blog away to your heart&rsquo;s content. If not, well&hellip;<br /> <br /> Interested? <a href="javascript:location.href='mailto:'+String.fromCharCode(115,116,101,112,104,101,110,104,64,112,101,110,110,119,101,108,108,46,99,111,109)+'?subject=I%20want%20to%20blog%20for%20Lightwave'">Drop me an email</a>. All this could be yours&hellip;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/01/you-think-this-is-easy.html2010-01-12T21:06:00.000ZVerizon's Wellbrock on router OC-768 port pricesnoemail@noemail.orgStephen Hardy<p>As mentioned <a href="/networking/news/Verizon-plans-more-100G-and-40G-deployments-80638462.html">in the story I posted yesterday</a>, Glenn Wellbrock, director of backbone network design at Verizon, said that the carrier will likely buy &ldquo;a lot&rdquo; of 4x10G muxponders to carry 10-Gigabit Ethernet streams from core routers at a more spectrally efficient 40 Gbps. This is because the router vendors are asking too much money for OC-768 interfaces.<br /> <br /> &ldquo;OC-768 has been way higher than 4x10-GigE LAN PHY. And that&rsquo;s what&rsquo;s kept it from being, at least in our network, highly used or used everywhere,&rdquo; he said. While praising transport vendors for getting 40G DWDM interfaces down to an attractive level, &ldquo;it&rsquo;s still not that way on the router side. They&rsquo;ve been charging a real premium, and that doesn&rsquo;t seem to be changing,&rdquo; Wellbrock said.<br /> <br /> That doesn&rsquo;t mean Verizon hasn&rsquo;t deployed any 40G router interfaces, however. &ldquo;We use some OC-768, where you really need that amount of capacity and you need it all at once,&rdquo; Wellbrock said. &ldquo;But in many cases we use the 10-GigE LAN PHY because the cost delta is so big.&rdquo;</p> <p>Meanwhile, he's hoping that 100G router ports, which he expects to see this year, will be in the neighborhood of 5X or 6X 10G LAN PHY prices -- but that 10X might be attractive enough to get deployment rolling.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2010/01/verizons-wellbrock-on-router-oc-768-port-prices.html2010-01-05T20:27:00.000ZNortel: OIF 100G components yes, modules probably notnoemail@noemail.orgStephen Hardy<p>In an interview last week in the wake of <a href="/networking/news/Nortel-unveils-commercially-available-100G-optical-system-79220847.html">Verizon's commercial deployment of Nortel's 100-Gbps technology</a>, Metro Ethernet Networks group VP of R&amp;D Dino DiPerna talked with me about what he expects out of the <a href="/mso-optics/transport/news/oif-issues-100g-framework-document-54895452.html">Optical Internetworking Forum's efforts to create a supplier eco-system</a> around dual-polarization quadrature phase-shift keying (DP-QPSK) with coherent detection.</p> <p>His views won't warm the hearts of module vendors.</p> <p>Nortel is probably in the best position to reap immediate benefits from the OIF's efforts, since it now has a commercial system based on the format the OIF&nbsp;has adopted.</p> <p>&ldquo;We use two subcarriers, but it&rsquo;s exactly that modulation scheme and that coherent detection. Obviously the specifics of the DSP algorithms and that will vary between vendors I imagine down the road,&rdquo; he said. &ldquo;But I don&rsquo;t see a big disconnect between where they&rsquo;re going and where we already are.&rdquo;</p> <p>DiPerna is therefore eager to see the results of the OIF's work, particularly in the area of electro-optic subcomponents -- detectors, polarization combiners, and coherent mixers were three areas he mentioned specifically. &quot;We&rsquo;ll have access to more components and get the cost down, etc., etc. As opposed to the first time, [when] we had to build it ourselves,&rdquo; he explained.</p> <p>He wasn't quite as optimistic about getting help immediately at the module level because of what he termed the &quot;serious hurdles&quot; such vendors face in developing useful products quickly. The electronic parts of coherent detection, particularly the algorithms associated with cleaning up signals, represent one major example.</p> <p>Nortel's algorithms are &quot;one of our critical secret sauces,&quot; he said. &quot;What we&rsquo;ve built into our CMOS custom chips is exactly that -- the digital coherent reception, the high-speed analog-to-digital and digital-to-analog [converters], and then all of those algorithms. So right now I still believe this is one of our critical differentiators,&quot; he continued. &quot;So they would certainly have to show that they&rsquo;ve taken it to the next step beyond what we have at this point. And certainly there&rsquo;s no evidence of that yet.&rdquo;</p> <p>&nbsp;</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/12/nortel-oif-100g-components-yes-modules-probably-not.html2009-12-22T20:14:00.000ZMotorola plans XG-PON field trialsnoemail@noemail.orgStephen Hardy<p>In the wake of <a href="/fttx/news/Verizon-tests-XG-PON-10G-GPON-with-Huawei-equipment-79551162.html">Verizon's recent XG-PON field trial </a>with Huawei, Motorola's director of global marketing and communications for Access Networks, Floyd Wagoner, says that his company also has field trials of 10-Gbps PON technology on tap.</p> <p>Wagoner expects the trials to take place in the first quarter of next year. While he wouldn't confirm that the trials will be with Verizon (rather, &quot;one of the largest deployers in the world&quot;), that seems a pretty safe bet. Motorola is one of Verizon's current GPON suppliers.</p> <p>Wagoner says that current worldwide interest in 10G PON technology -- including 10G EPON -- is pretty much of the tire-kicking variety. He placed the number of carriers with even that much interest at less than 10.</p> <p>He believes 10G PON won't reach wide deployment for another three or four years, due to a combination of lack of need and a fairly large price tag.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/12/motorola-plans-xg-pon-field-trials.html2009-12-18T22:05:00.000ZCiena overcomes NSN protest to win Nortel MEN court approvalnoemail@noemail.orgStephen Hardy<p>The bankruptcy courts overseeing Nortel's breakup turned aside objections from Nokia Siemens Networks yesterday and approved the sale of the Nortel Metro Ethernet Networks (MEN) division to Ciena.</p> <p>Apparently, the promise of $810 million was 11 days late and about $20 million short.</p> <p>Both Nortel and Ciena announced the result early this morning. Ciena now has full court approval to purchase essentially all of the MEN optical and Ethernet assets for $530 million in cash and $239 million in convertible securities. Ciena expects the deal to close in the first quarter of calendar 2010. Ciena has been granted early termination of the antitrust waiting period under the Hart-Scott-Rodino Act as well as termination of the applicable waiting period under the Canadian Competition Act. It still needs to clear some regional regulatory hurdles as well as statutory information-sharing and consultation processes with employee representatives in EMEA, and court approvals in France and Israel.</p> <p>In the asset auction, held November 20-22, NSN actually bid $1 million more than Ciena's $769 million total. However, a provision in <a href="/business/mergers-and-acquisitions/news/Ciena-becomes-Nortel-MEN-stalking-horse-63669792.html">the stalking horse bid</a> from Ciena called for the systems vendor to receive a $21 million breakup fee should the assets eventually go to anyone else. Thus, NSN's $770 million bid represented a net weaker offer by $20 million.</p> <p>However, after the admitting defeat in a press release early on the morning of November 23, NSN decided to raise an objection during yesterday's hearings to approve the auction's outcome that since so much of Ciena's bid was based on securities, which inherently carry risk, its true value was uncertain. The objection called for the bidding to be reopened -- and NSN promised as an incentive that it would submit an $810 million all-cash bid that would cover the $21 million breakup fee and more.</p> <p>Details of this aspect of the hearing remain sketchy, although <a target="_blank" href="">Reuters reports</a> that the courts ruled that NSN and its bidding partner, One Equity Partners, &quot;did not have standing to object to Ciena's bid.&quot;</p> <p>In fact, Reuters' report implies, Ciena and Nortel may have had more trouble from a parallel objection filed by a Nortel creditor. Matlinpatterson Global Investors asked the courts to demand further information about the securities for reasons similar to NSN's. Reuters reports that after hours of negotiation, Ciena agreed to change the pricing on its convertible securities under certain conditions and Matlinpatterson withdrew its objection.</p> <p>&nbsp;</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/12/ciena-overcomes-nsn-protest-to-win-nortel-men-court-approval.html2009-12-03T14:50:00.000ZNSN takes one more run at Nortel MENnoemail@noemail.orgStephen Hardy<p><strong>UPDATED </strong><a href="">Reuters is reporting</a> that Nokia Siemens Networks plans to petition the courts overseeing the Nortel bankruptcy proceedings to reject Ciena's winning bid for the Metro Ethernet Networks (MEN) group when they hold joint hearings today. And it's willing to put up $810 million in cash to make it worth the courts' consideration, according to the news service.</p> <p>Ciena <a href="/networking/news/Details-reactions-emerge-on-Cienas-win-of-Nortel-MEN-auction-71817652.html">was declared the winner of the auction</a> for most of the optical and Ethernet assets within MEN with a bid of $769 million -- $530 million in cash and $239 million in convertible notes. The auction began on November 20 and lasted until the night of November 22.</p> <p>Reuters quotes NSN spokeman Barry French as asserting the fact that Ciena's notes carry some risk put the true value of the winning bid in question.</p> <p>Reuters and <a href="" target="_blank">other media outlets</a> also say that NSN bid $770 million for Nortel's MEN assets, which gives the initial appearance that Nortel and its creditors left $1 million on the table. However, Nortel's <a href="/business/mergers-and-acquisitions/news/Ciena-becomes-Nortel-MEN-stalking-horse-63669792.html">stalking horse agreement</a> with Ciena provides Ciena with a $21 million breakup fee should another party end up with MEN. Thus, NSN would have had to exceed any Ciena bid by at least that amount to offer a better deal. An $810 million bid would do the trick.</p> <p>The court hearings are expected to begin at 11 AM EST in Delaware. However, the MEN assets are not the only sale on the courts' docket. Nortel's sale of its GSM/GSR assets to Ericsson and Kapsch CarrierCom AG also will be reviewed. Therefore, the session may not end until well into this afternoon.</p> <p>&nbsp;</p> <p>&nbsp;</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/12/nsn-takes-one-more-run-at-nortel-men.html2009-12-02T23:33:00.000ZCiena wins Nortel MEN auctionnoemail@noemail.orgStephen Hardy<p><a target="_blank" href="">Reuters is reporting</a> that Ciena has emerged victorious in the auction for most of Nortel's Metro Ethernet Networks (MEN) optical and Ethernet assets. Ciena will pay $769 million for the assets -- well above the $521 million it originally bid.</p> <p>Nokia Siemens Networks issued a statement just after midnight this morning confirming that &quot;it did not submit the highest bid for Nortel&rsquo;s optical networking and carrier Ethernet assets.&quot;</p> <p>The company statement added, &quot;Nokia Siemens Networks believes that its final offer represented fair value for the assets, and further bidding could not be financially justified.&quot;</p> <p>The auction process started Friday morning and lasted until Sunday.</p> <p>Reuters quotes an anonymous source as revealing that Ciena will surrender $530 million in cash and $239 million in convertible notes.</p> <p>This is the second portion of Nortel that Nokia Siemens Networks attempted to buy without success. The company paired with One Equity Partners, a private equity firm, in pursuit of the MEN assets. The NSN/One Equity Partners pairing apparently was the only competition Ciena faced in the bidding.</p> <p><a href="/networking/news/Details-reactions-emerge-on-Cienas-win-of-Nortel-MEN-auction-71817652.html">More details here in a separate article.</a></p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/11/ciena-wins-nortel-men-auction.html2009-11-23T05:42:00.000ZNortel's MEN on the auction block todaynoemail@noemail.orgStephen Hardy<p>The auction for most of the optical and Ethernet assets of Nortel's Metro Etherent Networks (MEN) division was slated to start right about now (10 AM EST). We'll keep you posted on the action as details leak, leading up to the announcement of the winner. Be patient -- depending upon how many bidders are involved and how many rounds are necessary, this could take a while.</p> <p>The only known bidder is Ciena, although both Reuters and Bloomberg have named Nokia Siemens Networks, partnered with private equity firm One Equity Partners as a competitor. That said, speculation remains that others may be involved.</p> <p>Nortel has cleared the decks for this event, pushing back the auction of its GSM/GSM-R assets, originally scheduled for today, to some unspecified date in the future.</p> <p><strong>UPDATE 11/23 12:53 AM EST</strong> <a href="/blog/Ciena-wins-Nortel-MEN-auction.html">Reuters says Ciena wins!</a></p> <p><strong>UPDATE 4:00 PM EST</strong> Nothin' but crickets chirping so far. Meanwhile, here are a few words about One Equity Partners, from the firm's website:</p> <p style="margin-left: 40px;"><span class="bodyText">One Equity Partners acts as a lead equity investor in management-led buyouts, and growth capital financing with a particular emphasis on corporate partnerships and divestitures. OEP typically invests $50 million to $200 million per transaction but historical investments have ranged from $3 million to over $500 million. </span></p> <p><span class="bodyText">The firm has made investments in the communications space previously, principally Genband</span>.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/11/what-am-i-bid-for-nortels-men.html2009-11-20T15:00:00.000ZCiena has competition for Nortel's MEN assetsnoemail@noemail.orgStephen Hardy<p>Ciena is going to have to pay more if it wants Nortel's optical and Ethernet assets.</p> <p>Nortel has announced this Friday, November 20, 2009, as the auction date for the assets for which <a href="/business/mergers-and-acquisitions/news/Ciena-becomes-Nortel-MEN-stalking-horse-63669792.html">Ciena earlier bid approximately $521 million</a>. That means at least one other qualified bidder was found yesterday.</p> <p>No word yet on who the competition is -- Nortel source says it's a private auction, so their lips are sealed. Naturally, I'll keep digging. Meanwhile, care to make any guesses?</p> <p><strong>UPDATE 11/19 11:00 AM EST </strong><a href="" target="_blank">Reuters is reporting</a> that Nokia Siemens Networks paired with private equity firm One Equity Partners to submit a qualified bid. One Equity manages $8 billion for JPMorgan -- which is a lot of fire power for Ciena to overcome. Ciena's stock rose $0.40 to $14 in after-hours trading yesterday when it became clear that someone else might buy MEN. It will be interesting to see how NSN fairs today.</p> <p><strong>UPDATE 11/19 9:41 AM</strong> <strong>EST</strong> Paul Bonenfant of Morgan Keegan issued a note on Opnext in light of a &quot;reported NSN bid for Nortel optical business.&quot; He doesn't see a Nokia Siemens Networks acquistion of MEN as a positive for Opnext, since an outside supplier of 40G and 100G technology would no longer be necessary.</p> <p>Meanwhile, I'm in touch with Nortel media relations to find out just how this auction process works. Not surprisingly, the judge doesn't stand up with his/her gavel and start yodeling. In fact, the judge isn't involved at all in the actual bidding process. Apparently, ground rules are first set, then Nortel and the attending creditors identify the initial high bid. The other potential bidders decide how they want to respond. New bids are then assessed by Nortel and the creditors. Other rounds ensue as necessary. My media relations contact said that one of the previous auctions required six rounds &quot;and that took all day.&quot; The judge gets involved a few days later at a hearing to approve the winning bid.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/11/ciena-has-competition-for-nortels-men-assets.html2009-11-18T21:40:00.000ZNortel MEN bid due date nears endnoemail@noemail.orgStephen Hardy<p>Today's the deadline for parties interested in pulling Nortel's MEN optical and Ethernet assets out from under Ciena. Nortel had <a href="/networking/news/Nortel-postpones-optical-networking-and-Carrier-Ethernet-asset-auction-69979697.html">extended the bid deadline last week</a>.</p> <p>A Nortel source says that an auction for the assets would likely follow a few days from now &quot;if there are other bidders.&quot; Ciena has <a href="/business/mergers-and-acquisitions/news/Ciena-becomes-Nortel-MEN-stalking-horse-63669792.html">submitted a stalking horse bid</a> of approximately $521 million that will win the assets if no other bidders materialize.</p> <p>The company will likely issue an update tomorrow, November 18.</p> <p><strong>UPDATE: </strong>11:25 AM EST, November 18: &quot;Short release&quot; in works, says Nortel source.</p> <p><strong>UPDATE: </strong>2:55 PM EST Still nothing. Perhaps they're waiting for the market to close.</p> <p><strong>UPDATE: </strong>4:45 PM EST <a href="/blog/Ciena-has-competition-for-Nortels-MEN-assets.html">The auction is on for this Friday!</a></p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/11/nortel-men-bid-due-date-nears-end.html2009-11-17T22:36:00.000ZSCTE Cable-Tec Reporter's Notebook, Day 3noemail@noemail.orgStephen Hardy<p>10/30 11:15 AM: Wait -- what's that thing up&nbsp;in the sky? Is that the sun?</p> <p>Another new-to-me company here is InnoTrans Communications, which specializes in 1550-nm transmitters. its Chromadigm product can support a wide variety of cable architectures, from HFC cable deep to RFoG and PON. The product has seen application for node splitting and fiber reclamation via WDM (it can be configured for WDM support of&nbsp;4 to 16 wavelengths), as well as upgrades from 1310 nm.&nbsp;The company's most recently announced customer is Cox Communications Northern Virginia for a 1-GHz optical transport upgrade project.</p> <p>NEC is also here in an attempt to get a foothold in the MSO space. Rich Moran says the company is leveraging its relationship with Transmode (which is also here at the show) to offer the Scandinavian company's WDM systems as well as its own packet transport platforms, particularly the recently announced MN Series. Large MSOs like to deal with large systems suppliers, Moran says, and that's where NEC comes in, particularly for equipment that supports business services delivery and wireless backhaul.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/10/scte-cable-tec-reporters-notebook----day-3.html2009-10-30T15:15:00.000ZSCTE Cable-Tec Reporter's Notebook, Day 2noemail@noemail.orgStephen Hardy<p>10/29 12:30 PM: The snow is still coming down...</p> <p>In a workshop session this morning, Mark Pelligri, manager, transport networks, at Cox, provided an overview of Lambda Zone, a proposed method of accommodating multiple business services over the same fiber-based collection ring. Lambda Zone would see passive OADM filters placed on the collection ring, from which optical access rings would deliver services to customers. Cox hopes to trial the technique in the first quarter of next year. More details to come.</p> <p>ZTE USA through its hat in the DOCSIS PON ring this morning. The ZXA10 DOCSIS Mediation Server enables the company's existing EPON OLT to integrate with an MSOs DOCSIS management structure.</p> <p>Oplink is at SCTE Cable-Tec for the first time, hoping to raise its profile among manufacturers and their customers in the MSO market. Among other things, the company is showing off a liquid-crystal-based 1x2 WSS, pluggable VOAs in SFP format, a wide range of XFP and SFP+ modules for 10-Gbps applications, as well as SFP BiDis and other modules for EPON applications. Oplink is also showing off WDM subsystems&nbsp;for prospective ODM/OEM customers.</p> <p>Alcatel-Lucent is showing off routers, but not its GPON gear. GPON and MSOs are a bad mix, a source at the booth told me.</p> <p>4:35 PM: Cisco is showing off its D-PON technology in a corner of its booth. (See the video <a href="http://video/video-display/51932327.html">&quot;Intro to DOCSIS PON&quot; </a>on our the Lightwave Channel's MSO Optics page for an introduction to Cisco's approach.) James Brannan, market manager, transport and access networks, told me that the company has changed its original plans for a standalone 3RU unit in the hub to a modular approach that sees transmitter and amplifier modules that fit into the Prisma II platform. Brannan said the corresponding ONT is a real differentiator; the &quot;RFoG plus&quot; capabilties include a proprietary FM scheme that gives the MSO the option of greater reach or more bandwidth. He says Cisco has customers for its D-PON technology in Europe and the U.S. (including Hawaii).</p> <p>Auxora's President and CEO Xin Zhang says his company is seeing success in the MSO space thanks to extended-temperature-range WDM products, which can be deployed in the outside plant.</p> <p>One of the good things about coming to shows is finding new (to me) companies. For example, Zycko is a Minnesota-based distributor of optical communications products. However, it has its own line of optical transceivers, branded as ProLabs. The line extends from 1 to 10 Gbps in the standard form factors. A source at the booth said that they think they can provide better quality control and customization when necessary through their own module line...Electroline is a Montreal-based company that's been in the RF business since the 1950s, but is now moving into the optical realm. The company recently was purchased by an investor group that includes Applied Optoelectronics. (The two companies bumped into each other when they used the same Chinese manufacturing facility.) The company's newest product is the very high density ELink Optical Transmission Platform, which can accommodate a wide range of optical services, including RFoG. The company has a companion RFoG micronode that can support PON overlay. Other items include a mini-node for MDU applications and a standalone transmitter unit....QAMnet also has a full fiber-optic line for MSOs. The line includes a variety of amplifiers and transmitters for HFC, FTTH, and RFoG applications. A lot of their work is custom, and their customers include several well-known government and academic research organizations.</p> <p>Pacific Broadband Networks (PBN) is attempting to mimic its Asian success in optical platforms for MSOs around the world. CTO Peter Saglietti reports the company is active in HFC, RFoG, and FTTH, with the modules that fit within the OCMR optical communications mainframe the cornerstone of PBNs optical approach. The OCMR modules provide support for HFC, 2.5G EPON, and Active Ethernet. The company also offers <a href="/mso-optics/products/PBN-to-unveil-new-ODN-RFoG-at-Cable-Tec-Expo-65515967.html">space-saving RFoG OSP nodes</a>&nbsp;as well as&nbsp;micronodes (it has an OEM relationship with Hitachi for the latter). Saglietti says the company has at least 10 million deployments of its platforms and asserts the company is the #1 supplier in China -- as well as the lead supplier to Telstra. Customers can also be found in Taiwan, Korea, Europe, and the U.S.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/10/scte-cable-tec-reporters-notebook-day-2.html2009-10-29T18:21:00.000ZSCTE: Phooey on FiOSnoemail@noemail.orgStephen Hardy<p>10/28 12:30 PM: Welcome to Denver, where the snow is piling up to the point where locals are having a tough time attending SCTE Cable-Tec without the aid of sled dogs. I'm opening up the reporter's notebook again this week, so check back for updates.</p> <p>The title of this morning's Technology Leadership Roundtable was &quot;Enough Already!&quot; &quot;Enough of what?&quot; you ask. Answers the roundtable description: &quot;Growing a little weary of all that FiOS&nbsp;in your face?&quot; The short answer, not surprisingly, is yes. Roundtable moderator Leslie Ellis (Ellis Edits LLC) opened the discussion by asking whether the cable-TV community should be defensive about the fact that it hasn't fully embraced FTTH -- particularly since the industry invented video over fiber and carries more video over fiber than anyone else.</p> <p>Much pooh-poohing of FTTH and telcos ensued. Paul Liao, president and CEO of CableLabs, said that the MSOs are the big dogs when it comes to video and becoming big dogs in voice delivery -- and when you're a big dog, you're going to attract competitive attention.</p> <p>Dermot O'Carroll, SVP, engineering and network operations, at Rogers Cable Communications up in Canada, asserted that fiber &quot;doesn't do much&quot; for voice or video (I assume he meant fiber access versus HFC) and perhaps only a little bit when it comes to Internet access. This last shortfall should go away with deployment of DOCSIS 3.0, he said.</p> <p>Liao agreed that DOCSIS 3.0-enabled HFC should prove more than adequate for customer needs today and into the future, adding that DOCSIS 3.0 should enable more bandwidth than anyone will ever need. (This sounds like one of those &quot;eat your words in 10 years or less&quot; statements, but Liao is certainly smarter than I am and more versed in DOCSIS 3.0 capabilities.)</p> <p>Meanwhile, at least two workshops later in the week will discuss how to migrate HFC networks to FTTH. It doesn't hurt to hedge your bets, apparently. Getting a better understanding of how MSOs really feel about FTTH is one of my goals here.</p> <p>Other nuggets from the roundtable: Migration to IPv6 will be a bear, particularly since you'll still need to support IPv4 for a while and hopefully find a way for customers with older IPv4-only devices to access new IPv6-enabled services. IP in general (including, of course, IPTV) will be the key to enabling &quot;anything to any device&quot; services. And as cablecos offer such services, they'll have to shift their focus from marketing to households to marketing to individuals.</p> <p>7:15 PM: Back here at the hotel after exhibits closed. The show floor was significantly busier than last week's Supercomm event. Opinions in the exhibit hall regarding how much of a market for FTTH MSOs represent were all over the map. (This will either be a standalone story or I'll run through them Friday once I talk to more people.) Meanwhile...Stefan Murry, VP, global sales and marketing, at Applied Optoelectronics reports a trend away from transceiver-based ONT designs for GPON toward designs based on discrete components. There's not enough volume in GPON to generate savings from mass deployment, so ONT designers are coming up with proprietary designs to extract costs.</p> <p>Hitachi is showing off three <a href="/mso-optics/products/Hitachi-designs-EPON-equipment-to-support-mobile-backhaul-residential-and-business-services-growth-67210527.html">new ONUs for the company's DOCSIS PON offering</a>, which carries the Salira label. (See the video interview from the recent FTTH Conference entitled <a href="/video/video-display/51932497.html">&quot;DOCSIS-based FTTH Options&quot;</a> for the company's take on the market.) The ONUs and the rest of the DePON gear leverages &quot;multi-wavelength&quot; (four of them, to be precise), which company sources have repeatedly emphasized isn't the same as WDM-PON. The company also has an SFP-based ONU (developed by Broadway) for enterprise applications...Motorola fresh off of <a href="/mso-optics/news/MetroCast-chooses-Motorolas-RF-over-glass-for-FTTH-66139017.html">announcing a customer for its RFoG gear</a> (see -- someone is buying this stuff!) and <a href="/mso-optics/products/Motorola-expands-Fiber-Deep-and-CablePON-portfolios-66147407.html">expansion of its CablePON and fiber deep portfolios</a>, also displayed an EPON ONU from Alloptic, with whom the company has a relationship. Moto's Floyd Wagoner suggested that MSOs are starting to show a preference for EPON over the GPON Motorola currently offers, and the company wants to be part of those conversations -- although a companion EPON blade for Motorola's OLT doesn't yet have a firm place on the company's product roadmap...Aurora Networks also announced <a href="/mso-optics/news/Aurora-Networks-MDU-and-rural-area-optical-nodes-now-available-in-the-Americas--66128322.html">expansion of a product line</a> that already spans HFC, fiber deep, RFoG, and <a href="/mso-optics/products/Aurora-Networks-debuts-new-RFPON-GEPON-CPEs-for-cable-market-67214832.html">RFPON</a> (the last an RFoG/EPON hybrid; see <a href="/mso-optics/featured-articles/RFoG-plus-PON--Enabling-cables-all-IP-future-64581082.html">&quot;RFoG plus PON &ndash; Enabling cable&rsquo;s &lsquo;all-IP&rsquo; future?&quot;</a>). VP of Marketing John Dahlquist says Aurora has more than 50 customers using fiber deep gear at about 30 using RFoG, with significant overlap among those two customer lists.</p> <p>Finally, Corning Cable Systems showed off new <a href="/mso-optics/access/products/Corning-Cable-Systems-to-display-all-fiber-access-network-offerings-at-SCTE-Cable-Tec-Expo-65519662.html">All-Fiber Access Network segmented-split OSP gear</a> designed to mimic their HFC counterparts. The complete offering includs the FlexNAP technology Corning is already offering to telcos for FTTH.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/10/scte-phooey-on-fios.html2009-10-28T19:40:00.000ZA few notes from Supercommnoemail@noemail.orgStephen Hardy<p>Wednesday the traffic was okay; Thursday it wasn't&nbsp;as good. Now, on Friday, it's a ghost town, with exhibitors itching to tear their booths down because the unions don't work on the weekend and they need to be out by the end of the day. Welcome to Supercomm 2009!</p> <p>Major optically related themes here at the show are mobile backhaul (synchronization capabilities based on 1588v2 and Synchronous Ethernet -- both providing them and testing them -- are big), service management, and packet optical transport. A fair amount of discussion about 100G, particularly when it might be deployed in more than one link. Estimates range from 2011 to years afterward.</p> <p>For you numbers junkies, Cisco issued some interesting statistics via its Visual Networking Index Usage study. They estimate that the average broadband connection generates 11.4 Gbytes of Internet traffic a month. They see overall broadband Internet usage climbing 5X between 2008 and 2013. However, they say carriers will have to engineer their networks to handle 7X for peak usage, which generally occurs between 9 PM and 1 AM.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/10/a-few-notes-from-supercomm.html2009-10-23T16:42:00.000ZCiena updates MEN shopping listnoemail@noemail.orgStephen Hardy<p><a href="">On October 8, I reported on a conversation</a> I had with John Marson, vice president of industry and strategic marketing at Ciena, regarding what his company wouldn't be buying as part of its proposed deal to acquire Nortel's optical and metro Ethernet assets. If you read the article, you'll recall that Marson wasn't entirely sure what Nortel products were on his company's shopping list.</p> <p>I recently received an email from Marty Querzoli of the Davies Murphy Group, which does PR for Ciena. Marty provided clarification of some of Marson's comments. According to Marty, Ciena indeed doesn't hope to acquire the Ethernet Access Service Terminal, the WDM-PON platform that's part of the LG-Nortel joint venture. The company does want Nortel's Optical Multiservice Edge 1000 family, as well as the HDX crossconnect family.</p> <p>Of course, wanting the products and actually acquiring them will be more complicated than your average M&amp;A activity, thanks to Nortel's bankruptcy. Ciena's $521 million offering merely serves as the opening bid for an auction of the assets, which is scheduled to begin November 13 at 9:30 AM ET. The deadline for initial submission of qualified bids is November 9 at 4:00 PM ET. The US and Canadian court hearing to approve the successful bid is slated to follow November 19 at 1:00 PM ET.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/10/ciena-updates-men-shopping-list.html2009-10-20T16:17:00.000ZCongratulations to fiber pioneer on Nobel nodnoemail@noemail.orgStephen Hardy<p>As you may have read recently, Charles K. Kao, one of the pioneers of optical fiber, received the Nobel Prize in Physics. (He's actually sharing the prize with two Bell Labs researchers for their work developing CCD sensors, but let's not quibble.) Not surprisingly, Kao was recognized &quot;for groundbreaking achievements concerning the transmission of light in fibers for optical communication.&quot;</p> <p>The Nobel committee described Kao's work as follows:</p> <p style="margin-left: 40px;">&quot;In 1966, Charles K. Kao made a discovery that led to a breakthrough in fiber optics. He carefully calculated how to transmit light over long distances via optical glass fibers. With a fiber of purest glass it would be possible to transmit light signals over 100 kilometers, compared to only 20 meters for the fibers available in the 1960s. Kao's enthusiasm inspired other researchers to share his vision of the future potential of fiber optics. The first ultrapure fiber was successfully fabricated just four years later, in 1970.&quot;</p> <p>The Nobel Prize is but the latest recognition of Kao's work; the list includes <a href="/lightwave-issue-archives/issue/engineers-honored-for-making-the-telecommunications-revolution-possible-53461642.html">the Draper Award from the National Academy of Engineering in 2000</a>. Based on the email I've received since the award was announced, Kao's achievement has cast a reflective glow across the entire optical communications community.</p> <p>Congratulations to Dr. Kao -- and to fiber optics in general.</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/10/congratulations-to-fiber-pioneer-on-nobel-nod.html2009-10-07T15:33:00.000ZA few notes from the FTTH Conferencenoemail@noemail.orgStephen Hardy<p>I'm reviewing my notebook from last week's FTTH Conference, and here are some of the tidbits that haven't already appeared on the site...For all the talk about how many broadband stimulus proposals were filed, several carriers at the show who might have filed decided not to because of &quot;too many strings&quot; attached to the potential funding...DSM Desotech beat the drum loudly for <a href="/fttx/news/DSM-Telcordia-team-for-fibercable-microbending-specification-57718937.html#">the microbend standard they're attempting to put together </a>with Telcordia. You'll see a video interview about it on our site in the near future...Telco Systems touted video surveillance as a service supported by the company's Active Ethernet approach. Telco Systems acquired the service technology when it bought Vigilant Technologies. City security services are a big pull, I'm told...Not surprisingly, bend-insensitive fiber cabling for MDU applications remained a point of emphasis across the show floor. Corning showed off its Rapid Pass drop cables, OFS crowed about its <a href="/fttx/news/Verizon-selects-OFSs-EZ-Bend-cable-to-support-MDU-FIOS-deployments-60087957.html">selection for Verizon applications</a>, and Draka highlighted its new <a href="/fttx/products/Draka-unveils-300-lb-drop-cable-with-bend-insensitive-fiber-62325362.html">300-lb drop cable</a>...OFS, meanwhile, is urging the ITU to come up with a C version of G.657 for &quot;ultra&quot; bend-insensitve fiber...LG-Nortel showed off 3D TV as a bandwidth driver for its WDM-PON gear...Sumitomo Electric Lightwave says it has a big-name customer for its indoor/outdoor ribbon drop cable...Gustavo Welkner, CTO at Chilean carrier GTD Manquehue, SA, estimated that there are 450,000 homes passed by FTTH technology in Latin America; 80% of these are in Brazil. His company is doing its part with a deployment in Santiago, Chile...</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/10/a-few-notes-from-the-ftth-conference.html2009-10-05T15:04:00.000ZRural telco offers clever parry to incumbent pricing schemenoemail@noemail.orgStephen Hardy<p>War stories of the competition between upstart rural or municipal FTTH service providers and the local incumbent MSO or telco are staples of the FTTH Conference. This year's event in Houston, wrapping up today, is no exception. I heard one such story yesterday that was fairly amusing.</p> <p>EATEL is a small carrier in Louisiana that, like the <a href="/fttx/news/ftth-council-joins-lus-fight-in-supreme-court-53433707.html">Lafayette Utilities System</a> in nearby Lafayette, was attempting to get <a href="/fttx/featured-articles/eatel-brings-triple-play-to-cajun-country-53429417.html">its FTTH-based services off the ground</a> in the face of competition from Cox.&nbsp; As commonly happens, Cox aggressively dropped its service prices in EATEL's territory to hold onto its customers. However, the MSO's marketing strategy made it subject to a state regulation that the price structure had to be honored elsewhere in its footprint, which included not only Lafayette but also Baton Rouge.</p> <p>So EATEL took out ads in Lafayette newspapers and on radio stations (and had sympathetic bloggers spread the word as well) to alert the citizens of Lafayette that Cox was selling communications services in EATEL's territory for a much lower price than was being offered in Lafayette -- but if they called Cox and cited the state regulation, they could get that same bargain price.</p> <p>Needless to say, Cox wasn't pleased -- particularly with the implication that EATEL would undoubtedly be willing to do the same thing in Baton Rouge.</p> <p>Cox didn't raise its prices (it changed its approach to remove the obligation to offer the low prices outside of EATEL's territory) but, according to EATEL's president, it &quot;got their attention&quot; and &quot;took some of the pressure off.&quot; Sometimes tweaking the tiger's tail is victory enough.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/09/rural-telco-offers-clever-parry-to-incumbent-pricing-scheme.html2009-09-30T14:06:00.000ZECOC Reporter's Notebook -- Day 3noemail@noemail.orgStephen Hardy<p>[8:36 AM CET] Sitting in the press room before the final day of the show begins, saying last rites for my cell phone and watch batteries, both of which are about to die. Should make keeping appointments an interesting challenge. Meanwhile, here are a few stray notes from yesterday I didn't have a chance to cover...Opnext is showing off <a href="/top-stories/Opnext-showcases-40100G-products-at-ECOC-60448982.html">a prototype 100G CFP for LR4</a>. They're also highlighting a 40G DQPSK module in 300-pin format and an SFF VSR module. The DQPSK module has begun shipping to customers and complements the <a href="/video/video-display/52689047.html?player=32673428001&amp;title=19091734001">continuously optimized DPSK technology</a> introduced at OFC. The latter is good for applications involving lots of ROADMs, the former for links where fiber impairments are the primary impediment to reach...Along with MergeOptics (see the Day 1 blog) and Yenista (whom I neglected to mention in yesterday's test roundup) Proximion is another European company claiming to be going great guns despite the economy. The company is doing well with its tunable optical dispersion compensators for 40G, and is showing off a&nbsp; <a href="/equipment-design/proximion-releases-patch-cord-dispersion-compensation-module-54895322.html">compensator in a patch cord format</a> introduced in June. The company's fiber grating expertise enables the design... 3S Photonics of France (ex-Alcatel Optronics, ex-Avanex) is highlighting its pump lasers, including a new, <a href="/equipment-design/components/products/3S-PHOTONICS-unveils-980-nm-pump-laser-module-59779697.html">uncooled device for submarine applications</a>, as well as <a href="/equipment-design/components/news/3S-PHOTONICS-shows-analog-laser-module-protoype-at-ECOC-60457582.html">laser technology for OFDM</a>...Sumitomo Electric Device Innovations is quietly touting a tunable XFP-E which company sources say should beging sampling the fourth quarter of this year. Meanwhile, a <a href="/top-stories/Sumitomo-Electric-develops-40GbE-transceiver--60446587.html">40GbE CFP module</a> will sample in Q2 2010 and a 100GbE version the folowing quarter, if current plans hold...</p> <p>[8:00 PM CET] The exhibition and the poster sessions are over, but this blog isn't...The 100G test conversations continued today.&nbsp; Daniel van der Weide, vice president of engineering at Optametra, asked to meet me to make the case for the real-time approaches on which <a href="/test-and-measurement/products/Optametra-Coherent-Lightwave-Signal-Analyzers-available-for-100G-physical-layer-test-60085087.html">his company's offering</a> (as well as Agilent's) is based, versus the EXFO approach he had seen highlighted <a href="/blog/ECOC-Reporters-Notebook----Day-2.html">in this blog yesterday</a>. van der Weide suggested that people who are evaluating 100G test platforms should look closely at bandwidth claims and focus on informational bandwidth. He said that real-time approaches can more closely follow what's going on in a signal, since it's tracking all the bits, as opposed to analyzing optical samples. Among the benefits of the real-time approach is a greater capability to provide accuragte BER measurements, he asserted. You can bet we'll here more about this in the near future...van der Weide also expressed the hope that oscilloscope manufacturers would increase the bandwidth of their systems...Meanwhile, Aragon Photonics of Spain is also taking an optical approach to 100G and related testing. However, their approach leverages stimulated Brillouin scattering...Synthesys Research demonstrated brand new (as in &quot;We just finished this box before we got on the plane to Vienna&quot;) capabilities to do stressed eye testing at 28.7812 Gbps, plus other capabilities for 100GbE requirements. The demonstration occurred in EXFO's booth, as the companies' equipment complements each other. Official announcement of the new capabilties (and, perhaps, details of the relationship with EXFO?) should occur around OFC next March...I neglected to mention yesterday that JDSU displayed its full <a href="/test-and-measurement/products/jdsu-reveals-100-gigabit-ethernet-test-suite-55093337.html">range of 100GbE test gear</a> as well.</p> <p>On the transceiver front, Finisar has joined the CFP party, with a demonstration of a prototype 40G LR4. It also touted the capabilities of its <a href="/equipment-design/transport/news/Finisar-optimizes-1x2-WSS-for-telecom-network-edge-60455032.html">Edge Wavelength Processor</a>, a 1x2 WSS based on the company's liquid crystal on silicon technology...Gigalight of China has a range of 10G ZR (80 km) transceivers that should be available over the next two months. These include XENPAK, X2, and DWDM XFP. A company source asserted that Gigalight is one of the few Chinese module vendors with its own SFP+; the company hopes to offer an ER module in this form factor next year.</p> <p>&nbsp;Inphi touted its differential 2811DZ 40G DQPSK modulator drivers in surface-mount packages. The company has demostrated interoperability of these devices with the SMT versions of Sierra Monolithics' 40G mux/demux devices. Loi Nguyen, vice president of broadband analog products for Inphi, expressed a sense of vindication that the company pursued DQPSK when many others were betting on DPSK for 40G. He said that the extremely small package sizes of the drivers might obviate the need for quad chips, at least from an economic perspective...Narda, meanwhile, touted <a href="/general/narda-to-exhibit-modulator-drivers-at-ecoc-55093347.html">its modulator drivers</a> for 40G VSR, ODB, DPSK, and DQPSK drivers in GPPO packages. A source at the booth confirmed the obvious sense of playing in the 100G market.</p> <p>And speaking of 100G, sources at the Discovery Semiconductor booth reported the company is making good progress on shrinking its KittyHawk coherent transmitter/receiver technology into something smaller than a rack unit. The hope, the sources said, is to come up with a line card that might be of interest to systems companies. Meanwhile, the company spread the word about its new 100GbE (4x25G) quad PIN-TIA optical receiver for LR and ER applications.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/09/ecoc-reporters-notebook----day-3.html2009-09-23T07:36:00.000ZECOC Reporter's Notebook -- Day 2noemail@noemail.orgStephen Hardy<p>[12:45 PM CET] First, some leftovers from yesterday. JDSU European Director Sales Sinclair Vass expanded on the company's <a href="/equipment-design/transmission/news/JDSU-advances-tunable-strategy-with-new-component-building-blocks-60090712.html">tunability announcement</a> by saying that the company will sell the ITLA and even the TOSA that enable their tunable XFP, but only to &quot;select people&quot; -- mostly those that would use them in non-telecom applications. The tunable XFP is aimed first at replacing fixed-wavelength XFPs, with unseating 300-pin transceivers next. A tunable SFP remains in the works. Sinclair also offered his opinions on coherent technology at 40G as well as the company's strategy for coming up with the coverter, DSP, and algorithms for coherent detection, which I hope to include in a separate story on 100G strategies (so stay tuned)...Speaking this morning, Javed Patel, CEO of Sierra Monolithics, is confident in his company's competitive position at 100G. When it comes to the receiver electronics, he sees two camps -- all CMOS (such as Fujitsu Microelectronics Europe is pursuing) and SiGe for the front end and CMOS for the rest, which is what his company is doing in its partnership with IBM. He expects to be working mainly with systems companies rather than transceiver vendors because the first generation of coherent receivers will by ASIC based -- and the ASICs will come from the systems houses....</p> <p>&nbsp;[3:35 PM CET] And now, a round up of what's new here in test. The major thrust, not surprisingly, is in 40G/100G test capabilities. Agilent has expanded the capabilities of its optical modulation analyzer by <a href="/equipment-design/test/news/Agilent-adds-bit-error-ratio-analysis-to-optical-modulation-analyzer--60256947.html">including BER functionality</a>. The feature is designed to complement other BER measurements to enable users to not only detect the fact that errors occur, but what part of the system is causing them...EXFO, meanwhile, is leveraging its Picosolve acquisition <a href="/test-and-measurement/products/EXFO-launches-turnkey-optical-modulation-analyzer-for-100-Gbaud-signal-characterization-59648697.html">to offer its own optical modulation analyzer</a> as competition. Peter Andrekson, who joined EXFO upon the acquistion and heads the company's activities in its Swedish facilities (as well as teaching at the Chalmers University of Technology), touts the fact that the PSO-200 is based on an all-optical sampling system (versus the real-time, electrically based approaches of companies like Agilent and Optametra). Among other benefits, the all-optical approach enables the testing of higher-speed and/or more complex modulation approaches. So, for example, the instrument is already able to handle 400-Gbps DQPSK as well as 16QAM...Anritsu is demonstrating that it has <a href="/test-and-measurement/design-and-manufacturing/news/Anritsu-offers-signal-generationanalyis-up-to-56-Gbps-59991047.html">increased the performance of its MP1800A Signal Quality Analyzer Series</a>. The unit now offers signal generation and analysis up to 56 Gbps. Mux/demux options enable testing of 100G applications as well...In other test applications, mdi says that Optus, SingTel's division in Australia, has decided to deploy the company's eyeD 360 network monitor. The company has also developed a version of the eyeD for lab applications...Luna Technologies is showing off its OVA 5000 Optical Vector Analyzer, <a href="/test-and-measurement/design-and-manufacturing/news/Luna-Technologies-offers-new-Optical-Vector-Analyzer-59647572.html">introduced earlier this month</a>.&nbsp;</p> <p>&nbsp;</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/09/ecoc-reporters-notebook----day-2.html2009-09-22T11:41:00.000ZECOC Reporter's Notebook -- Day 1noemail@noemail.orgStephen Hardy<p>The Reporter's Notebook returns, this time for ECOC in Vienna. Check back regularly, as I'll be adding to the notebook as the day progresses. (Twitter? Who needs Twitter?)</p> <p>[3:30 PM CET] Skipped opening address by George Gilder, based on first experience hearing him speak at this year's Executive Forum. Apparently, he is a true oracle -- because, without a high priestess to interpret his pronouncements, I couldn't follow what he was saying. Went to exhibit floor instead, where 85% of the vendors were still setting up their booths...u2t had its act together, which featured an integrated 43G DPSK receiver demo with two devices, from Kylia and Optoplex, respectively. u2t supplies the amplifier, photodiode, and related technology, and their opto customers add the delay line interferometer, etc. A lot of this work ends up being custom, they tell me; they have several such projects underway. Meanwhile, they're looking at integrated 40G DQPSK receivers, since the technology requirement is doubled. While a fully integrated receiver would be ideal, they may offer integrated versions of the two main parts as an initial, interim step. Oh -- and then there's the <a href="/equipment-design/news/Optical-receiver-manufacturers-announce-100G-MSA-59777412.html">MSA-compliant 100G receiver</a> prototype...Molex featured a connector with an integrated SFP, a relatively old offering that may find applications in LTE equipment. The arrangement removes the SFP cage from the PCB, enabling space savings as well as more flexibility in terms of where the interface can be located. The company also is finding success with its I-Pass connector technology. Standardized by the InfiniBand community, the I-Pass also shows promise for CXP applications, the source at the stand believes. He added that he sees the CXP significantly eating into the market for QSFP devices....Speaking of parallel interfaces, MergeOptics is going into parallel applications in a big way, particularly active optical cables (AOCs). These include 120G cables using both CXPs (featuring the I-Pass technology) and QSFPs. Company CEO Dag Neumeuer says that business is up 30% to 40% from last year, due primarily to the AOCs. Neumeuer sees QSFPs eroding demand for SFP+ devices (apparently before QSFPs give way to CXPs, if you follow Molex's thinking). MergeOptics has parallel products for the consumer space on the drawing board as well, Neumeuer hinted...JGR Services was showing off a pair of custom-built products they hope have wider appeal. They built the Functional Modular Platform, an eight-slot modular test platform, for Alcatel-Lucent for systems test. Meanwhile, they built a field-level backreflection measurement platform with 72 ports for AT&amp;T...</p> <p>[5:00 PM CET] Exhbition closing down. After slow start, attendance picked up dramatically this afternoon...RED-C Networks reports increasing interest in hybrid Raman/EDFA amplifiers with uptick in 40G deployments. Raman acts as a pre-amp, the EDFA as a booster...CoreOptics offering 40G SerDes so that Sierra Monolithics finally has some competition. Also debuted second-generation 40G VSR transponder. But the major source of conversation was the use of coherent detection for 40G applications. More on this to come separately...NeoPhotonics showing off coherent mixer (shipping for 40G applications but applicable to 100G) and second-gen DQPSK demodulator. By reducing birefringence problems with their PLC-based approach, they feel the new DQPSK demodulator equals the performance of free-space-based demodulators and outshine them in all other aspects. Meanwhile, they acknowledge that customers will want to see detectors and TIAs integrated into the mixer, and they're working on it...ADVA Optical Networking's 100G priorities, according to CTO Christoph Glingener, in descending order of importance (but increasing order of cost): four wavelengths of 28G, their <a href="/general/adva-demos-serial-100g-transmission-for-metroregional-applications-54893407.html">DPSK-3ASK modulation format,</a> and the OIF's 2Pol-QPSK with coherent detection. It's the cost aspect that has set these priorities, he says. Meanwhile, he says <a href="/fttx/ftth-b/featured-articles/Staying-ahead-of-the-access-bandwidth-curve-with-WDM-PON-56740887.html">the company's push into WDM-PON</a> has netted three trials so far. He notes that their implementation of WDM-PON technology, which is targeted at metro apps such as wireless backhaul, can be ring-based as well as tree or point-to-point. The company is pondering offering a residential version in the next few years. Finally, he also see ADVA targeting cloud computing applications via its FSP 3000...</p> <p>&nbsp;</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/09/ecoc-reporters-notebook----day-1.html2009-09-21T14:28:00.000ZWhat do you know -- IPOs!noemail@noemail.orgStephen Hardy<p>Here's something you probably didn't think you'd see again anytime soon: a pair of successful IPOs in the optical communications space.<br /> <br /> Avago Technologies started the ball rolling earlier this month when it offered 43,200,000 ordinary shares on NASDAQ at a public price of $15.00 per share. (Underwriters purchased an additional 6,480,000 ordinary shares from selling shareholders via the underwriters' over-allotment option.) The shares, under the listing AVGO, are trading at $17.78 as I write this at about 2:15 EDT. <br /> <br /> Meanwhile, on the other side of the world, <a href="">Accelink launched its IPO today on the Shenzhen Stock Exchange</a>. The initial asking price for the shares, trading under the number SZSE 002281, was approximately $2.34. However, online sources suggest it opened at $3.95 a share and closed at approximately $4.36.<br /> <br /> Not bad for an industry that's supposed to be hurting, eh? While one can point to the last round of optical IPOs and assert that early exuberance will quickly fade, I still think it's encouraging that the offerings even took place, never mind that the stocks are trading above the initial asking price.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/08/what-do-you-know----ipos.html2009-08-21T16:00:00.000ZRon Martin reduxnoemail@noemail.orgStephen HardyPosted by <a href=mailto:>Stephen Hardy</a><br /><br /><b>UPDATE:</b> So having belatedly blogged about Ron Martin's departure from ADVA Optical Networking Wednesday, it comes to light Thursday that he has a new job: vice president for worldwide sales at Infinera. (Scott Chandler and Howard Lukens remain at the Digital Optical Networking company.) As always, timing is everything in the media business...<br /><br />Here's the Wednesday post:<br /><br />In the midst of announcing the results for the second quarter of 2009, ADVA Optical Networking also revealed that Ron Martin, who had joined the company in November 2007 as chief marketing and strategy officer with a focus on expanding ADVA Optical Networking's North American customer base, left the company at the end of June. Brian Protiva, company CEO, positioned the departure as a cost-cutting move arrived at "by mutual agreement."<br /><br />Here's an interview I did with Martin at last year's NXTcomm in which he explained what he hoped to achieve.<br /><br /><embed src="" bgcolor="#FFFFFF" flashVars="playerId=31944703001&viewerSecureGatewayURL=" base="" name="flashObj" width="450" height="380" seamlesstabbing="false" type="application/x-shockwave-flash" swLiveConnect="true" pluginspage=""></embed>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/08/ron-martin-redux.html2009-08-05T14:40:00.000ZNSN: How to take down Cisconoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Today's news of <a href=,-IBM-enter-data-center-OEM-agreement/>Juniper's OEM agreement with IBM</a> puts me in mind of the switch/router vendor's link with Nokia Siemens Networks (NSN) for <a href=,-Juniper-pair-for-IP-over-DWDM/>an integrated approach to IP over DWDM</a> -- a concept <a href=>practically owned by Cisco</a>.<br /><br />I spoke earlier this month to Pathmal Gunawardana, director of business development at NSN, who explained why the combination of NSN and Juniper could beat Cisco at its own telco network game. While he declined to make a feature-to-feature comparison of Juniper's boxes with Cisco's, he was much more willing to suggest that NSN provided the stronger DWDM portfolio.<br /><br />Specifically, said Gunawardana:<br /><ul><br /><li> NSN has a broader, more scalable DWDM offering. While Cisco's gear is principally targeted at the metro, in Gunawardana's view, NSN can offer metro, long-haul, and ultra-long haul platforms, he said.<br /><li> Since NSN was the market leader in 40-Gbps sales in 2008, the company would be better able than Cisco to leverage volume to create a cost-effective 40-Gbps interface -- not to mention a future 100-Gbps interface.<br /></ul><br />Speaking of these interfaces, Gunawardana said that the 40-Gbps interface would likely leverage the company's current DPSK offerings, at least initially. The company has targeted what he called "CP-QPSK" -- for "coherent polarized QPSK" -- as the technology of choice for the 100-Gbps interface. CP-QPSK will be similar, if not identical, to the dual-polarization QPSK with coherent detection the OIF is currently working on. Once perfected, the technology would likely be used for 40 Gbps as well, he predicted.<br /><br />Gunawardana said that both interfaces should be available within the next 18-24 months, although he hinted that the 40-Gbps capability would likely appear in the early part of that window and the 100-Gbps technology toward the end. Since the interface will be integrated into the Juniper hardware, Gunawardana conceded that it could be used with DWDM equipment not from NSN. However, such a pairing would likely fall short of the performance a Juniper/NSN combination would provide, he asserted.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/07/nsn-how-to-take-down-cisco.html2009-07-22T18:38:00.000ZCommerce waives broadband 'Buy American' provisionnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />The Secretary of Commerce on June 19 granted a limited waiver of the American Recovery and Reinvestment Act's "Buy American" provision to most of the broadband equipment that would likely be purchased under the Broadband Technology Opportunities Program (BTOP). The sigh of relief coming from potential BTOP applicants -- not to mention their potential suppliers -- could almost be heard in every corner of the United States (and a few places outside of the country as well).<br /><br />The BTOP represents the lion's share of the money set aside for telecommunications projects under the Obama Administration's economic stimulus program. ($4.7 billion; the Department of Agriculture's Rural Utilities Service will oversee the dispersal of an additional $2.5 billion.) The "Buy American" provision in the Recovery Act states that no funds, including those earmarked for the BTOP, "may be used for a project for the construction, alteration, maintenance, or repair of a public building or public work unless all of the iron, steel, and manufactured goods used in the project are produced in the United States." Fortunately, the act also provides the head of a federal department or agency with the authority to waive this provision for any of three reasons, including a determination that applying the "Buy American" criterion would be against the public interest.<br /><br />That's exactly what the Department of Commerce decided. I'd make the case for why this action was necessary, but I don't have to -- the Department has done it for me within its waiver notice (which you can <a href=>find here</a>). The provision would put an unreasonable burden on potential applicants to prove that the hardware required to build its network was made from materials from U.S. sources, the Department concluded. There are several reasons for this, according to the notice:<br /><br />1) Much of the equipment used to manage and operate broadband networks is manufactured outside of the United States, using complicated and constantly varying supply chains.<br />2) The waiver will facilitate the roll out of modern broadband networks incorporating the latest technology, which is a large part of the BTOP's purpose.<br />3) As such networks are built and operated, jobs will be created.<br />4) While the Office of Management and Budget "has clarified which countries would be exempt rom the Buy American provision, some of the key countries that produce broadband equipment would not be exempt."<br />5) The broadband industry is "very dynamic and global" according to the notice, and therefore the equipment used in a project can change in the midst of network rollout.<br /><br />The waiver covers broadband switching equipment, routing equipment, transport equipment, access equipment, CPE and end-user devices, and billing/operations systems. It does not cover fiber and cable, the notice emphasizes. However, applicants can apply for a special waiver as part of their applications should they feel it necessary.<br /><br />It's gratifying that the Department of Commerce has shown some common sense in determing how best to meet the twin goals of aiding U.S. businesses and bringing broadband capabilties to currently underserved areas of the country. Now if we could just get the application process rolling...http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/06/commerce-waives-broadband-buy-american-provision.html2009-06-30T15:23:00.000ZVerizon expects 100G by end of yearnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Speaking to a group of media and analysts after last Friday's OIF interoperability demo at Verizon's Waltham, MA, facility, the carrier's vice president of network architecture, Stuart Elby, said he expects that at least one of his current vendors, and perhaps as many as three, will deliver tenable 100-Gbps networking platforms by the end of this year. He added that he expects the platforms will be based on technology "like" the dual-polarized QPSK with coherent detection around which the OIF has rallied the industry (including Verizon), saying he believes the platforms will be "as close to that as exists" at the time.<br /><br />Elby also said he expects that Verizon will deploy some of the equipment it receives, but not in large numbers. As was the case with 40G, he expects the first generation of 100G platforms will be extremely expensive, and greater deployment will wait until further iterations of the technology reduce 100G's price tag. He said he had doubts that 40G prices would ever reach a level 2.5 times that of 10G, partly because the price of 10G technology continues to shrink. He said that 100G might enjoy a more aggressive downward cost run than 40G, due to greater deployment in data center environments.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/06/verizon-expects-100g-by-end-of-year.html2009-06-22T14:56:00.000ZAT&T mum on potential procurement strategy changesnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />AT&T's response to inquiries regarding Morgan Keegan & Co.'s note yesterday regarding a plan to reduce its supplier count to two for each of about 14 technology domains: "We have no comment on this."<br /><br />Simon Leopold, communications equipment analyst and managing director at the broker-dealer, issued a note yesterday saying he was starting to take seriously information he had received regarding a potential AT&T plan to drastically reduce the number of equipment vendors with which the company does business. The goal, based on what Leopold said he had heard, was to have only two supplier in each of "roughly" 14 technology domains. (Leopold did not list the domains in the note.) Naturally, it seems likely that AT&T would end up with fewer than 28 suppliers, on the assumption that some suppliers would remain viable in multiple domains.<br /><br />The initiative has three goals, according to the note: cost savings, risk reduction (particularly in the face of Nortel's bankruptcy filing), and streamlining of major projects. AT&T's success with Alcatel-Lucent on the U-Verse roll out served as a proof point of this last element, Leopold suggests.<br /><br />While Leopold asserts, "[w]e consider it premature to panic," this news, if true, clearly would make systems suppliers nervous -- particularly smaller ones whose narrower product lines would potentially lessen their opportunities to stay engaged.<br /><br />As I've indicated above, AT&T isn't shedding light on Leopold's report. (The quote above came courtesy of Jenny Bridges, who handles trade media inquiries at the carrier.) If anyone has any further info, I'm all ears.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/06/at-t-mum-on-potential-procurement-strategy-changes.html2009-06-11T20:54:00.000ZCiena: Things may look better -- but aren't yetnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Ciena announced unaudited results for fiscal second quarter 2009 today. Not surprisingly, the company reported down revenue; what was perhaps unexpected was a whopping GAAP net loss of $503.2 million, most of which was a non-cash charge of $455.7 million for impairment of goodwill.<br /><br />The magnitude of the loss may have obscured the somewhat hopeful note of the commentary:<br /><br />"Our fiscal second quarter was particularly challenging, reflecting the difficult macro and industry environment and continued delays in customer spending," said Gary Smith, Ciena's CEO and president. "While recent service providers' public commentary about expected annual capital expenditures has given the industry reason to be more optimistic about the second half of the year, our customers continue to spend cautiously, and as a result, our visibility remains limited. However, based on our direct conversations with customers and supported by trends we are seeing currently in the business, including recently improved order flow, we expect to deliver sequential revenue growth in our fiscal third quarter."<br /><br />The question, of course, is whether that rebound will occur and will it be significant. Simon Leopold, communications equipment analyst at Morgan Keegan & Co., still thinks Ciena is a good bet. "We maintain our Outperform rating on Ciena," he wrote in a note issued today. "Despite the poor April quarter, good sequential improvement leaves us optimistic. Challenging visibility remains, but sequential improvement, a new product cycle, net cash per share near $3 and a CY10 EV to sales of 1.0x suggests the stock has upside potential. One could argue for a fair value near $15 based on an EV/Sales ratio of 1.5x."http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/06/ciena-things-may-look-better----but-aren-t-yet.html2009-06-04T17:13:00.000ZIEEE P802.3ba chief talks opticsnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />As part of my research for a June issue article on 100G semiconductors (When will we see them?), I spoke with John D'Ambrosia in his role as a spokesman for the Ethernet Alliance. John also happens to be the head of the IEEE P802.3ba task force charged with developing the 40- and 100-Gigabit Ethernet specifications. While talking about chips, John also had a few things to say about optics for these emerging applications. Herewith a sampling:<br /><br /><b>On keeping 40GbE and 100GbE optical specs as simple as possible:</b> "If you look back at the 10-gig optics, what initially came out of the group was more complicated than desirable, because most of the interfaces were XAUI based and you basically wound up having several layers that you put into your optics solution -- which quickly evolved to 'take all that stuff out.' And now we have simple solutions like the XFP or SFP+."<br /><br /><b>Which doesn't mean you won't see functions pulled out second- and third-generation 100GbE modules:</b> He foresees three stages of module evolution. which are spelled out in <a href=>a whitepaper on the Ethernet Alliance site</a>.<br /><br /><b>On the <a href=>recently announced 40GbE serial effort</a>:</b> "For this new effort, people are looking at doing a serial-type interface at 40-gig. And the carrier people are really driving this one. They really want to see a 40-gig serial interface that will allow them to coexist easier with their OC-768 equipment....I don't know where they're going to go with the electrical interfaces on that yet, and I think that remains to be seen for that project. I think that they'll probably leverage in the short term off of the same NAUI-type interface that we're talking about [for the current singlemode-based specs], and then have your internal muxes."<br /><br /><b>On the prospects for on-time ratification of the task force's current standards work:</b> "At this point, I don't really see anything that's going to throw us off of schedule."http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/05/ieee-p802-3ba-chief-talks-optics.html2009-05-21T20:42:00.000ZEuropean cablecos get special treatmentnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Aurora Networks today <a href=>announced the NC2000</a>, a new optical platform it developed for the European cable-TV market. In a pre-briefing I received on the announcement, I asked what was so European about it. The answers to that question provided an insight into the European cableco market.<br /><br />Not surprisingly, the NC2000 is based on Aurora Networks' existing NC4000 platform. However, it needed to be repackaged for the European market. The company's vice president of marketing, John Dahlquist, says that most European cablecos have buried plant and the NC4000 was designed primarly for pole mounting. So the platform had to be packaged to so that most of the outputs were on one end of a shorter, more compact package that could be vertically mounted. Naturally, the trend in new housing developments in the U.S. is toward underground cabling, so Dahlquist says Aurora plans to offer a similar configuration to U.S. cablecos in the future. The other major difference is that European cablecos operate in different wavelength bands than their U.S. counterparts.<br /><br />Meanwhile, Aurora Networks will offer its RFoG, RF PON, and related FTTH and "Fiber Deep" capabilities. Dahlquist says that European cablecos have expressed interest in RFoG and wouldn't necessarily insist on starting the standards process for such a capability from scratch. The fact that the current SCTE efforts are going to be PON friendly should make for a smooth transition to the European marketplace, he believes.<br /><br />Dahlquist says the platform was developed by specific customer requirements and that deployment announcements should come in the near future. It will be interesting to see how close to the customer those carriers run fiber.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/05/european-cablecos-get-special-treatment.html2009-05-19T18:31:00.000ZNortel breakup imminent?noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />The <i>Financial Post</i> of Canada has a story on its website today suggesting that Nortel could announce the first major asset sale tomorrow, with others following in the near future.<br /><br />The enterprise business group will be the first to go, the <i>Financial Post</i> reports, perhaps as soon as tomorrow and no later than next week. The <i>Post</i> identifies Avaya and Siemens Enterprise Communications as the chief competitors for the prize.<br /><br />Meanwhile, the MEN group could be sold in as soon as five weeks; the <i>Post</i> says the group was on the point of being sold when Nortel filed for bankruptcy. Huawei, Alcatel-Lucent, Fujitsu, "and at least one private equity firm" are said to be vying for the group.<br /><br />You can <a href=>read the story on the <i>Post</i>'s site</a>.<br /><br />If Nortel does go the breakup route, it will be a huge blow to the Canadian optical communications industry. Besides being the country's flagship fiber-optics company, Nortel has been home to many of the technologists and executives responsible for the other Canadian companies in the space. If the company that acquires MEN pulls the group's resources out of Canada, the vibrancy of Ontario's photonics community will diminish significantly.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/04/nortel-breakup-imminent-.html2009-04-30T18:32:00.000ZOclaro: What's in a name?noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />I'm always curious about how companies come up with their names, particularly in this age when nothing escapes the scrutiny of the spin doctors. So I asked Howard Jones, who does PR for what used to be Bookham, how the Bookham/Avanex combine came up with "Oclaro." He sent me the following, which I was told to consider as straight from the mouth of CEO Alain Couder:<br /><br /><ul>"The name Oclaro was created as a combination of 'Optical' and 'Clarity.' The word communicates our clarity of vision, with an emphasis on knowing where we are going with a clear direction in the photonics market.<br /><br />"The new tagline, 'Shining Light on Photonic Innovation,' further emphasizes our dedication to innovation, leveraging our components, modules, and systems-level expertise in photonics."<br /></ul><br />I have to admit that when I first saw it I immediately thought of a compact car (as in "...the new Ford Oclaro!"). That said, like any new name, it probably just needs a bit of time to settle in.<br /><br />What do you think?http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/04/oclaro-what-s-in-a-name-.html2009-04-28T14:41:00.000ZSaw taken to NoCal fiber optic cable?noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />In case you missed it -- which means you don't live in Northern California -- someone (or more than one) apparently took a hacksaw to AT&T and Sprint Nextel cables in four locations near San Jose and San Carlos early Thursday morning. The police are currently investigating, and no suspects or motives have been identified publicly. <br /><br />"We didn't do it," the AP quoted Libby Sayre, area director for the California chapter of the Communications Workers of America, as saying. The union currently is in contract negotiations with AT&T.<br /><br /><a href=>Service was restored in about 24 hours</a>.<br /><br />The affected cable ran in underground conduits about 10 feet below ground level, the AP reports.<br /><br /><a href=>AT&T has upped its reward offer from $100,000 to $250,000</a> for information leading to the arrest and conviction of anyone involved.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/04/saw-taken-to-nocal-fiber-optic-cable-.html2009-04-13T18:20:00.000ZMore info on Australia's NBNnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a> <br /> <br />A few stray bits of info to add to the story on the most recent twist in Australia's National Broadband Network (NBN) plans <a href='ll-do-broadband-build-ourselves/>that I posted yesterday</a>: <br /><ul> <br /><li> The Australian Government has posted a whitepaper entitled "National Broadband Network: Regulatory Reform for 21st Century Broadband" that lays out more details of what it has up its sleeve. <a href=>You can find it here</a>. <br /><li> One interesting tidbit contained in the paper that I neglected to include in my story is that FTTH will be mandated for greenfield developments starting in July 2010. <br /><li> Some have stated that the Government's new plan is a positive outcome for Telstra, since it potentially provides a way for the incumbent carrier to participate in the NBN <a href=>after being shut out of the original RFP</a>. However, the document makes clear that the Government plans to put the squeeze on Telstra to further open its network to competitors -- and this outcome was one of the major reasons Telstra balked at submitting a complete response to the RFP. The document also ponders the possibility of limiting the types of businesses Telstra could acquire or forcing it to divest its hybrid fiber/coax network. <br /><li> Reaction from the companies involved in the RFP were mixed. <a href=>Telstra</a> and <a href=>Singtel Optus</a> appear ready to make the best of things, while TransACT is "<a href=>disappointed</a>." <br /></ul> <br />One good place to follow the action is <a href=>Australian FTTH News</a>, a blog by consultant Stephen Davies. <br /> <br />http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/04/more-info-on-australia-s-nbn.html2009-04-08T15:13:00.000ZMSO PONs now or later?noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />The NCTA Cable Show in Washington, DC, this week inspired a few announcements targeting the use of FTTH technology by MSOs, based either on RFoG, PON (<a href=>here</a> and <a href=>here</a>), or <a href=>both</a>.<br /><br />But while <a href=,-too/>many have speculated that RFoG is merely a waypoint on the path to PONs</a>, John Dalquist, vice president of marketing at Aurora Networks, says MSOs aren't exactly stampeding toward the use of PON for residential applications. Most of his customers that are thinking about serving homes with fiber are going the RFoG route, he says. PON, for the most part, is being reserved right now for the delivery of commercial services to businesses.<br /><br />The exception is in Europe, where there's interest in what Dalquist called a "blast and split" approach that looks a lot like Verizon's combination of RF broadcast and IP services via PON.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/04/mso-pons-now-or-later-.html2009-04-03T20:10:00.000ZOFC Reporter's Notebook 3noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />I can tell it's the second day of exhibits by the fact that my handwriting has gotten appreciably more difficult to read. The over/under on the time at which my brain completes its transformation to cottage cheese has been set at 2:41 PM on Thursday.<br /><br />Meanwhile...<br /><ul><br /><li> The folks at EXFO are very happy with the reception of some of their new products, in particular their <a href=>FTB-5600 Distributed PMD Analyzer that measures PMD in a manner reminiscent of an OTDR</a>. However, given the current economic environment, there are a significant number of potential customers who would like access to the capabilities but can't afford to buy the box. EXFO is currently contemplating how to address this issue. The impression I got was that they're leaning towards enabling some sort of test service offering, either directly or through partners. It's all part of a new corporate strategy that I'm guessing we'll see roll out later this year.<br /><li> If you're feeling nostalgic for iolon's old MEMS-based tunable laser, Luna Technologies will sell you one--or more than one, since they've set a minimum lot size for the device, which they acquired from Coherent. Meanwhile, they've incorporated the laser into a line of tunable light sources with very fast sweep rates, with the Phoenix 1200 the most recent (and most compact) example.<br /><li> In discussions regarding <a href=>ADVA's announcement of a phase-based modulation format</a> as an alternative to the dual-polarization QPSK the OIF favors for 100G, there seems to be concensus on the show floor that the long haul and metro will indeed see the application of different modulation schemes, at least initially.<br /><li> Harry Bosco, who will move from president and CEO of Opnext to chairman at the end of the month, suggests the company could take the opportunity provided by the downturn to do a bit of vertical integration via M&A.<br /><li> Oh yeah -- <a href=>I'm doing videos, too.</a><br /></ul>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/03/ofc-reporter-s-notebook-3.html2009-03-26T06:45:00.000ZOFC Reporter's Notebook 2noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a> <br /> <br />Here's a look at what people told me today: <br /><ul> <br /><li>JDSU introduced its tunable XFP, based on its integrated laser and Mach-Zehnder (ILMZ) modulator. The ILMZ is packaged in a special TOSA. The device also features a JDSU-designed controller ASIC. In addition to the module, a JDSU spokesman said the company will offer the TOSA and ASIC as separate products. <br /><li> <a href='s-TXFP-transceiver-tunes-across-more-than-90-channels/>EMCORE also unveiled plans for a tunable XFP</a>. However, ExceLight Communications, the US arm of Sumitomo Electric Industries (SEI), thinks there will still be demand for the tunable XFP-E it has on display. The XFP-E will be a more capable replacement for very high performance 300-pin transceivers than your average XFP, said SEI staffers at the booth. Replied a JDSU spokesman, it sounds like SEI is targeting ultra-long-haul applications -- and if SEI thinks they can make a living off of those applications, good luck to them. <br /><li> Execs at Ignis Photonyx, which makes WDM-PON subsystems (among other things), foresee strong pull from business service applications for WDM-PON. <br /><li><a href=>AMCC's extremely flexible Yahara 10GbE framer/mapper/PHY</a> can be used in metro packet optical transport platforms as well as 100G applications. But AMCC sources said they'd need to develop new chips to meet the requirements of the long-haul packet optical transport platform <a href=>Verizon is talking up</a>. <br /><li>At a press luncheon, University of Minnesota Andrew Odlyzko suggested that the current "bandwidth is growing at 50% annually" rule of thumb could be overshooting the mark by about 15% if you're talking worldwide figures. <br /><li>OFS is now in the splicer and optical components business, thanks to a North American reorganization its parent, Furukawa Electric Co., Ltd., has initiated. OFS will now market pump lasers, signal lasers, connector parts, and other components, as well as the fusion splicers, under the Fitel name. In addition to publicizing a tunable dispersion compensator, OFS also highlighted a reconfigurable dispersion compensation module, which comprises multiple lengths of dispersion compensating fiber and a small switch. <br /><li> They're hiring in the market research space. Andrew Schmitt has shut down Nyquist Capital (where he authored <a href=>one of the more interesting blogs in the space</a>) to join Infonetics Research as directing analyst for optical. Meanwhile, LightCounting has hired Brad Smith as senior VP. Smith will be responsible for the development of LightCounting's new market coverage in transceiver-related semiconductor and optical markets and for managing the company's consulting operations. <br /> <br /> <br />http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/03/ofc-reporter-s-notebook-2.html2009-03-25T04:26:00.000ZOFC Reporter's Notebook Inoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />With all due modesty, the OSA/Lightwave Executive Forum was the place to be on the first day of OFC/NFOEC week. In addition to <a href=>Verizon's pitch for a long-haul packet optical transport platform</a>, the day featured several interesting tidbits:<br /><ul><br /><li> What's the killer app? Keynoter Surya Panditi, vice president and general manager of Cisco's Access and Transport Technology Group, touted video. However, Steve Carlton, VP of planning and product management at Fujitsu Network Communications, suggested that applications such as medical imaging, which require terabits of capacity at a time, will have an even greater impact. But closer to home, Joseph Huggins, director, access and transport technology management at Qwest, said not to forget about online gaming.<br /><li> Kou Miyake, director, NTT Service Integration Laboratories, revealed that it will take 20 million subscribers for NTT to make money with its FTTH services. The company hopes to reach that figure by the end of fiscal 2010.<br /><li> Vik Saxena, senior director, network architecture, office of the CTO at Comcast Cable, said his company would deploy 100GbE technology today if it were available, driven more by operational efficiencies than pure bandwidth demand. However, Verizon's Elby said he expects that it will take two or three generations of 100G technology development before the technology meets the necessary price points for wide acceptance.<br /><li> Despite a darker economic environment than at this time in 2008, the Component Vendors panelists weren't nearly as cranky as they were last year. A wag in the audience suggested that's because they were in shock. However, I think that last year, component execs felt that they hadn't derived the rewards they thought were due them from a rebounded market. In the current environment, depressed margins and company valuations make more sense.<br /><li> Along these lines, Source Photonics chief Near Margalit suggested that a significant percentage of optical components vendors should get used to the idea that their margins will never exceed 25%.<br /><li> The panelists on our M&A panel don't expect a big play along the lines of the Finisar/Optium merger this year. With credit tight, cash will be precious -- and won't be spent without an extremely good reason. That said, private companies' expected valuations appear to be going down to increasingly tempting levels. But no one is interested in buying market share. Acquisitions must increase product breadth or meet some other strategic objective.<br /></ul>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/03/ofc-reporter-s-notebook-i.html2009-03-24T06:39:00.000ZWhy Meghan won't be at OFC/NFOECnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />If I may be allowed a digression from the usual line of conversation in this space, it is my honor to announce that Senior Editor Meghan Fuller Hanna gave birth last night to the Red Sox' 2027 first round draft choice. John "Trip" Hanna III entered the world at 9:34 PM weighing 5 lbs and measuring 18 inches. (I'm assuming that's head to toe.)<br /><br />Now it just so happens that yesterday was my birthday as well. Is the timing of Trip's birth a coincidence in light of the fact that, in addition to Trip, Meghan also delivered a partially finished Tech Trends article for our April issue that I'm now on the hook to complete? You decide.<br /><br />Needless to say, you won't see Meghan at OFC/NFOEC in a couple of weeks. Sadly, you also won't see her byline here on the site or in the magazine for a few months. Our expectation is that she'll have time for us again in June -- assuming she's not too busy hitting grounders to #1 Son.<br /><br />We'll have pics when they're available. In the meantime, please join me in congratulating Meghan on her new product introduction.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/03/why-meghan-won-t-be-at-ofc-nfoec.html2009-03-12T19:57:00.000ZMenara Networks, one year laternoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />A year after <a href=>debuting its "OTN in a module" concept</a> at OFC/NFOEC 2008, <a href=>Menara Networks</a> plans to use this year's event to highlight its electronic dispersion compensation (EDC) technology, company founder, president, and CEO Siraj Nour ElAhmadi told me earlier this week.<br /><br />ElAhmadi says that the company has garnered six design wins for its upcoming XFP device, which he expects will reach general availability in either the second or third quarter of this year. Like Menara Network's other 10G modules, the XFPs leverage the company's ASIC technology to incorporate the electronic functions required to support OTN capabilities within the transceiver, rather than on the board.<br /><br />Menara Networks produces devices for both long-haul and short-reach applications. The long-haul market was the first target; the company successfully completed a 1300-km trial in Sweden, ElAhmadi says. However, the company also sees interest in its technology for intra-CO applications, which are normally dark spots for OTN-based network management. He sees a role for Menara Networks' technology for OTN-based service demarcation as well.<br /><br />Both switch/router vendors and telecom equipment developers are working with the Menara Networks' modules, ElAhmadi reveals. The company also has seen interest from service providers, including cable companies, he adds.<br /><br />In addition to fully implementing its EDC technology, Menara Networks also has full C-Band tunability on its development roadmap. ElAhmadi hopes to have both capabilities available sometime next year.<br /><br />But perhaps the company's most noteworthy achievement was landing third-round funding last November. ElAhmadi terms the pursuit "extremely difficult -- we were lucky." He says this round should take the company through break even or cash positive.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/03/menara-networks-one-year-later.html2009-03-06T19:31:00.000ZOpVista on the chopping block?noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Paul Bonenfant, communications components analyst at Morgan Keegan & Co., released a note today saying his sources suggest that DWDM supplier OpVista may be on its last legs. Bonenfant says "industry chatter" indicates that the company has furloughed staff and is searching for a buyer; if it doesn't find one, it could close as early as this week. Bonenfant suggests that Emcore is the company most likely to snap up OpVista's assets.<br /><br />Contacted for comment, an OpVista source offered an interview with President and CEO Karl May scheduled for tomorrow, February 25, at noon EST. (February 25 update: <a href=>Read the results of the interview</a>.)<br /><br />OpVista has seen most of its success in the cable-TV space; it includes Cox Communications and Time Warner Cable on its international roster of customers. Its differentiation is the ability to cram multiple wavelengths into the space normally required for one. The company leveraged this technology last year to <a href=>introduce its CX8 optical networking system</a>, a 40-Gbps platform that used what the company called "Dense Multi-Carrier" technology to transmit four 10-Gbps wavelengths in each window of the ITU-T grid. The transmission would therefore act much like a 10-Gbps transmission, which would enable carriers to maintain their current 10-Gbps network design rules and obviate concerns about such impairments as polarization mode dispersion.<br /><br />The company <a href=,-IP-services/>also offers ROADM capabilities</a>.<br /><br />You can <a href=>see a video interview with Karl May</a> on the Lightwave Channel.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/02/opvista-on-the-chopping-block-.html2009-02-24T15:09:00.000ZOvum: European FTTH needs governmental rolenoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Inspired by his attendance at <a href=>this week's FTTH Council Europe conference</a> in Copenhagen, <a href=>Ovum</a> Senior Analyst Charlie Davies has issued a few thoughts on the catalysts necessary for FTTH to grow more rapidly on The Continent. <br /><br />Noting that a busy exhibition floor highlighted the growing importance placed on next-generation access (NGA) infrastructures, Davies added that "there was an equal recognition that for fibre to flower, it needs a lot more nurturing: from private investors, governments and regulators alike." However, governments in particular will have a primary role to play, he said. For example:<br /><br /><ul>"The pragmatic approach of Danish incumbent TDC – rolling out fibre as and when the market requires it and when there is a clear business case to do so – would seem sensible to many commercial companies. But at a time when governments and business leaders posit the fundamental importance of a robust NGA infrastructure to the economy/recovery/future of their respective countries, this approach leaves a considerable chasm. This is making the prospect of 'patchwork quilts' of NGA access unfolding out over Europe a distinct possibility with top down (large telcos/cablecos/private-funding) meeting bottom up (utility/community/public & private funding.<br /><br />"Key to the 'bottom up' is a more active role by regional and local authorities who have a much more vested interest in their region’s overall economic and social progress. So rather than being purely operator driven broadband rollout, with an eye on NGA access and more ubiquitous coverage comes under the wing of the region, working in partnership with private companies. This approach has been key to rollout of fibre in Sweden and is being replicated elsewhere in Europe on a larger scale. A number of regions including South-West France (Pyrenees-Atlantiques) and Southern Spain (Catalonia) are embarking down this road."<br /></ul><br />Davies notes that regulators can either help or hurt this process -- particularly as they attempt to encourage private investment. "We expect the next 6 months to be crucial as regulators at both an EU and a national level get to grips with a complex, changing, but at least vibrant landscape," he concluded.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/02/ovum-european-ftth-needs-governmental-role.html2009-02-13T21:44:00.000ZBookham/Avanex: Why now?noemail@noemail.orgStephen Hardy<p>Posted by <a href="">Stephen Hardy</a><br /> <br /> One of the more interesting aspects of <a href="/general/bookham-avanex-finally-agree-to-merger-54892242.html">yesterday's announcement that Bookham and Avanex had agreed to merge</a> was the timing -- what took so long?<br /> <br /> Alain Couder and Giovanni Barbarossa were asked during yesterday's conference call about the merger's timing. While Couder answered that there was no reason to wait out the downturn before consumating the deal, Barbarossa piped in with &quot;My question was why we didn't do it yesterday!&quot;<br /> <br /> Given that has been pretty much the sentiment among most observers of the space, I got through to Yves LeMaitre (thanks Howard!), currently Bookham's vice president of telecommunication sales and vice president of corporate marketing and future head of the combined company's non-telecom business, and posed the question to him.<br /> <br /> According to LeMaitre, the timing had a lot to do with the current macro-economic environment. The time is right for consolidation in the industry, he feels. Also, the companies' current valuations increased the feasibility of a merger. Finally, LeMaitre echoed a story Couder told during the call about a &quot;trial run&quot; interaction between the executives of the two firms that convinced the Bookham president and CEO that the two companies could work compatibility together.<br /> <br /> One has to wonder at this point whether <a href="/general/avanex-announces-executive-management-changes-54888202.html">the departure of Jo Major</a> as Avanex president and CEO last July was the result of a difference of opinion within the senior management ranks about whether it was time to find a buyer. Given Barbarossa's enthusiasm for the deal, it certainly appears he's happy with what has taken place.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/01/bookhamavanex-why-now.html2009-01-28T19:44:00.000ZThe Avanex/Bookham merger is on!noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />The long-rumored merger of Avanex and Bookham is a rumor no longer. I'm on hold as I write this, waiting for the conference call to start.<br /><br />According to someone at Bookham's PR firm, it's believed that Bookham will own a majority stake in the combined company, something like 53.25%. Bookham also will have a 4-3 majority on the board.<br /><br />More soon...the conference is starting now.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/01/the-avanex-bookham-merger-is-on-.html2009-01-27T23:02:00.000ZHow do you like us now?noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />As you may remember, <a href=>I highlighted back in August</a> the fact that Lightwave magazine was undergoing a redesign. The fruits of our labors have finally been unleashed on the world -- and I'm curious about what the world thinks.<br /><br />Those of you who subscribe to the magazine should have received an email yesterday from me offering links to both HTML and PDF versions of the January issue. If you know us only from this website, <a href=>you can check out the 'zine here</a>.<br /><br />My <a href=>editorial in the December issue</a> described what we were hoping to accomplish. Basically, we've tried to combine the readability of a printed magazine (you can print out the articles easily if you really like paper) with the multimedia and interactive aspects of websites.<br /><br />How did we do? You can comment below or <a href=>go to our corner of the Interconnection World community</a> on our sister site, Connector Specifier.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/01/how-do-you-like-us-now-.html2009-01-22T15:33:00.000ZOvum: Nortel's move may result in more balanced industrynoemail@noemail.orgMeghan Fuller HannaPosted by <a>Meghan Fuller Hanna</a><br /><br /><a href=>Ovum</a> analysts Dana Cooperson and Matt Walker believe that Nortel's bankruptcy filing may result in a more balanced industry structure for communications equipment.<br /><br />In a comment issued this morning, Cooperson and Walker note that when the latest phase of Nortel's downward spiral took shape four months ago--with the announcement that it would explore divestiture of its Metro Ethernet Networks (MEN) division--they argued that "a more radical approach than divestiture was needed to cure the company's woes." <br /><br />While the bankruptcy filing is radical, the analysts believe it will also be exploited by the company's competitors, who are "now in a strong position to remind customers that Nortel can no longer give assurances of continued development of any specific products, which will surely impede Nortel's ability to bring in new business."<br /><br />In fact, say Cooperson and Walker, Nortel's bankruptcy may open the door for broader industry rebalancing. They cite four specific examples, excerpted here:<br /><br /><blockquote><br /><b>Taking on Cisco</b><br />In data networking, Cisco remains dominant in every region and in most product segments . . . . For a company targeting Cisco, bits and pieces of Nortel's Enterprise and MEN units are clearly attractive. Juniper, Tellabs, and Ciena would benefit from looking carefully at Nortel. All have some experience with growth through M&A, and have geographic and cultural similarities. They also have some product overlap, but buying a competitor just to get them out of the market is not an unheard of strategy. More important, Nortel has channel depth outside of North America, which is of high value to these companies.<br /><br /><b>40G/100G jumpstart</b><br />The MEN's 40G/100G business, which Nortel now puts at 42 customer wins--defined as purchase orders or contracts that include 40G, not deployments--is an attractive focal point for a slimmed down Nortel or a competitor looking to limit its own R&D and jumpstart its customer list. Nortel has done a good job promoting its solution's viability over competitors' networks, so virtually all its competitors should be interested in this asset.<br /><br /><b>Diversification through acquisition</b><br />Ericsson's purchase of Marconi and Redback--after their respective bankruptcies-- provides a blueprint for what is likely to be the ultimate outcome for Nortel: acquisition by a firm looking to fill out its equipment product line. For example, Ericsson and Nokia Siemens Networks both have gaps in their wireline portfolios and little position in enterprise. Acquiring significant chunks of Nortel may be attractive to both, and their relatively high cash reserves could make it possible; as of September 2008, both vendors had just under $10 billion in cash and short-term investments. <br /><br /><b>Evolution to 4G</b><br />Nortel's mobile infrastructure business is now focused on LTE/SAE (long term evolution/system architecture evolution). It is working hard to develop a strong LTE/SAE ecosystem, including LG Electronics, LG Nortel, and other partners. It is doing its best to demonstrate its capabilities through trials (e.g., Verizon and T-Mobile Germany) and announcements (e.g., a deal with KDDI) and expects some commercial launches in 2009. Its LTE assets (part of the Carrier division) may be attractive for another player, perhaps Alcatel-Lucent, NEC, or ZTE.<br /></blockquote><br /><br />Finally, Cooperson and Walker concede that Nortel's decision to file for bankruptcy now, when it still has $2.6 billion in cash reserve, may enable it to re-emerge as a smaller, more focused version of itself. However, they believe the scenarios they mapped out above, in which "rivals use Nortel's bankruptcy as a chance to reshuffle the supplier landscape dramatically to their benefit, seem more likely."http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/01/ovum-nortel-s-move-may-result-in-more-balanced-industry.html2009-01-16T18:47:00.000ZIndustry reaction to Nortel's bankruptcy filingnoemail@noemail.orgMeghan Fuller HannaPosted by <a>Meghan Fuller Hanna</a><br /><br /><a href=>Nortel</a> yesterday announced that it was filing for bankruptcy, a move that did not exactly come as a surprise, given the rumors swirling around the industry over the last several months. <br /><br />In September, Nortel announced that it would <a href=>explore a divestiture of its Metro Ethernet Networks (MEN) Business</a>, including its optical and Carrier Ethernet portfolios. Its most frequently cited suitors included Huawei, Cisco, and Nokia Siemens Networks, with Alcatel-Lucent and Motorola considered long shots as both struggle with problems of their own. <br /><br />In December, <i>The Toronto Globe and Mail</i> reported that Nortel had received offers from "three serious bidders" and was considering selling off additional assets in lieu of seeking bankruptcy protection. In the meantime, the company continued to burn cash, the value of its shares continued their free-fall, and the company inched closer to NYSE delisting. In late December, news broke that the company was, in fact, exploring bankruptcy as an option, and several analysts argued that this could be its best course of action. In a research note dated December 19, 2008, UBS analyst Nikos Theodosopoulos offered the following viewpoint:<br /><br /><blockquote><br />NT has an interest payment of $100-$120 million due on Jan 15. We believe a likely challenging 4Q08, weakening '09 outlook, and tightening DIP financing may cause NT to withhold its interest payment and possibly pursue an early bankruptcy. If the company cannot sell its MEN division or get additional assistance from the Canadian government, an early bankruptcy may make sense to maximize franchise value.<br /></blockquote><br /><br />And now the deed is done, leaving the telecom industry to ponder what may be next for the Canadian telecom giant. Will the company emerge from bankruptcy stronger than ever, or will it be forced to sell assets in what could amount to a fire-sale?<br /><br />In an article from yesterday's <i>New York Times</i>, <a href=>"Nortel Seeks Bankruptcy Protection"</a>, Ian Austen cites several analysts who believe that the company is likely headed for liquidation. If they are correct, he writes, <br /><br /><blockquote><br /> . . . the end of Nortel would be one of largest failures in the telecommunications equipment business . . . . Nortel's demise would also be among the biggest business failures in Canadian history. During the zenith of the technology boom, Nortel's market value accounted for about a third of all equity traded on the Toronto Stock Exchange.<br /></blockquote><br /><br />A <a href=>news article</a> in yesterday's <i>Wall Street Journal</i> quoted telecom analyst Ping Zhao of CreditSights, who "gave Nortel little hope of emerging from bankruptcy. 'They were already out of favor due to their weak finances,' [she said,] 'but for any of the new projects, they are definitely out of the picture' due to the filing."<br /><br />Today's <i>Toronto Globe and Mail</i> argues that <a href=>"Nortel needs a plan, fast</a>." Writers Simon Avery, Jacquie McNish, and Shawn McCarthy note that Nortel has not yet formulated "a master plan on how to re-emerge a stronger company."<br /><br /><blockquote><br />An approved agenda will most likely include the sale of assets, but whether that disposition leaves the Nortel brand alive or amounts to a full liquidation of the century-old company remains hotly debated.<br /><br />"A breakup is not a top priority for the business. On the contrary, it's to be able to come out the other side as a nimbler, more focused, successful technology company," Mike Zafirovski, [Nortel's] president and chief executive officer, said in an interview. "There is no announcement today regarding strategy."<br /></blockquote><br /><br />Over at <a href=></a>, reporter Amy Thomson notes that some of Nortel's customers, including Verizon, are vowing to stick around, but they may already be weighing their options.<br /><br /><blockquote><br />Verizon, the largest U.S. phone company and Nortel's biggest customer, "isn't doing . . . anything different about Nortel today than yesterday," spokesman Eric Rabe said. The company accounted for 11 percent of Nortel's $10.9 billion in 2007 sales. Rabe said Verizon probably won’t change its relationship with Nortel in the short term.<br /><br />Should Nortel be dissolved, Verizon has agreements with other network providers, including Cisco, for the parts it needs, Rabe said. Verizon has already moved some of its business to Cisco to meet demands for new technology, he said.<br /></blockquote><br /><br />Finally, <i>Toronto Globe and Mail</i> columnist Derek DeCloet argues in <a href=>today's edition</a> that Nortel should not rely on a government bailout but instead needs to refine--or perhaps redefine--its corporate vision:<br /><br /><blockquote><br />It's too early for government help. Nortel must first undergo a corporate soul-searching exercise: What does the company want to <i>be</i>? Where does it want to compete, and where does it want to give up? Assuming it is not dismantled entirely, the future Nortel will certainly be smaller, and focused on perhaps one or two lines of business.<br /></blockquote><br /><br />So what do you think? Does bankruptcy represent Nortel's best path forward?http://localhost:4503/content/lw/en/blogs/lightwave-blog/2009/01/industry-reaction-to-nortel-s-bankruptcy-filing.html2009-01-15T20:58:00.000ZTrack Santa at the speed of lightnoemail@noemail.orgMeghan Fuller Hanna<a href=""><img style="DISPLAY: block; MARGIN: 0px auto 10px; WIDTH: 152px; CURSOR: hand; HEIGHT: 122px; TEXT-ALIGN: center" alt="" src="" border="0" /></a><br /><br /><div>Posted by <a href="">Meghan Fuller Hanna</a><br /><br />In the words of that wonderful 1897 editorial from the New York Sun, "Yes, Virginia, there is a Santa Claus." And now you can track him in real time as he makes his way around the globe.<br /><br />For the 50th year, the <a href="">North American Aerospace and Defense Command (NORAD)</a>, the bi-national U.S. and Canadian military organization responsible for the aerospace defense of both countries, will be tracking Santa Claus' Christmas Eve flight. </div><br /><div>Personnel at the NORAD Tracks Santa (NTS) operations center, which is located at the Peterson Air Force Base in Colorado, say they will use fighter jets, satellites, radar systems, and <a href="">Qwest Communications'</a> high-speed fiber-optic network to keep a watchful eye on that jolly old elf.<br /><br />"While Santa's eight tiny reindeer can move his sleigh full of toys pretty fast, Qwest's fiber-optic network moves at the speed of light, so we'll be able to stay at least one step ahead of them and quickly report their whereabouts to everyone tracking the flight," says Diana Gowen, senior vice president and general manager of Qwest Government Services.<br /><br />Beginning at 6:00 a.m. EST tomorrow, children will be able to track Santa's flight and view live video feeds (captured by special Santa Cams in space, of course) via the <a href="">NORAD Tracks Santa Website</a>, which is available in English, French, German, Italian, Japanese, Spanish, and Chinese.<br /><br />You may also want to explore the site, which features lots of interesting information, particularly in the <a href="">FAQ about Santa</a> section. Do you know, for example, the path of Santa's flight?</div><br /><div>According to NORAD scientists, "Santa usually starts at the International Date Line in the Pacific Ocean and travels west. Historically, Santa visits the South Pacific first, then New Zealand and Australia. After that, he shoots up to Japan, over to Asia, across to Africa, then onto Western Europe, Canada, the United States, Mexico, and Central and South America. But keep in mind, Santa's route can be affected by weather," they warn, "so it's really unpredictable. NORAD coordinates with Santa's Elf launch staff to confirm his launch time, but from that point on, Santa calls the shots."<br /><br />Here's another one for you: NORAD's satellites, positioned in geo-synchronous orbit 22,300 miles from the Earth's surface, use infrared sensors to detect heat--including the infrared signature of Rudolph's red nose, allowing the satellites to detect Santa's sleigh wherever it is. And you thought Rudolph's nose was only good for guiding Santa's sleigh through the fog!<br /><br />For the science nerds (and who among us isn't?), the Website even provides technical data about Santa's sleigh--the length, height, and width of which are measured in candy canes (cc) divided by lollipops (lp). You can also learn about the propulsion and the fuel, but not the emissions: That's classified.<br /><br />Finally, for all of us who are just too darn excited to sleep, NORAD scientists issue the following warning: "We cannot predict where and when [Santa] will arrive at your house. But we do know from history that it appears he arrives only when children are asleep. In most countries, it seems Santa arrives between 9:00 p.m. and midnight on Christmas Eve. If children are still awake when Santa arrives, he moves on to other houses. He returns later . . . but only when the children are asleep."<br /><br />Here's to a good night's sleep!</div>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/12/track-santa-at-the-speed-of-light.html2008-12-23T20:30:00.000ZVerizon: No rollout plans yet for 100 megnoemail@noemail.orgStephen HardyPosted by <a href="">Stephen Hardy</a><br /><br /><a href=",-weighs-what">Despite the impression</a> he gave us (<a href="">not to mention Telephony</a>) during his keynote presentation at Optical Access '08 last Thursday, Verizon Director of FTTP Architecture and Design Vincent O'Byrne did not announce Verizon's intention to roll out 100-Mbps services over its FiOS FTTH network in 2009. So says Jim Smith, Verizon Telecom's director of media relations, in an email we recently received.<br /><br />"What I and he thinks he said was we'd make the engineering choices and construction to make 100 meg possible; neither of us remember him saying we'd sell it," Smith wrote.<br /><br />Just to put a cap on the matter, Smith concluded, "It is not on anyone's actual product rollout plans at this point."<br /><br />Were we (and others) hearing things? You can judge for yourself by <a href="">listening to O'Byrne's keynote</a>.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/12/verizon-no-rollout-plans-yet-for-100-meg.html2008-12-11T15:40:00.000ZAvanex preparing for boarders?noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Paul Bonenfant, communications components analyst at Morgan Keegan & Co., reports that Avanex filed an 8-K after the market closed last night revealing that the compensation committee of the board of directors had approved "change in control agreements" for the CEO (which, as of November 18, is now officially Giovanni Barbarossa, who also serves as company president and member of the board), senior vice presidents of sales and operations, and other VPs (including the interim CFO). The agreements are meant to provide incentives for these executives to remain with the company upon change of control -- in other words, if it's sold or merged.<br /><br />This led Bonenfant to wonder if the much discussed hook up with Bookham is on again. He points out that while the combination looks good on paper, it would hold execution risks for Bookham and likely would delay its move toward profitability (as if the current economic situation hadn't done that already).<br /><br />On top of Bonenfant's concerns, I recall that Bookham chief Alain Couder has repeatedly told me that if he gets involved in M&A, it will be to move into new markets. "My experience in merging two companies with similar product lines in the same market is catastrophic," <a href=>he told me once</a>. And while the overlap between Avanex's and Bookham's activities isn't 100%, there would be a lot of redundancy in amplifiers and transceivers, as well as in the markets served. (Couder also discussed his M&A philosophy during a <a href=>video interview at OFC/NFOEC</a>.)<br /><br />So if the Bookham/Avanex marriage hasn't happened yet, it may be because Bookham is looking for something else in a partner.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/11/avanex-preparing-for-boarders-.html2008-11-26T16:17:00.000ZGemfire restartsnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Gemfire CEO Rick Tompane reports that his company is back up and running. With the Thanksgiving holiday this week, he plans to bring "the people associated with the fab and equipment" back first. That's likely good news for the folks in Livingston, Scotland.<br /><br />Tompane credits cooperation from his customers, who have either accelerated payment terms or increased their orders. Despite discussing the possibility earlier (see post below), Tompane says his customer did not participate in a bridge loan.<br /><br />"The bridge was an advance on the financing to fund additional growth since we had doubled our revenue last year and planned to double again this year," he explained i an email. "In this market, both additional financing and growth is uncertain so we have shifted out some new programs and focused on current customers who have been kind enough to increase orders and accelerate payments so that we can keep going."http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/11/gemfire-restarts.html2008-11-24T21:59:00.000ZGemfire 'on vacation' after new funding falls throughnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />PLC-based optical components and technology supplier <a href=>Gemfire</a> has asked its employees to go on a two-week vacation while it attempts to replace the funds a skittish investor decided not to provide.<br /><br />According to company CEO Rick Tompane, a new investor that had been lined up before the stock market crash decided to pull back its offer. Meanwhile, a couple of large customers pushed out their receivables. The combination of the two events has led the company to suspend operations for two weeks while it looks for a new source of funding.<br /><br />Tompane says that in informing his customers of the situation, several offered to participate in a bridge loan that would get Gemfire through its current difficulties. He described himself as "pretty optimistic" that such a customer-generated loan will be successfully put in place sometime next week, which is when the two-week suspension is scheduled to end.<br /><br />Meanwhile, Tompane revealed that competitors had approached him about acquiring Gemfire. He said that no active discussions along these lines are currently underway, although that might change if the bridge loan fell apart or didn't meet the company's long-term needs.<br /><br />As for the company's current investors, Tompane says they are not currently part of the bridge loan effort. He declined to speculate whether they would be willing to make up the difference if the bridge loan fell short of requirements.<br /><br />The current hiatus caused considerable uproar in Scotland, where Gemfire has a plant in Livingston (see <a href=>a BBC story here</a> and a story from <a href=>the <i>Daily Record</i> here</a>). Tompane says that UK labor law doesn't allow the "two week vacation" option; the only way to temporarily suspend operations is to shut down the plant, then attempt to recall the workers later. Gemfire would reopen the facility should it succeed in gaining the additional funds.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/11/gemfire-on-vacation-after-new-funding-falls-through.html2008-11-07T18:47:00.000ZThe scoop on Gudmundsonnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Here is JDSU's official take on Gudmundson's move:<br /><br /><i>The JDSU Optical Communications and Lasers segments are being combined into one segment called Communications and Commercial Optical Products (CCOP). This combination will enable JDSU to leverage its technology, its manufacturing model and its people to continue to improve profitability. Alan Lowe will assume the role of President of the Communications and Commercial Optical Products segment. David Gudmundson, who has driven the strategy and positive change for Optical Communications, will serve in an advisory role.</i><br /><br />Lowe previously headed up the Commercial Lasers business.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/10/the-scoop-on-gudmundson.html2008-10-31T21:54:00.000ZMore changes at JDSUnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />While attention has focused on <a href=>President and CEO Kevin Kennedy's imminent departure from JDSU</a>, he's not the only example of high-level personnel shuffling at the company.<br /><br />According to the same October 30 8K filing that details Kennedy's news, Executive Vice President and President, Optical Communications Products Group David Gudmundson stepped down from his position as head of the company's optical comms business to become vice president, senior advisor, optical technologies on October 28.<br /><br />I've got a query into JDSU to find out more.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/10/more-changes-at-jdsu.html2008-10-31T19:54:00.000ZWavelength debate makes RFoG standard unclearnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />As you'll see in an article that will appear in the November issue of Lightwave, the <a href=>working group within SCTE</a> that is shaping the <a href=>RF over Glass (RFoG) standards</a> has hit a stumbling block. The group has reached consensus on 1550 nm as the downstream wavelength. However, determining which upstream wavelengths to use has become an issue.<br /><br />The problem is that the working group, which contains several equipment vendors that also make PON systems, is attempting to construct the specifications in such a way that carriers can overlay a PON over the RFoG infrastructure. Given the wavelengths already in play for GPON and EPON, plus those expected to be used for the 10-Gbit/sec standards now under development within FSAN and the IEEE, there aren't a lot of attractive wavelengths left. And if you can't leverage the volumes that lasers tuned to the already popular wavelengths enjoy, how can you keep costs down?<br /><br />At the heart of the conundrum is the question of the role RFoG will play on the path toward all-optical MSO networks. One source for my article, whose company currently supplies RFoG-like equipment, asserts that the PON suppliers are pushing an agenda in which RFoG is merely an interim step toward PONs. The source disagrees with this philosophy; his feeling is that RFoG architectures will have a long life within MSO networks, and that the SCTE should therefore focus on what's the most cost-effective way to deploy them, PONs be damned.<br /><br />Despite this hiccough, consensus indicates that the SCTE working group will complete its task by the middle of next year. But it would appear that carriers could play a very useful role by communicating their viewpoints on the expected relationship (or lack of it) between RFoG and PONs.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/10/wavelength-debate-makes-rfog-standard-unclear.html2008-10-22T15:19:00.000ZThe 100G conundrumnoemail@noemail.orgMeghan Fuller HannaPosted by <a>Meghan Fuller Hanna</a><br /><br />This week, I've been trying to wrap the ol' cerebral cortex around dual-polarization quadrature phase-shift keying, or DP-QPSK to those of us in the know. (Since it took me most of the week just to memorize the acronym, I probably shouldn't count myself among that esteemed company.)<br /><br />At any rate, I started my research with the <a href=>Optical Internetworking Forum</a>, which recently announced that its 100G project would focus on DP-QPSK as the advanced modulation format for 100G long-haul transmission. When I asked Joe Berthold, former president of the OIF and one of the authors of the current 100G project, why the organization has decided to standardize on DP-QPSK, he was quick to note that this is <b>not</b> a standard, per se. <br /><br />Instead, the OIF's goal is to "create a critical mass of vendors in the market to buy technology components of a particular type so that we can create an ecosystem," he explained. "We are not looking at standardizing the be-all-and-end-all. We're looking at selecting a particular implementation that looks attractive enough that we think a lot of us are going to follow [it], and we think it makes a lot of sense for component companies to invest in the various piece-parts for it."<br /><br />In other words, the OIF is aiming to 1) mitigate the risk inherent in component R&D, and 2) drive down the cost of the resultant components by creating a critical mass of buyers. <br /><br />On the one hand, I applaud the OIF for trying to ease the financial burden on the component vendors by creating an ecosystem of buyers. In an article I wrote summarizing the mood at this year's OFC Conference, <a href=>"No relief in sight for optical components sector,"</a> I noted that the component sector was still too fragmented, gross margins remained tight at 20% to 30%, and no one seemed to know which part of the telecom food chain should bear the brunt of the R&D burden going forward. In that environment, what impetus does a component vendor have to develop the high-speed electronics and integrated photonic components that DP-QPSK will require? These companies poured hundreds of millions of R&D dollars into 40G, and . . . well, we all know what happened there. <br /><br />But here's the other side of this particular conundrum: While Berthold says the OIF was perfectly willing to "back another horse" should someone come up with a viable alternative to DP-QPSK, some wonder whether it's simply too early for the OIF to be backing <b><i>any</b></i> horse in this particular race. <br /><br />DP-QPSK still presents formidable challenges. For real 100G transmission, digital signal processing should be at least two times faster than for 40G, and the industry just isn't there yet. And it will take far more photonic integration than is currently available. You can build a system out of discrete components, but it's cost-prohibitive. Speaking of which, several people have told me that it's still at least--AT LEAST--six times more expensive than currently available 40G options, which, frankly, isn't going to fly with any of the carriers I know. <br /><br />To be fair, no one I interviewed for my story disputed the fact that DP-QPSK holds a lot of promise, but some wonder whether we can conclusively say it holds the <b>most</b> promise at this point. <br /><br />Because here's another interesting thing about DP-QPSK: it's relatively immature. As one industry insider told me, there is usually a three- to five-year lag before research concepts are commercialized, but "this is like lab-to-the-field almost immediately." There are still a lot of folks in research labs and university settings working on variations of APSK (amplitude and phase shift keying), for example, and there seems to be a recent groundswell of interest in OFDM (optical frequency domain multiplexing). <br /><br />So what do you think? Is now the right time for the industry to rally around DP-QPSK, or has the OIF jumped the gun? <br /><br />(FYI: My article--tentatively titled "Is DP-QPSK the end-game for 100 Gbits/sec?"-- is slated to appear in <i>Lightwave</i>'s November issue.)http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/10/the-100g-conundrum.html2008-10-16T15:52:00.000ZWhere FTTH falls shortnoemail@noemail.orgStephen HardyPosted by <a>Tim Pritchard, publisher, Lightwave</a><br /><br /><i>Editor's Note: Our boss, Tim Pritchard, asked for "equal access" to the blog to get something off his chest that's been bugging him about his current FTTH experience.</i><br /><br />When I first thought of entering the world of blogging, I was getting myself psyched up for yanking on my pant belt to let loose a good, "Why you, I, or somebody oughtta…!!!" But cooler headwinds prevailed.<br /><br />As both Group Publisher of LIGHTWAVE and an FTTH customer, my take on fiber service is different than an average customer. I see everything through the lens of our industry's work. Are we getting the promise of fiber out of our fiber-optic works? So far I say no.<br /><br />At the heart of what it seems to me to be a missed opportunity that lies between what <i>is</i> happening in the fiber roll out to the home and what <i>could</i> happen is the fact that wireline service providers focus on getting fiber service to the NCU -- and then stop there.<br /><br />If the wireline side of telecom looked at customers the way their wireless counterparts do -- then they would forever have their eyes on the device and the end user, not just the backbone.<br /><br />I have FTTH. At my NCU the Cat5 runs to a router that has an Ethernet attachment for my desktop computer and a wireless interface for the laptops in my home. Also out of my NCU, my telecom provider repurposed the previous MSO's coax between the NCU and my set top boxes that serve the televisions in my home.<br /><br />Consider the opportunity missed. Verizon, Cingular/ATT, and other wireless providers partner with Motorola, Nokia, and other handheld device makers and in doing so not only continually drive next-generation cell phone activity -- always coming out with and promoting next-generation devices which in turn drive more applications and revenue for the wireless phone companies -- but also drive brand as the carriers' brand footprint is all over the applications (see Verizon VCast as an example).<br /><br />What if Verizon and other wireline carriers were to partner with Sony, Sanyo, RCA, Dell, Gateway, and other wired device makers in the same way their wireless counterparts partner with consumer wireless device makers?<br /><br />Imagine now an FTTH customer experience that brings fiber all the way to the device. Instead of supporting 56-kbit/sec service on my laptops and scrambled, latency-affected viewing on my TV, I would get something completely out of the ordinary. Awesome HDTV and lightning fast Internet connectivity coupled with interfaces and applications that the bright engineers at the consumer device makers can and would dream up.<br /><br />I think that for the full reality of the fiber to the home dream -- of what we in this industry know we can build -- to come true, smart carriers need to find consumer device makers with which to partner. Customers like me would gladly add $20 to our monthly bill for a device (or devices) integrated into our packages that would work with extraordinary efficiency and a full menu of otherwise unattainable applications.<br /><br />Hey, the phone companies may even get back to a model where they simply lease the devices (like the 1980's AT&T phone models) -- only this go around they would offer regular upgrades for everyone's benefit.<br /><br />I am one FTTH customer that would sign onto that type of integrated service with a leased device model. How about you?http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/10/where-ftth-falls-short.html2008-10-06T20:48:00.000ZWDM-PON: Is it better suited to the European marketplace?noemail@noemail.orgMeghan Fuller HannaPosted by <a>Meghan Fuller Hanna</a><br /><br />At ECOC last week, France Telecom spoke publicly about its participation in the EU's SARDANA research initiative and the cost savings it hopes to achieve with the resultant technology. The carrier plans to evaluate SARDANA's WDM-PON during a trial in the town of Lannion, France in 2010. <br /><br />The EU has also funded a WDM-PON initiative known as GigaWaM, championed by the likes of Ignis Photonyx (See <a href=>"EU to fund Ignis-led WDM-PON project</a>.")<br /><br />And I couldn't help but notice that the first publicly announced customer for LG-Nortel's WDM-PON-based Ethernet Access system is a Dutch broadband service provider (See <a href=>"UNET deploys Nortel's WDM-PON based system</a>.")<br /><br />All of which makes me wonder: Is WDM-PON better suited to the European marketplace? Or are the Europeans simply going to be the early adopters of the technology, much like Japan led the way with its early deployments of PON technology? <br /><br />I recently interviewed Giovanni Manto, leader of Nortel's Ethernet Fiber Access Solutions Division, on the subject of WDM-PON. He noted that in North America, Verizon has settled on GPON, and in Asia, NTT has standardized on GE-PON. But in Europe, there is no such "900-lb. gorilla," as he called them. <br /><br />"A lot of the EMEA customers we deal with historically are in two camps right now," Manto told me. "One is basically saying, 'We want to build point-to-point networks. We believe point-to-point networks are scalable networks, they are foundational, and we can change the personality of that particular fiber connection depending on what our customers want.' Some of them have actually stipulated publicly that GPON will never be deployed in their network because of all the encumbrances and issues that GPON brings to their network, specifically from an operations standpoint and from a scalability standpoint."<br /><br />"Then there are customers in Europe that are looking at GPON," he admitted, "but they are reevaluating their decisions based on the fact that now there’s another solution in town."<br /><br />While there is certainly a great deal of PON and active Ethernet already deployed in Europe, there isn't a clear-cut front-runner, and that makes me wonder if European operators will be more likely to consider WDM-PON versus some version of 10G PON going forward. <br /><br />What do you think? Feel free to respond to this post or drop me an email (<a></a>) with your thoughts on the subject. I'm thinking about writing a follow-up article that tackles this very question. <br /> <br /><br />. . . . Speaking of tackling difficult questions, some of you have asked for my predictions now that baseball's post-season is upon us. Keeping in mind that I am a Red Sox fan and not selecting them would be sacrilegious no matter how formidable the opponent, I offer you the following picks for the Division Series: <br /><br />National League Division Series:<br /><b>Cubs over Dodgers in five games.</b> As a Red Sox fan, this is a tough one. I mean, really tough. For obvious reasons, a Red Sox/Dodgers World Series would be compelling, thanks to the return of former Sox Manny-being-Manny, Nomah, and Derek Lowe. Plus, I'd love to see Terry Francona and Joe Torre managing head-to-head in the Fall Classic. That said, not rooting for the Cubs seems a little self-centered, what with the Red Sox winning two titles in the last four years and the Cubs in the midst of a hundred-year drought. So I have to go with the Cubs on this one. Besides, now we can safely see a Red Sox/Cubs World Series without worrying about the impending apocalypse.<br /><br /><b>Phillies over Brewers in four games.</b> CC Sabathia is no doubt going to register at least one W, but I still think the Phillies will take this series. <br /><br />American League Division Series:<br /><b>Tampa Bay Rays sweep Chicago.</b> Let's face it: We can no longer question whether these Rays are for real, and I think they're going to mop the floor with the White Sox. (For the record, I think they would have mopped the floor with the Twins, too.)<br /><br />And, without further ado . . . <br /><br /><b>Red Sox over the Angels in five games.</b> Okay, so in my heart I know this will be a tough series. On paper, it looks like it could be the Angels' year, but here's the great thing about baseball: It's not played on paper. As Stephen reminded me this morning, all we have to do is take one of the two games in Anaheim, and then we're back at Fenway where this team simply plays great baseball. The bottom line: I &#9829 Jon Lester, I think he's going to beat Lackey tonight, and I'd take my chances against anyone at the Fens.<br /><br />I'll check in with my Championship Series predictions next week.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/10/wdm-pon-is-it-better-suited-to-the-european-marketplace-.html2008-10-01T16:19:00.000ZGlasvezelNet Amsterdam trials 1-Gbit/sec to the homenoemail@noemail.orgMeghan Fuller HannaPosted by <a href=>Meghan Fuller Hanna</a><br /><br /><a href=>GlasvezelNet Amsterdam</a> (GNA), a driving force behind Amsterdam's 40,000-home FTTH initiative <a href=>CityNet</a>, in conjunction with BBned and InterNLnet, recently conducted a trial to demonstrate the feasibility of delivering 1-Gbit/sec symmetrical service to the home. <br /><br />Yesterday, GNA posted a video to YouTube entitled "1000/1000 Mbps FttH test Amsterdam." The images are somewhat grainy, and there is no sound, but I still think it's cool that they YouTubed their results. <br /><br /><object width="425" height="344"><param name="movie" value=""></param><param name="allowFullScreen" value="true"></param><embed src="" type="application/x-shockwave-flash" allowfullscreen="true" width="425" height="344"></embed></object>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/09/glasvezelnet-amsterdam-trials-1-gbit-sec-to-the-home.html2008-09-12T20:46:00.000ZIt's the price, right? Wrong.noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />We continually hear component and subsystem vendors complain about how often their customers grind them down on price. They can't maintain decent margins when systems houses refuse to pay a reasonable price for their wares or when their competitors agree to ship a few dollars out the door with each device they sell. It's all about how low your price is, one would conclude.<br /><br />In our upcoming September issue, we report the results of our second annual survey of Lightwave readers who buy transceivers and transponders. The survey is designed to uncover the factors that go into the purchase decision, including attributes of the products themselves as well as the vendors who supply them. And, for the second year in a row, our readers tell us that price isn't the most important factor when it comes to choosing a transceiver or transponder. Reliability is #1, with performance second and cost coming in third.<br /><br />So on the one hand, we hear constant grumbling from the component and subsystem vendors about price; on the other, we have their customers saying that price isn't the first, or even the second, most important factor in choosing a device for their application. What gives?<br /><br />There could be several factors at play. First, engineers may choose the device, and someone else in the company may choose how much they're going to pay for it. Second, there may be even fewer points of differentiation among vendors when it comes to reliability and performance than one would think, meaning that the only way they can compete is on price. That's the way a commoditized market works -- and heaven help the industry if that's the case more often than not.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/09/it-s-the-price-right-wrong-.html2008-09-11T20:36:00.000ZDraka makes fiber-to-the-houseboat a realitynoemail@noemail.orgMeghan Fuller HannaPosted by <a>Meghan Fuller Hanna</a><br /><br />Several days ago, I received an email from <a href=>Draka</a> touting "the world's first fiber to the houseboat." It seems the company has developed a new type of optical connector that allows houseboat owners to physically connect to Amsterdam's <a href=>CityNet</a> fiber-optic network upon mooring and disconnect whenever a trip is necessary.<br /><br />Intrigued, I immediately asked for more information. <br /><br />Unfortunately, the lead engineer on the project is currently on vacation, but I've been in email communication with a Draka spokesperson, who provided a few additional details. As part of the Amsterdam CityNet broadband project, Draka was challenged to develop a connector to meet the unique requirements of FTTH customers living on houseboats. <br /><br />Not an easy task, as the Draka spokesperson noted. These houseboats are mobile; now and then, they go sailing on the Ijsselmeer, and sometimes they must sail to shipyards for maintenance. <br /><br />(For the record, the Netherlands Board of Tourism & Conventions (NBTC) reports that there are some 2,500 houseboats docked in the city of Amsterdam.) <br /><br />A standard, fixed optical fiber connection for houseboat applications was obviously out of the question. Instead, a fiber-to-the-houseboat application requires an optical connection that is robust enough for the houseboat owners themselves to connect and disconnect. Moreover, standard connectors are sensitive to dirt and dust, so a fiber-to-the-houseboat-optimized optical connector would have to be easy to clean and dry. (Draka tells me the connector it has developed can be dropped into the water.)<br /><br />Writes the Draka spokesperson, "Trials proved that with minor modifications, a very robust beam connector originally developed for military applications is ideally suited for houseboat applications. It can be cleaned easily and has good transmission characteristics."<br /><br />To underscore the viability of the new connector, Draka released this photo of Oliver Ax, proud owner of the world's first fiber-connected houseboat. <br /><br /><a href=""><img style="display:block; margin:0px auto 10px; text-align:center;cursor:pointer; cursor:hand;" src="" border="0" alt="" /></a><br /><br />"I now have ultra-fast Internet, TV, and telephone connection through one single cable," reports Ax, who says he has always been interested in technology and is delighted with the new services provided by GNA (Glasvezelnet Amsterdam) and local Internet provider Alice. <br /><br />Fiber-to-the-houseboat. Kinda makes you wonder what's next. Fiber-to-the-RV? What about fiber-to-the-car? Maybe some day, we'll be able to plug into a fiber connection while we're waiting for our electric cars to charge. <br /><br />In the meantime, anyone else have any cool "Fiber to the . . . ." stories?http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/09/draka-makes-fiber-to-the-houseboat-a-reality.html2008-09-05T21:25:00.000ZA new look for Lightwavenoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Those of you who receive Lightwave magazine may be interested to learn that we're in the process of redesigning it. This will be our first major face-lift since 2004.<br /><br />The redesign will focus particularly on <a href=>the digital edition of the publication</a>. (The link is to the current version, by the way -- we're still working on the new one.) That's because we think that we have the opportunity to marry the best of print and online within the electronic version of the magazine. From the print world, we'll keep the attractive, full-page layout of articles that makes them easy to read. To this end, one thing we'll do within the redesign is minimize the number of times articles jump from one part of the magazine to another. This will make the articles easier to follow -- and easier to print out and take with you, if you'd still prefer to read them in paper format.<br /><br />But we'll also leverage the capabilities of a digital format. We'll add links at the end of our articles that will make additional information resources available to you with a single click of your mouse. We'll also add video and animation capabilities. For example, we hope to add brief "newscasts" to <a href=>the "Update" section</a> that will fill you in on the important events that occurred between the time our deadlines forced us stop writing and the time you receive your digital edition. We also hope to be able to replace some static photos and figures with animation that will provide a clearer understanding of whatever point we're trying to illustrate. (<a href=>This animation</a> JDSU supplied for our <a href=>Optical Equipment Design Center</a> might be an example of something we'd use in our "<a href=>Product Profiles</a>.") Naturally, we'll also look to include <a href=>some of the video</a> and other multimedia features you can find right now on our website.<br /><br />In short, our goal is make Lightwave an even more useful tool for you than it is now. If you're getting the print edition, you'll still benefit from the redesign -- but I urge you to consider switching to the digital edition to get the full benefits of our beefed up, multimedia coverage of the optical communications market.<br /><br />We plan to unveil the new-look Lightwave this January. That means there's time for you to offer suggestions on how we can shape the publication to better meet your needs. Have an idea? Email me using the link above or offer a comment below. We'd love to hear from you.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/08/a-new-look-for-lightwave.html2008-08-29T15:12:00.000ZWDM-PON chatternoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br /><a href=>LG Nortel's recent acquisition of Novera Optics</a> has WDM-PON back in the news -- and a continuing subject of debate. The latest to weigh in is Lynn Hutcheson, VP and analyst at Ovum.<br /><br />"It most definitely is a good deal for Novera as this acquisition couldn't have come at a better time, with the company approaching an end to its funding and with no significant customer orders in sight. Nortel, on the other hand, abandoned the access market in 2000 as the downturn in the telecommunications market was just beginning. This will be a small step back into the broadband access market," Hutcheson comments. "At the Executive Forum as part of the Optical Fiber Communications Conference, Michael Adams, Nortel's VP of strategy and architecture for Metro Ethernet Solutions, said that WDM-PON was going to play a major role in its broadband access strategy. At best it will provide a few bragging rights for Nortel as it will be the only broadband access player with a WDM-PON product, but that is all as the market for this product is still a few years away."<br /><br />Despite labeling Nortel as "the only broadband access player with a WDM-PON product," Hutchinson does note that Novera has a strategic marketing agreement with ADC. The latter used NXTcomm to debut the results of that agreement, the PONy Express. Meghan talked to Tom Devittorio, product line manager at ADC, about the platform, and you can see the interview right now:<br /><br /><embed src="" bgcolor="#FFFFFF" flashVars="playerId=1726686677&viewerSecureGatewayURL=" base="" name="flashObj" width="486" height="412" seamlesstabbing="false" type="application/x-shockwave-flash" swLiveConnect="true" pluginspage=""></embed><br />Of course, <a href=>we just posted a video on next-gen PON as well</a> that includes a discussion of WDM-PON. You'll note that ADVA Optical Networking is offering CWDM-PON as a way to support business services.<br /><br />I agree with Hitachi America's Scott Wilkinson (as captured in the NG PON video) that WDM-PON's short-term future for FTTH outside of South Korea will largely depend on how long it takes 10G PON technology to reach the marketplace. However, I could also envision some pull from carriers (MSOs?) who might use it to support business services. In either case, however, I don't see the technology sweeping the marketplace.<br /><br />What do you think?http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/08/wdm-pon-chatter.html2008-08-13T17:17:00.000ZNG PON synergiesnoemail@noemail.orgMeghan Fuller HannaPosted by <a>Meghan Fuller Hanna</a><br /><br />I've spoken with several people lately about the likelihood of the IEEE and ITU finding common ground in the development of their separate 10-Gbit PON specifications. The two standards bodies met last month in Geneva to discuss next-generation optical access and explore any potential synergies that may exist between future incarnations of EPON and GPON. <br /><br />My sources tell me that the standards bodies have not made a lot of progress, and the current status is "not real rosy," as one industry insider put it. Differences at the TC/MAC layer, in particular, may be difficult to reconcile. That said, some people believe it may be possible to find some synergies at the optical layer, which strikes me as very good news.<br /><br />I understand there are numerous technical hurdles to overcome in order to achieve these synergies, but imagine the economies of scale that could be derived from, say, a common laser specification for both 10-Gbit GPON and EPON. The laser is, after all, the most expensive part of the ONU. <br /><br />For a lot of folks, EPON versus GPON is something of a religious debate, but if we could lower the price of 10G PON optics by doubling the volume, it may well be worth it to find some common ground. Even if the two standards continued to be developed separately but shared some key optical building blocks, it seems to me a win-win for both camps. <br /><br />What do you think?http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/07/ng-pon-synergies.html2008-07-30T20:07:00.000ZNew tunable pluggable MSA?noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br /><a href=>Last week's announcement</a> that Ciena will design in a tunable pluggable transceiver from Bookham that doesn't conform to current multisource agreements (MSAs) -- and apparently expects to have a second source deliver similar modules -- raises the question of whether there is going to be a demand for a new MSA. Such an form factor would serve as an interim step down in size between, say, an X2 and an XFP-E.<br /><br />Why would the industry need such an MSA? Clearly, Ciena wanted to deliver the benefits of tunable pluggables now and couldn't find a supplier capable of meeting its requirements in an XFP-E or smaller form factor. If Ciena is willing to go with a custom size, will its competitors do the same to avoid falling behind?<br /><br />Yves LeMaitre, VP of telecom sales at Bookham, says there currently aren't MSA discussions focused on his company's device. "At the same time, you are absolutely right about the need for an MSA in the coming two years," he wrote in an email when I posed the question to him. "The value of a tunable-pluggable solution is evident and we believe that the adoption rate will be extremely high amongst optical systems manufacturers."<br /><br />Unless someone comes up with a tunable X2, XFP, or XFP-E soon, the pressure for a new MSA will rise. The fact that there will be at least two companies making "Ciena-sized" modules might move such an MSA into the development fast lane.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/07/new-tunable-pluggable-msa-.html2008-07-24T18:56:00.000ZCisco/Comcast 100G opticsnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />The folks at StrataLight Communications followed up <a href=>the Cisco/Comcast 100G demo announcement</a> by confirming that they had provided the optics. But according to Ross Saunders, general manager of next-generation transport at StrataLight, the company isn't focusing on productizing the interface used in the demo in the short term.<br /><br />Saunders wouldn't provide many details (such as modulation format) of the "DWDM optics" his company supplied for the demo, other than to confirm that it involved a single wavelength running at 100G. The interface did not use <a href=>the 100G chipset the company recently announced</a>, either.<br /><br />As the IC announcement implied and Saunders confirmed, StrataLight is focusing its 100G attention in the near term on a muxponder product that could be used to aggregate 10G and 40G wavelengths as an interim step toward serial 100G transmission. Saunders says the muxponder or "multiplexing transponder" is slated to be delivered to an OEM and a carrier for trials within the next month. When and if the prototype evolves into a commercially available product will hinge largely on the reaction it generates during these trials.<br /><br />Meanwhile, the company <a href=>continues its work in the 40G space</a>.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/06/cisco-comcast-100g-optics.html2008-06-30T16:47:00.000ZThe buzz from Cable Tecnoemail@noemail.orgMeghan Fuller HannaPosted by <a>Meghan Fuller Hanna</a><br /><br />Lots to talk about from the Cable Tec Expo, but the biggest question seems to be around FTTH. Will they? Won't they? What is the real market opportunity for PON vendors? <br /><br />I attended a session early yesterday morning during which independent consultant Victor Blake gave a presentation entitiled, "Cable's Competitive Response to Verizon's FiOS." He began by asking for a show of hands: How many believe FiOS represents a competitive threat? From my vantage point, it looked like the majority of hands went up. <br /><br />When Blake recapped Verizon's future plans--eventually migrating to 40G upstream / 10G downstream--you could hear murmurs from the crowd. Assuming Verizon doesn't change its split ratios, that's 40G x 32 subscribers, which yields more than 1 Gbit per subscriber. And when you add a WDM-PON overlay, said Blake, it gets even worse--for the cable competitor, that is. There seems to be a growing awareness that their existing architecture just isn't going to cut it. <br /><br />Blake believes that FTTH is more capable than any current or foreseeable coax technology, and he had some interesting things to say about EPON in particular. (Look for more on that later.) That said, he also cautioned that the MSOs need to get the timing just right--too early and they tie up precious capital, too late and they may miss the opportunity to compete. "It's not a crisis now," he said, "but we can't wait until it becomes a crisis either." Just when the timing will be right remains a subject for debate.<br /><br />The PON vendors--including Tellabs, Alcatel-Lucent, Hitachi Telecom, Calix, Alloptic, Enablence, and Salira--are all here this week, many touting RFoG or some version of DOCSIS-over-PON. The consensus seems to be that the MSOs will deploy PON for commercial services first. In fact, both Calix and Motorola announced GPON wins this week, specifically for commercial services delivery. From there, the MSOs may begin to deploy PON for residential services in Greenfields, but it will likely be some time before we see them overbuilding their existing architectures. <br /><br />In fact, if the MSOs do decided to make a wholesale change and deploy PON in overbuild situations, the PON architectures they are most likely to deploy probably won't look much like what's available today. It may be something developed by CableLabs. It may be derived from EPON or GPON. It may be some version of 10G PON. At this point, it's anyone's guess.<br /><br />That said, almost everyone I spoke with at the show believes the MSOs are serious about PON. Said one source: "The interesting thing about the cable industry and PON is that PON is actually well suited to the MSOs. It's point-to-multipoint. It uses the same wavelengths that the cable industry is used to worrying about. In fact, talking to the telcos about PON is sometimes more difficult than talking to the MSOs." <br /><br />I plan to write an article for Lightwave's August issue about these trends and others, including some discussion about RFoG as well as the need to integrate with DOCSIS provisioning systems. Keep your eye out for that.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/06/the-buzz-from-cable-tec.html2008-06-27T14:09:00.000ZCable Tec 2008: A fact-finding missionnoemail@noemail.orgMeghan Fuller HannaPosted by <a>Meghan Fuller Hanna</a><br /><br />I'm leaving in about two hours for the SCTE's Cable Tec Expo, where I hope to find the answers to several questions: <br /><br /><br />*How serious are the Cable MSOs about FTTH technology?<br /><br />*Most of the MSOs currently deploying PON technologies are using them to support commercial services. When, if ever, can we expect to see MSO deploying PON for residential services? <br /><br />*What challenges will they face in migrating to FTTH?<br /><br />*How is the SCTE's RF-over-glass (RFoG) standard progressing?<br /><br />*Is DOCSIS 3.0 an interim fix, a stop-gap measure on the way to FTTH, or will it provide enough bandwidth to allow the MSOs to compete with the likes of Verizon's FiOS and AT&T's U-verse? <br /><br />*If the MSOs deploy "telco-grade" PON, do they care about GPON versus EPON? Is WDM-PON on their radar screen?<br /><br /><br />Last week, I spoke with Jeff Stribling, vice president of marketing and customer service at <a href=>Salira Systems</a> about the vendor's multi-wavelength PON (MW-PON) and DOCSIS-over-PON (DePON) technologies. (Watch for more on this later.) He told me that Salira sees "a major opportunity, right now, today, in commercial services"--so much so that Salira recently announced its strategic focus on the MSO market. <br /><br />Stribling says the MSOs' existing HFC infrastructure is, in many cases, insufficient for supporting the kinds of high-bandwidth services that the MSOs must deliver to compete with the likes of FiOS. Traditionally, they have deployed point-to-point CWDM to overcome the bandwidth limitations of HFC, but Salira advocates the use of PON--specifically EPON--as a cost-effective, higher bandwidth alternative that also leverages the ubiquity of Ethernet in their backbone networks. <br /><br />I suspect other PON vendors will put forth similar arguments; we'll see what the next few days bring.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/06/cable-tec-2008-a-fact-finding-mission.html2008-06-25T13:06:00.000ZNXTcomm Notesnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Quickly reviewing my reporter's notebook while I chomp on a bagel in the press room Thursday morning, here's what I find:<br /><br />Net Insight, which has ridden its DTM technology to success in the video broadcast market, hopes to make a bigger name for itself in North America through support of telepresence and multiservice delivery...Mintera execs Terry Unter (pres and CEO) and Niall Robinson (VP, prod marketing) claim that they've shipped their 40G transponders to "seven or eight" customers, five of whom have designed the module into a system -- and, in some cases, more than one...Speaking of 40G, rumor has it that the "high speed" company Opnext currently is looking at most closely for acquisition (see blog post below) is StrataLight Communications...Having speculated in the same blog post that it might make sense for Opnext to look at CoreOptics, I asked Saeid Aramideh, VP of global sales, marketing and business Development, about the possibility. He said his company sees itself as a consolidator, not a consolidatee -- and hinted that CoreOptics might prove that in the reasonably near future...Everyone is talking about having an RF return path capability on their PON or other FTTX product to address the MSO market as well as other carriers who want to add on-demand services without abandoning their current RF infrastructure. EPON vendor Alloptic claims that at least some of the GPON vendors touting this capability are using their technology -- which they refuse to sell to other EPON vendors...For every vendor touting WDM-PON at the show (and there are several), there's another saying that there probably isn't much of a market for it between now and the time more affordable 10G PON technology will be available...While the <a href=>OIF has given its blessing to dual-polarization QPSK for 100G transport</a>, neither Hitachi Telecom nor Adva Optical Networking think that's the right way to go. They don't seem too concerned about the OIF's current direction. If AT&T or Verizon announced they'd prefer a different modulation format, the OIF would undoubtedly pull an about-face, one source speculated.<br /><br />More to come later...http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/06/nxtcomm-notes.html2008-06-19T16:56:00.000ZThe most interesting thing so farnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />NXTcomm is in full swing, with the buzz surrounding high-speed networking and FTTH, particularly WDM-PON, attaining the anticipated high pitch. The most interesting thing I've heard about so far, in fact, pertains to WDM-PON, specifically Tellabs' <a href=>upcoming work for the SARDANA program</a> funded by the European Commission.<br /><br />SARDANA stands for "Scalable Advance Ring-based passive Dense Access Network Architecture." As the name implies, the goal of the three-year program is to develop a ring-based WDM-PON architecture with a reach of around 100 km. (You can read more about it <a href=>at the project's website</a>.) The key to the extended reach is including amplification in the network. According to Tellabs, they plan to base the amplification on EDFA technology. But how do you put EDFAs, which require power for the pump laser, into an passive network infrastructure? You "disintegrate" it, Tellabs says; you put the pump laser function in the central office, where you already have power, shoot the pump laser light down the network on one of the PON wavelengths, and have passive amplification units that leverage the pump to amplify the surrounding wavelengths.<br /><br />If it works, Tellabs thinks they can find more applications for it beyond SARDANA. I bet they can too.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/06/the-most-interesting-thing-so-far.html2008-06-18T17:39:00.000ZVCs: Consolidation effects may not be significantnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />While researching an article I'm doing for our July issue on the current funding environment, I've been talking to VCs about whether their community's ongoing lack of enthusiasm for the optical communications space (when it comes to photonics, it's all about green technology, photovoltaics, and sold-state lighting these days, they say) and the hostile IPO environment might finally drive consolidation, particularly in the components and subsystems space. It might, they say -- but some wonder whether the sector will see as much benefit as we've all assumed.<br /><br />That's because most of the observers whose opinions on consolidation we hear about are focused on Western companies. Yes, moves such as <a href=>the Finisar/Optium merger should be good for the industry</a>. Indeed, other well-known companies will be looking to M&A to increase their scale or broaden their product lines, taking smaller companies off the board.<br /><br />But while the number of Western companies may start shrinking, the number of companies in the East, particularly in China and India, continues to grow. They're looking to be <a href=,-Inc-files-for-IPO/>the next Fiberxon</a> or <a href=>Opnext</a>, I'm hearing.<br /><br />The trend may mean that the total number of companies in components and subsystems may not change much at all in the short term -- and the new entrants will be focusing on low-cost approaches that may continue to put margin pressure on the entire sector.<br /><br />One VC described the phenomenon quite apply, I thought: "Scary."http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/06/vcs-consolidation-effects-may-not-be-significant.html2008-06-13T18:47:00.000ZOpnext shopping for 40G/100G?noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />As reported by Morgan Keegan's Paul Bonenfant, Opnext President and CEO Harry Bosco yesterday morning told the audience at Cowen and Company's 20/20 TMT Conference in New York that the reason he hasn't executed a previously announced stock buy back is that he's "looking at different things" now that the acquisition price for companies has come down. In particular, he said, he's looking at high-end optics, specifically 40G and 100G module and technology companies. "Some are small and some are fairly significant," he said.<br /><br />Bonenfant identified Mintera and StrataLight as the two most prominent companies of interest. He believes a StrataLight acquisition is more likely, given the fact that <a href=$19-million/>JDSU has invested in Mintera</a>. The fact that StrataLight supplies technology to Nokia Siemens Networks, as does Opnext, doesn't hurt either.<br /><br />One company that Bonenfant didn't mention that Opnext also might be looking at is CoreOptics. The company is <a href=$25-million-funding-round/>already active in 40G development as well as coherent detection technology for 100G</a>. (See my May 15 blog posting below.) The fact that <a href=>the OIF has decided to build its 100G DWDM implentation agreement</a> around dual-polarization quadrature phase-shift keying with coherent detection indicates that CoreOptics is moving in the right direction for 100G.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/05/opnext-shopping-for-40g-100g-.html2008-05-30T15:11:00.000ZZeugma Systems conducts video consumption surveynoemail@noemail.orgMeghan Fuller HannaPosted by <a>Meghan Fuller Hanna</a><br /><br />Last week, I spoke with Kevin Walsh, vice president of marketing for startup Zeugma Systems, about the company's flagship product, the Zeugma Services Node. (See our coverage of the product launch <a href=>here</a>.)<br /><br />Basically, the ZSN is a service delivery router that enables service providers to identify and monitor each and every session flow traversing their networks, which, in turn, enables them to better manage and even customize those flows. <br /><br />One of the potential new sources of revenue that the ZSN's visibility would enable is something Walsh calls "content acceleration," which posits that subscribers would be willing to pay incremental fees in order to enjoy faster downloading of content, specifically video content. <br /><br />Walsh cites a June 2007 IDC study in which analysts forecast that almost 50% of traffic growth between now and 2012 will be attributable to over-the-top (OTT) video or Internet video services delivered by the likes of <a href=>Joost</a> or <a href=>Hulu</a>. "If you're a service provider," he says, "it's this over-the-top video that scares you the most, because it's the source of the greatest growth." <br /><br />To test its content acceleration theory, Zeugma Systems retained <a href=>IDC</a> to conduct a survey of 800 U.S. consumers this past March. Respondents were asked a number of questions related to their video consumption habits--and their willingness to pay for value-added services. <br /><br />"We asked first, would you even be interested in finding, buying, and watching video delivered over the Internet directly to your 50-inch flat-panel television using only your remote control?" Walsh recalls. "In other words, no computer is involved, and this is high-definition video suitable for what we call living-room-quality video."<br /><br />Thirty-five percent of the respondents said they were very interested, with an additional 41% of respondents indicating somewhat to moderate interest. The question played even more favorably among the 18-34 age group, with 54% of respondents expressing high interest and 41% indicating moderate interest. <br /><br />But now comes the tricky part: If a movie costs $5 to rent and takes several hours to download, would you pay more for a faster download so you could begin watching that video immediately? <br /><br />The survey revealed the following: <br />&#8226 38% of respondents expressed a moderate to high interest in paying an additional $0.25 for faster downloading capabilities, while an additional 18% said they'd be somewhat to moderately interested. <br />&#8226 14% said they would be moderately to highly interested in paying an extra $1.00 for faster service, while 26% said they would be somewhat to moderately interested. <br />&#8226 41% of respondents said they would be willing to watch a 30-second ad for the privilege of faster downloads.<br /><br />"The take-away from this is not how much subscribers would be willing to pay," says Walsh, "rather, that they are willing to pay." <br /><br />"A lot of service providers are sitting there right now scratching their heads, wondering what sorts of services they can roll out that will generate revenue," he continues. "Here's just one example of the many that consumers are saying they would be willing to pay for."<br /><br />So now I'm asking the Lightwave audience the same questions: Would you be interested in receiving video content over the Internet straight to your TV set? And would you be willing to pay more to download that video content faster?http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/05/zeugma-systems-conducts-video-consumption-survey.html2008-05-28T21:34:00.000Z3D tennis, anyone?noemail@noemail.orgMeghan Fuller HannaPosted by <a>Meghan Fuller Hanna</a><br /><br />An interesting press release crossed my desk this morning . . . France Telecom's <a href=>Orange</a> today announced that it will film and broadcast its first live 3D sports event at next week's French Open. <br /><br />The trial is set to take place next Monday (5/26) and Tuesday (5/27). Orange says it will use 3D cameras to film all the matches on the Suzanne Lenglen court and broadcast them live. The 3D matches will also be available on VoD until the end of the tournament. At Roland Garros and in its two flagship stores located at Champs Elysées and Paris Madeleine, Orange will be providing its guests with 3D glasses to watch the matches on 3D TV. <br /><br />The carrier says this trial is "preparing the arrival of 3D television in [its] customers' homes."<br /><br />While tennis is not really my proverbial cup of tea, 3D sports would be VERY MUCH my cup of tea, and I applaud Orange for upping the ante when it comes to sports broadcasting. How long before the North American telcos and cable MSOs try to use such technology to their competitive advantage? I may be biased, but I can't imagine a better use for all that fiber they've deployed. <br /><br />Last night's frustrating 103-97 Celtics loss (at home, no less!) notwithstanding, this would have been a great week to watch Boston sports in 3D. Jon Lester's no-hitter was spectacular enough on the flat screen; imagine watching it in three dimensions. Jacoby Ellsbury's diving catch in the fourth would have sent him sprawling into my living room. And how much more imposing would Paul Pierce and LeBron James have looked if last Sunday's nerve-wracking, chest-pain-inducing Game Seven, Celtics v. Cavaliers, had been broadcast in 3D?<br /><br />Speaking of the Celtics, Verizon announced yesterday that reigning NBA Defensive Player of the Year Kevin Garnett recently completed filming a TV commercial for its FiOS offering. (To view an MPG preview, click <a href=>here</a>.) <br /><br /><a href=""><img style="display:block; margin:0px auto 10px; text-align:center;cursor:pointer; cursor:hand;" src="" border="0" alt="" /></a><br /><br />Part of the carrier's "This is FiOS; This is Big" campaign, the ad showcases Verizon's Home Media DVR, which allows subscribers to record a standard-definition program and view it anytime on up to seven TVs in the home, not just the one hooked up to the DVR. <br /><br />On any other day, that would have been pretty cool. But today Orange announced that it's going to offer a live sporting event in 3D, and to the sports fanatic, that's a slam dunk.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/05/3d-tennis-anyone-.html2008-05-23T16:49:00.000ZFinisar/Optium, Part IInoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />So where was I? Oh yeah -- scale.<br /><br />According to Gertel, his customers were looking for Optium to supply a wider range of products, which the company wasn't set up to do efficiently. So merging with a large, complementary company like Finisar made sense.<br /><br />Both Rawls and Gertel say that these days, size definitely matters. "We think this industry needs a dominant player that has all the answers for the customer," Gertel says. "I think the customer wants a broad supply from a stable company that has the infrastructure and the critical mass to supply all the R&D that's needed to support them in the long run. We think that's what we've created today."<br /><br />The merged entity certainly will be big. According to information presented during an analysts' call this morning, the two companies' annual revenue (based on the last reported quarter, annualized) would be $612 million, greater than $544 million JDSU would be expected to garner from its optical communications segment (computed the same way). It would also have the best gross margin 35.3%, non-GAAP) and EBITDA margin (11.3%, also non-GAAP) in the industry. The company would also have the broadest product line.<br /><br />"All of our customers are big equipment companies, whether they're developing network equipment or telecom equipment. And the end users are all huge companies as well," Rawls concludes. "I think there will always be a place for little companies to get funded to develop a specific technology in a niche product that addresses some inflection point in the market. Those companies, though, if they're going to be successful with these huge customers, they're going to have to be bought or merged or something into some bigger entity. Because they won't be able to scale. It's very difficult for little companies to scale to satisfy the needs of these increasingly consolidated, large equipment companies."http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/05/finisar-optium-part-ii.html2008-05-16T22:45:00.000ZRawls, Gertel explain Finisar/Optium mergernoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Alright, so I got off the phone with Jerry Rawls, chairman, president, and CEO of Finisar, and Eitan Gertel, chairman and CEO of Optium, about 15 minutes ago -- and I have a newsletter to put out. So here are few choice quotes about <a href=>the merger announced today</a> that I promise to come back and flesh out later.<br /><br />First, the merger was Jerry's idea. "I was on a trip in Europe and I got an email from Jerry and we started from there," says Eitan.<br /><br />Here's why Jerry called:<br /><br />"One of the things about Optium was that they were in my view clearly the most attractive partner in this industry because their product lines were so complementary to us and because they were a profitable, successful company. And all of our customers, where we shared customers in common, our customers thought highly of Optium. And there are so many of the potential partners in our industry that you might try to acquire or merge with and either the customers didn't like them, the company was wounded, they weren't profitable, or you had this huge overlap in products that you would have to now rationalize -- or you would lose some of the business at customers because you were competitors, you were both qualified. And Optium was the one company in the industry where that didn't exist. So the combination for us was clearly the best in the industry."<br /><br />Rawls admitted that he was shopping around. This isn't surprising, given <a href=>what he told me just before OFC/NFOEC</a> about his desire to expand Finisar's offerings in the telecom space.<br /><br />From Eitan's view, the deal is all about scale. "Optium was running at a pretty good growth rate, but at some point I looked at this industry and I said that the industry was too fragmented -- there's no critical mass. If we keep growing, even at 30% per year, it's still going to take a few years to get to a critical mass to start making a difference in this industry. So Finisar for us was the perfect company to do this deal with, because there's no duplication of products."<br /><br />More to come in a little bit...http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/05/rawls-gertel-explain-finisar-optium-merger.html2008-05-16T21:02:00.000ZCoreOptics tries for coherence on 40G, 100Gnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />As <a href=$25-million-funding-round/>yesterday's funding announcement</a> pointed out, CoreOptics plans to use part of its $25 million capital influx on new product development. In the short term, that means 40G technology; according to CoreOptics VP of Global Sales, Marketing and Business Development Saeid Aramedeh, a new IC is expected to be announced in the next few weeks, to be followed by the second half of next year by a 40G transponder. A 100G product will follow. The trick for CoreOptics, he says, is timing the 100G introduction in time to meet market requirements without killing the 40G offering's prospects.<br /><br />The company plans to pair <a href=>its MLSE-2 electronic dispersion compensation technology</a> -- yes, it's based on maximum likelihood sequence estimation and it's the company's second version of the technology -- with coherent optical receiver/detection formats. (Coherent receiver technology also is being <a href=>used by Discovery Semiconductor</a> and Nortel, the latter <a href=,-announces-new-customers/>pairing it with dual-polarization quadrature phase-shift keying</a>.)<br /><br />CoreOptics already offers 40G technology -- it was designed into the platforms Marconi (now part of Ericsson) sold to Deutsche Telekom -- but this part of the company's business represents only about 10% of sales. Aramedeh naturally expects this to grow, albeit slowly.<br /><br />"Personally, I don't see this happening over the next six months or 12 months. It's probably more of a two-year journey before we see 40G in mass deployment from our perspective," he says.<br /><br />Which brings us to the dilemma 100G poses. "Certainly if you look at the architecture of 100-gig, it has much promise for being more cost-effective than any 40-gig solution," Aramedeh explains. "So the question becomes if we were to expedite 100-gig delivery, would that make sense and then 40-gig would have a shorter life or never go into mass deployment."<br /><br />A short tenure for 40G would pose a problem for CoreOptics, Aramedeh implied. "When we look at technologies that are needed to make 100-gig a reality, electronic dispersion compensation is definitely a part of that. And for us, the technology roadmap to support 100 gig goes through 40 -- we have to do 40 and 40 coherent to get the learnings from the experience before we can confidently say what architecture we could implement for 100-gig transmission," he explains.<br /><br />For now, Aramedeh thinks he'll have enough time with 40G to get 100G right. "From a timing perspective, certainly we don't see that [100G] as an immediate need. And the market is divided on that," he admits. "You talk to some folks and they say, 'You need to go into trials the end of next year.' But for volume ramp, we certainly don't see it before 2011 or 2012."http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/05/coreoptics-tries-for-coherence-on-40g-100g.html2008-05-15T19:53:00.000ZQwest aims at UTOPIA, iProvonoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />When I put together <a href=>our article on Qwest's new FTTN roll out</a>, we took up their PR firm's offer to field some questions I had regarding how they plan to provide the services and where. I didn't receive answers until after I had posted the story and included it in our e-newsletter that day. But thinking that late is indeed better than never, here they are.<br /><br />The most interesting tidbit, given the <a href=>recent troubles that the UTOPIA</a> and <a href=>iProvo</a> FTTH initiatives are having, is that Provo and Salt Lake City are two of the markets in which Qwest's Connect Quantum and Connect Titanium will appear first. (No, Salt Lake City isn't part of the UTOPIA project -- but several of the munis involved in the project surround the city.) You don't need to be Sun Tzu to know that there's nothing like attacking a adversary at a moment of extreme weakness.<br /><br />Here's a list of the 23 markets where Qwest will roll out the FTTH-enabled services first:<br /><ul><br /><li> Arizona: Phoenix, Tucson<br /><li> Colorado: Denver, Colorado Springs, Ft. Collins<br /><li> Idaho: Boise<br /><li> Iowa: Des Moines, Iowa City, Cedar Rapids<br /><li> Minnesota: Minneapolis/St. Paul<br /><li> Nebraska: Omaha<br /><li> New Mexico: Albuquerque, Santa Fe, Las Cruces<br /><li> Oregon: Portland, Salem, Bend, Eugene<br /><li> Utah: Salt Lake City, Provo, St. George<br /><li> Washington: Seattle, Vancouver<br /></ul><br /><br />As mentioned in the article, Connect Quantum will offer download speeds of up to 20 Mbits/sec and Connect Titanium up to 12 Mbits/sec. Qwest now tells me that the upload speed for both is 896 kbits/sec. "We are continually looking at new technology that will expand both download and upload connection speeds, including options such as pair bonding and VDSL2," according to the Qwest source who provided the information. (His/her identity was not revealed.)<br /><br />In response to a question about future video services over the FTTN network, the Qwest source cited the carrier's existing relationship with DirecTV. The source added that DirecTV has announced plans to add video on demand later this year, "so that will be an exciting video product for our customers and it's a complement to these faster speeds."<br /><br />Finally, when it comes to technology suppliers, the source said, "We working with a variety of vendor partners, including Motorola, for our fiber-to-the-neighborhood roll out."http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/04/qwest-aims-at-utopia-iprovo.html2008-04-28T16:02:00.000ZFrance champions fiber-to-the-apartmentnoemail@noemail.orgMeghan Fuller HannaPosted by <a>Meghan Fuller Hanna</a><br /><br />In the last issue of <i>Lightwave Europe</i>, European editor Kurt Ruderman noted that in less than two years, France has become Europe's most competitive FTTH market. (Click <a href=>here</a> for that story.)<br /><br />It may have taken another step forward today.<br /><br />The <i>International Herald Tribune</i> is reporting that the French government plans to require all new apartment buildings with more than 25 units to be outfitted with fiber by 2010. The cost of installing fiber would be included in the sale price of the apartment. (See <a href=>"France to require apartments to come wired."</a>)<br /><br />"The government's goal is to give very fast broadband a push in the back," government spokesman Luc Chatel was quoted as saying. "There's an obstacle to access, which is the entrance to the building."<br /><br />The proposed law, to be voted on by Parliament this summer, would allow all network operators access to the buildings. Those operators would then decide among themselves whether and how to share local neighborhood switching nodes, says the article. And for issues that go beyond building access, the government would defer to the judgment of France's telecommunications regulator, ARCEP. <br /><br />There is no shortage of competition in France today, particularly around Paris. France Telecom--the first European incumbent to commit to FTTH--and alternative providers like Free and Neuf Cegetel, plus cable TV provider Numericable, are deploying fiber at a rapid clip. Each of their targeted homes passed numbers in the millions. And all have pursued these plans aggressively despite regulatory uncertainty in areas like vertical deployment and building access. It will be interesting to watch what happens once these regulations are in place.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/04/france-champions-fiber-to-the-apartment.html2008-04-24T21:06:00.000ZA guest blognoemail@noemail.orgFrank Coluccio<p><i>Editor's Note: One of our visitors asked to start a new thread. Here it is.</i><br /> <br /> Upon my first reading the heading, &quot;A first step toward sanity,&quot; I'd have bet money that you were referring to Oki's breakthrough in the ultra long haul space, by achieving, reportedly, up to 160 Gbps at the present time over thousands of km's, thus appearing to find a pathway to breaking the back of PMD's stultifying effects on the upper bounds of throughput over ULH distances: <a href=""></a>.<br /> <br /> Until now, the industry seems to have almost surrendered to the idea that parallel optics (i.e., using various methods of effecting &quot;inverse multiplexing&quot; over multiple wavelengths) is the only approach to achieving speeds greater than 10G over extended distances, and in many cases even 10G, never mind 100G, 160G, or beyond.<br /> <br /> As recently as last month I read an account that covered this topic in Lightwave by an author who stated that PMD has little effect on singlemode fiber these days. (See the article here.) Of course, many manufacturers of &quot;newer&quot; single mode are affected less than most variants of &quot;older&quot; single mode due to recent improvements in manufacture, but PMD at some point (as symbol rates are raised sufficiently high) is an anomalous characteristic that affects all types of fiber, through both static and dynamic manifestations.<br /> <br /> It occurs to me that many authors who make these assertions either have products to sell that use parallel optics, or they have come to a foregone conclusion for other reasons that symbol rates will remain at 10 Gigabaud or below indefinitely, or so it seems, while employing various forms of analog (duo-binary, quadrature-based, etc.) modulation techniques, hence making it easy to suggest that higher speeds can indeed be achieved in the presence of PMD.<br /> <br /> While this may be true, at what cost is this occurring in terms of the number of wavelengths that must be used in support of ever-higher throughput rates? And at what cost is it occurring in terms of forfeiting optical-level compatibility between a growing number of dissimilar analog modulation schemes as they begin to converge in large numbers and require handing off to one another at optical network nodes? Thoughts?<br /> <br /> Frank A. Coluccio, DTI Consulting Inc.<br /> <a href=""></a></p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/04/a-guest-blog.html2008-04-15T22:29:00.000ZA first step toward sanity?noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br /><a href=>As we reported yesterday</a>, Tellabs has decided to walk away from Verizon's GPON program. As the story indicates, the company isn't providing many details about its thinking, other than "we did not find the deal economical."<br /><br />However, I have to give Tellabs credit for having the sense to walk away from a high-profile program when the numbers just didn't add up. Yeah, sticking around would mean the company could still say it was a part of the country's biggest FTTH deployment. But being the #3 supplier for a GPON rollout that's still in its nascent stages couldn't have offered much hope of return on R&D investment. The situation reminds me of those stories you hear about titled families in Europe living almost like paupers in the ancestral castle; at some point, the sacrifices necessary to keep up appearances become irrational.<br /><br />In an industry that is known for wrapping products in dollars just to keep them moving out the door, Tellabs' ability to say, "Enough!" provides an example that others ought to ponder.<br /><br />Of course, the smartest move is to avoid getting involved in programs like this in the first place. But, hey, one step at a time.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/04/a-first-step-toward-sanity-.html2008-04-03T15:56:00.000ZIs WDM-PON on your radar?noemail@noemail.orgMeghan Fuller HannaBy <a>Meghan Fuller Hanna</a><br /><br />Last week, Infonetics Research released its latest PON market forecast, which included, for the first time, a quantifiable analysis of WDM-PON. The market research firm believes annual port shipments for WDM-PON will grow at a 41% CAGR between 2007 and 2011. <br /><br />Neither the 41% CAGR nor the fact that Infonetics has included an actual forecast of the WDM-PON market surprises me. I would characterize WDM-PON as the sleeper trend of last month's OFC conference. Representatives from no less than a half dozen Tier-1system vendors confirmed to me that they are exploring the technology.<br /><br />Here are some random tidbits I've been able to glean thus far:<br /><br />* Some believe WDM-PON has a solid business case now relative to GPON, but the sticking point, as you might imagine, appears to be the cost. System vendors will lean heavily on the component suppliers (like they aren't under enough pressure already) to get the price reductions required before widespread deployment is economically feasible.<br /><br />* When we think of WDM-PON, we tend to think of it in terms of residential service delivery, but early deployments will likely deliver high-capacity business services and/or IPTV backhaul. <br /><br />* There is a big difference between CWDM-PON and DWDM-PON. CWDM-PON would be less expensive and a far easier proposition; pluggable CWDM optics are available now. However, operators would be limited to a capacity of eight to 16 wavelengths. DWDM-PON, by contrast, could support upwards of 40 to 80 wavelengths and beyond, but it represents a much greater technical challenge. The AWGs, for example, must be athermal and environmentally hardened. And the cost of the optics is higher because of the narrower channel spacings. <br /><br />At a macro-level, WDM-PON makes a lot of sense. We're already seeing the optical layer begin to migrate further and further into the access network (the emergence of edge ROADMs would be a key example), and some sort of access WDM would seem like a natural extension to this trend.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/03/is-wdm-pon-on-your-radar-.html2008-03-26T21:24:00.000ZBarriers to consolidationnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Everyone says that the optical components space needs to consolidate -- but, based on observations made at the OSA/Lightwave Executive Forum right before OFC/NFOEC, industry executives don't think consolidation will reach the levels necessary for a healthy industry any time soon. Here's a list of some of the barriers to consolidation that I've heard about over the last three weeks. For more details, keep an eye out for an article I'm in the middle of writing for our April issue.<br /><br /><b>Why Consolidation is Happening Slower than Necessary</b><br /><br />1) Mergers between weak companies won't work -- and how many strong, profitable companies do you know of in the optical communications space?<br />2) You can do a deal that favors the shareholders of one company, but it's tough in the current environment to benefit shareholders of both companies.<br />3) Japanese companies don't do M&A.<br />4) The mathematics of merging two companies in the same market space look like this: 1 + 1 = 1.5.<br />5) Greed gets in the way sometimes.<br />6) That war chest I hoped to build with my IPO is much smaller than expected.<br />7) There are enough stupid VCs around to keep bad companies afloat.<br />8) Ego gets in the way sometimes.<br /><br />Perhaps you've heard other reasons. If so, let's hear them.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/03/barriers-to-consolidation.html2008-03-13T20:20:00.000ZAre designers ready for JDSU's Superblade?noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br /><a href=,-EDFA,-OCM-in-a-single-blade/>JDSU's Superblade announcement</a> was one of the highlights of last week's OFC/NFOEC. The integration of a wavelength-selective switch (WSS), EDFA, pre-amplification, and optical channel monitoring (OCM) onto a single-slot blade with its own OSS certainly represents a step forward in terms of functional integration.<br /><br />A few hours after the Superblade announcement, I visited the booth of Israeli EDFA and OCM specialists RED-C Optical Networks. A source there said that his company had attempted a similar integration in 2006, <a href=>combining in-house EDFA and OCM expertise with Capella's WSS</a>. The source said the combination hadn't been as warmly received as RED-C and Capella had hoped, mainly because systems designers didn't want to give up software control of the WSS.<br /><br />RED-C has a partnership with JDSU, so I asked the source if his company had passed along its experiences to JDSU. The source said they had, but that JDSU suggested that RED-C's experience was a couple of years old and the market was now ready for what the Superblade has to offer, including the OSS.<br /><br />Contacted later at his booth, JDSU's General Manager, Subsystem Products, Optical Communications Doug Alteen said that his team hadn't heard about RED-C's experiences. He added that the OSS doesn't prevent the host system from controlling the operation of the blade.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/03/are-designers-ready-for-jdsu-s-superblade-.html2008-03-03T19:54:00.000ZAbout our Top 5 vendorsnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a> <br /> <br />If you're reading this, you've probably already visited the story describing our picks for the Top 5 component/subsystems vendors for 2007. (And if you haven't read it, <a href=>you can find it here.</a>) Before you offer your comments, let me tell you about some of the companies that <i>almost</i> made it. <br /> <br />In particular, we were impressed with companies who made aggressive moves to expand into new areas or improve market share through acquisition. Oplink acquired Optical Communication Products (<a href=>see here</a>) and moved within Ovum RHK's Top 10 list of vendors for the four quarters beginning with Q406. <a href=>Luminent's acquisition of Fiberxon</a> to create Source Photonics had the same effect. And <a href=>Emcore's acquisition of Opticomm</a> in April and <a href='s-Optical-Platform-Division/>the telecom portions of Intel's Optical Platform Division</a> in December may produce a similar result once the latter deal closes. <br /> <br />However, these moves weren't quite enough to overcome the momentum we believe our Top 5 picks generated during 2007. <br /> <br />You may fire when ready...http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/02/about-our-top-5-vendors.html2008-02-20T23:31:00.000ZTo PONP or not to PONP: For some, that is a questionnoemail@noemail.orgMeghan Fuller HannaPosted by <a>Meghan Fuller Hanna</a><br /><br />In this space a few months ago, I cited the emergence of the packet optical network platform (PONP) as key trend to watch in 2008. (Click <a href=>here</a> for that blog entry.) Certainly, the analysts' numbers indicate a steep deployment curve, with the market netting as much as $1.7 billion in annual sales by 2010, by some accounts. <br /><br />But not everyone advocates this approach. I spoke recently with Russ Esmacher, manager of technical marketing for Cisco's Optical Transport Business Unit, on the subject of ROADMs, which lead to a discussion about the PONP. Esmacher gave me Cisco's admittedly contrarian view that the all-in-one-box approach to packet optical networking does not make sense for the many carriers that have already invested heavily in ROADM technology. <br /><br />Instead, Cisco gives its customers the option of taking an ITU wavelength off a router and plugging it directly into the ROADM, a technology it calls IP over DWDM. Cisco also offers an MSPP on a blade that performs classic SONET/SDH ADM and Ethernet. Finally, the vendor rounds out its packet capability with another card form factor, the Ethernet Xponder, that does Layer 2 Ethernet with 50-msec restoration. <br /><br />"We saw that the ROADM layer is in place," said Esmacher. "It is a protocol-agnostic converged layer. To our customers who have routers next to it, who have ADM needs next to it, we are basically saying, 'Don't change your back office OSS. Don't change how you deploy things. You've already got this ROADM in place. Just change the on-ramp to it.' This is different from what a lot of our competitors are doing," he admitted, "but, we have enough market input from our customers to think this is the right strategy." <br /><br />To be fair, the PONP does represent a paradigmatic shift in network build out, particularly from a backoffice and OSS perspective. The folks at Cisco maintain that carriers simply aren't interested in changing their procedures in such a radical fashion.<br /><br />But Verizon's recent packet optical transport platform (P-OTP) RFP suggests otherwise--to the tune of $500 million or more, which is the estimate Morgan Keegan analysts have given for the contract. <br /><br />True, this isn't a Blu-Ray versus HD DVD-kind of debate where one format will eventually win the day; individual carriers will continue to deploy whatever architecture suits them best. But like Blu-Ray, the PONP does seem to have some industry heavyweights lining up behind it. And, as is often the case, the smaller guys may well follow suit.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/02/to-ponp-or-not-to-ponp-for-some-that-is-a-question.html2008-02-15T21:48:00.000ZLean and meannoemail@noemail.orgStephen HardyPosted by <a href="">Stephen Hardy</a><br /><br />Remember those "lean manufacturing" initiatives we heard optical component and subsystem companies blame for last year's poor start? (See <a href=",-but-growth-ahead/">here</a> and <a href=$936-million-in-Q'02-2007/>here</a> for reminders.) Well, they're not over yet -- they're just moving farther down the food chain.<br /><br />In a recent conversation, JDSU's Craig Iwata, senior director of marketing and business operations, told me that the adoption of lean business practices by systems houses last year had two major effects. The first was that it gave the systems vendors an opportunity to re-examine their inventory requirements; several decided they didn't need to carry as many spare parts, which led to the aforementioned slowdown in orders.<br /><br />The second effect was a shortening of lead times, down to three weeks in some cases. So what can a company like JDSU do to meet these reduced lead times? Why, institute lean business practices itself. The company feels so strongly about its decision that it's hosting a webcast on the subject February 13. (<a href=";pagepath=News/News_Releases&amp;id=1851">Find details here</a>.)<br /><br />You can't blame JDSU and other subsystem vendors for taking this approach. But this certainly can't be good news for component suppliers to JDSU and companies of its ilk. They already suffered a trickle-down effect from the spread of lean practices at the systems level in 2007; a repeat at the subsystems level suggests component companies will see another year get off to a rocky start.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/02/lean-and-mean.html2008-02-08T18:51:00.000ZSuper Bowl XLII: A multimedia extravaganzanoemail@noemail.orgMeghan Fuller HannaPosted by <a>Meghan Fuller Hanna</a><br /><br />This year's Super Bowl will have a high fiber content, and I'm not talking about the chili I'm planning to make for the big game. <br /><br />All the glamour and excitement of Super Bowl XLII will unfold Sunday night at the University of Phoenix Stadium in Glendale, AZ, a stadium I profiled last year in the pages of <i>Lightwave</i>. (See <a href=>"Air-blown fiber is an MVP at University of Phoenix Stadium"</a>.) <br /><br />I had the pleasure of interviewing Cardinals IT director Mark Feller, who told me that the stadium features 71,000 feet of air-blown multimode fiber and 30,000 feet of air-blown singlemode fiber. <a href=>Sumitomo Electric Lightwave</a> is the supplier. This fiber infrastructure supports the team's internal communications needs and provides the backbone over which live games are broadcast to the world.<br /><br />But network traffic does not begin or end with the television broadcast. I've done a little research, and it turns out the Super Bowl generates a veritable bonanza of online traffic, most of which, presumably, rides over a fiber network at some point in its journey from origination to destination. For football fans, the Internet has become a big part of their pre- and post-game activities. <br /><br />This year's Super Bowl is expected to drive 2.4 million high-definition (HD) television units, generating some $2.2 billion in sales, reveals the third annual "Sports and Technology" survey conducted by the <a href=>Consumer Electronics Association (CEA)</a> and the <a href=>Sports Video Group</a>. That's hardly surprising. But what is surprising is just how many of the respondents plan to watch the game on more than one screen. <br /><br />"We have long known that the Super Bowl influences HDTV unit sales," notes Tim Herbert, CEA's senior director of market research. "We are now finding ties to other technologies consumers use to enhance their Super Bowl experience. This year, 18% of consumers watching the game expect to have a laptop PC nearby to check stats, IM with friends, or check betting lines," he says. "Another 12% plan to use their mobile phone for the same purpose."<br /><br />And this interactivity does not end with the fourth quarter. According to the CEA, 57% of HDTV owners say they will go online after the game to view memorable moments or TV commercials. <br /><br />In fact, the Super Bowl now generates a flurry of online activity even after the last of the confetti has rained down on the triumphant team. Last year's Super Bowl advertisers saw a collective 50% increase in Web traffic the day after the game, from 8.5 million unique visitors on Super Bowl Sunday to 12.7 million unique visitors on Monday, notes <a href=>The Nielsen Company</a> in its "2008 Guide to the Super Bowl." FedEx saw the biggest jump in Internet traffic last year; its Web site was below the reporting cutoff on Super Bowl Sunday (the sample size was insufficient for a reliable projection), but jumped to a whopping 1.1 million visitors the next day, report Nielsen analysts.<br /><br />And the pundits, once dedicated to analyzing the success or failure of TV ads in isolation, are now examining the level of integration between an advertiser's TV ads and its presence in online search and social websites like MySpace, Facebook, and YouTube.<br /><br />"Advertisers are becoming savvier about creating online buzz around events," says Peter Hershberg, managing partner at <a href=>Reprise Media</a>, which is conducting its fourth annual Super Bowl Search Marketing Scorecard. "Unlike many lost in the previous years, marketers are expected to finally use search and social media sites to capitalize on the excitement and brand awareness generated by their ads in the big game."<br /><br />So if you play your cards right, it sounds like $2.7 million--the reported price for a 30-second ad--buys you more than it used to . . . <br /><br /><br />In my research, I uncovered a few more fun facts about this year's big game, and since it's Friday afternoon and--let's face it--we're all just counting down the minutes until the big weekend, I thought I'd share them here: <br /><br />*Scarborough Sports Marketing estimates that 63% of Patriots fans in Boston have a broadband Internet connection. <br /><br />*According to a <a href=>Harris Interactive</a> survey sponsored by The Workforce Institute and <a href=>Kronos Inc.</a>, 1.5 million U.S. adults may call in sick to work the day after the Super Bowl, and an additional 4.4 million employees say they may arrive late. "Super Bowl-related absences could be particularly striking for organizations with a high population of Gen X and Gen Y employees," says the report, "as the majority of employed adults who say they may call in sick the day after the Super Bowl are males and females between the ages of 18-34 years."<br /><br />*I can't argue with the results of this national consumer survey commissioned by <a href=>Comcast</a>. When asked who would look best in HD at the Super Bowl, Patriots quarterback Tom Brady was the most popular choice among six players, garnering 27% of the vote. (Personally, I think Tom Brady would look good in any and all media, but I'm admittedly a little biased.)<br /><br />*And here's something I bet you didn't know: <a href=>The Hass Avocado Board</a> estimates that Americans will eat 49.5 million pounds of avocados on Game Day, enough to cover University of Phoenix Stadium end zone to end zone in more than 19 feet of avocados. In an unrelated study, The Nielsen Company found that tortilla chips were the most popular snack food during last year's Super Bowl, with a 29% sales increase and a $13.4 million incremental sales boost. Hence the need for 49.5 million pounds of avocados, I guess. Still, that's a lot of guacamole.<br /><br />Well, that's all I've got. Have a great weekend, everyone. I'll see you back here on Tuesday. (Just kidding, Stephen.)http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/02/super-bowl-xlii-a-multimedia-extravaganza.html2008-02-01T17:33:00.000ZTidbits from Tokyonoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Here are a few post-Fiber Optic Expo tidbits to tide you over until I'm back in the States and (hopefully) able to get back on a normal bio rhythm:<br /><br />The PON transceiver product line from Xponent Photonics seemed a bit out of place among the fiber and other glass component products at the <a href=>Hoya Communications</a> booth. Hoya was an investor in the company and is keeping its technology alive, at least for the time being. No word on whether Hoya is hoping to find a buyer or to succeed where the original management failed...Lots of 40G on the show floor from Japanese companies, but almost all of it was in the form of modulators. Yokogawa was the only Japanese company touting a 40G transceiver. More on this and other 40G technology at FOE next week...China Telecom appears to be moving forward with GEPON deployments, with an RFP for about 60,000 lines discussed on the show floor. The problem with tracking what China Telecom is up to, said a source, is that provincial management often acts independently from central management, so programs and purchase decisions could come from almost anywhere...Both Telekom Malaysia and BSNL in India are reported to be looking at FTTH deployments. More on this next week as well...There's apparently life in the XFPe form factor. Mitsubishi introduced a fixed-wavelength device at the show and has a full C-Band, 50-GHz tunable version in development. Expect to see more XFPe talk on the show floor at OFC/NFOEC...Finally (for now), SunSea Telecommunications of China is aiming to position itself as a domestic alternative to 3M and Fujikura for Chinese carriers looking for mechanical splicers. The company expects to have a product ready by May.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/01/tidbits-from-tokyo.html2008-01-18T11:59:00.000ZTime to worry?noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />If a presentation given this morning at the Fiber Optic Expo in Tokyo is any indication, the financial community is already getting nervous about this year's prospects for the carrier market -- and by extension, capex levels and funding for FTTH initiatives, particularly in developed countries such as the U.S. and Japan.<br /><br />Atsushi Yamaguchi, managing director, UBS Investment Research and senior analyst within the Equity Research Department at UBS Securities Japan Ltd., pointed to a general weakening in the U.S. economy and indications from Cisco that enterprise sales have slowed to suggest that carrier revenues will show decline this year. Such a decline will lead to investor pressure to reduce spending on costly projects -- such as FTTH. In general, Yamaguchi foresees restrictions on fixed-line capex growth among the U.S. RBOCs this year.<br /><br />Meanwhile, in Japan, Yamaguchi pointed to the fact that NTT has lowered its goal for FTTH subscribers and a slowdon in revenues as indicators of a similar scenario when it comes to NTT's fixed-line spending.<br /><br />Yamaguchi's thesis was that growth will be found in emerging markets, such as China, rather than in established markets. Yamaguchi said upcoming quarterly revenue figures from Verizon and AT&T will provide early signs of how much of a revenue decline is in the offing in the U.S.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/01/time-to-worry-.html2008-01-17T15:14:00.000ZNintendo's Wii: Driving demand for fiber in Japannoemail@noemail.orgMeghan Fuller HannaBy <a>Meghan Fuller Hanna</a><br /><br />I have a confession to make: I've never gotten the whole video game thing. I never played Pacman or Donkey Kong as a kid. I never waited with bated breath on Christmas morning, as so many of my friends did, to see if Santa had left an Atari system under the tree. Even today, I don't usually have anything to add when friends and co-workers wax poetic about this new game or that new game . . . and I'm starting to wonder if there's something wrong with me.<br /><br />So popular is Nintendo's new Wii gaming system, for example, that Japan's NTT recently announced a deal with the company to co-promote the Wii and its FTTH service. <br /><br />Nintendo and NTT say they will jointly operate call centers to provide support for those connecting to the Internet via their Wii console, and NTT is offering a price break on start-up packages for Wii users. <br /><br />It's easy to see how both companies might benefit from the deal. Later this year, Nintendo plans to offer a new service, WiiWare, which will enable Wii users to purchase new games via the Internet--a feature that presumably requires the kind of bandwidth that only fiber can deliver. And NTT, which hopes to pass 20 million subscribers by 2011, likely realized the chance to market its FTTH service to all those Wii fanatics was too good to pass up. <br /><br />The crazy thing is it just might work. If Verizon offered such a deal here, my brother-in-law absolutely would be the first to sign up. He and my sister have a Wii console (I think they had to give up their firstborn to get it), not to mention an Xbox 360 and PlayStation2. <br /><br />On a recent visit to their house, my brother-in-law showed off his most recent purchase: A game called "Rock Band" that today runs on the Xbox 360 and PlayStation2, though a Wii version is expected at some point in the future. It's sort of like next-generation karaoke, complete with color-coded instruments that match patterns on the TV screen. You just follow the pattern, press the corresponding buttons on the guitar or the drum set, and voila! You're a rock star. Through an Internet connection, you can even play with band mates around the world, and new songs are available for download every week. <br /><br />Of course, we had to try it. In our little band, my brother-in-law cut loose on the drums, my husband jammed on lead guitar, and my sister tackled base guitar. Which left me--whose only musical ability is turning the dial on my stereo to the "on" position--to wow the virtual crowd on lead vocals. <br /><br />I have to admit, it was fun. Actually, it was a lot of fun, and I can sort of see how people might get addicted to it. <br /><br />That said, <i>Lightwave</i>'s readers are forever asking me which application I think will serve as <B>THE</B> tipping point for FTTH in North America. Do I really think it's the Wii or the Xbox 360 or games like "Rock Band" and "Guitar Hero"? With all due respect to my brother-in-law and other virtual rockers, probably not. <br /><br />But I do think the NTT/Nintendo partnership underscores the fact that there likely will not be one "killer app." In reality, an amalgam of applications will spur the further penetration of FTTH, and we shouldn't discount any of them (no matter how silly we may feel singing The Police's "Roxanne" into a plastic microphone. For the record, it was like an American Idol tryout, with my husband playing the role of Simon Cowell. Every once in a while, he still busts out a good-natured "Roxannnnnne!" to remind me of my less-than-Grammy-Award-winning performance.)<br /><br />The bottom line is this: I applaud NTT for taking advantage of the Wii's popularity to promote its FTTH service, for recognizing that any chance it gets to introduce new subscribers to the benefits of fiber is a good one. And I think it's only a matter of time before other carriers follow suit. <br /><br /><br />[Thank you to Bill St. Arnaud for alerting his mailing list to this announcement. If you aren't already a regular reader of his various blogs, they are worth a look: <a href=></a>]http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/01/nintendo-s-wii-driving-demand-for-fiber-in-japan.html2008-01-11T18:33:00.000ZHybrid ONT/gateways? Bah humbug!noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />While competitors such as PMC-Sierra, BroadLight, and Conexant Systems have released chips designed to support the development of hybrid GPON ONT/residential gateway platforms (see announcements <a href='fiber-access'-chip-for-ONT/gateway-hybrids/>here</a>, <a href=>here</a>, and <a href=>there</a>), iamba Networks CEO Moshe Nattiv recently told me he isn't sure what all the fuss is about, at least in the short term.<br /><br />iamba develops and markets a wide range of PON products, from chips and software to ONTs.<br /><br />The regulations and a requirement for network demarcation points their technicians can access easily will continue to drive Verizon, AT&T, and most other U.S. carriers to have the ONT on the outside of the home and the gateway inside, he believes. And while others have suggested France Telecom might have interest in a hybrid platform, Nattiv expects that unbundling requirements imposed by French regulatory authorities will lead to configurations in which a single ONT can be shared by multiple service providers -- each of whom will want to use their own gateway.<br /><br />MDU deployments will require separate ONTs and gateways as well, he says.<br /><br />Nattiv was quick to add that this viewpoint isn't meant to imply that iamba Networks doesn't have a chip for such hybrid platforms on the roadmap. It's just that he doesn't see any reason to rush it to market.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/01/hybrid-ont-gateways-bah-humbug-.html2008-01-10T21:16:00.000ZMSOs go public with FTTH interestnoemail@noemail.orgMeghan Fuller HannaBy <a>Meghan Fuller Hanna</a><br /><br />An interesting little item just crossed my desk from the Society of Cable Telecommunications Engineers' Ontario Chapter. The group today issued a call for papers for its general meeting, to be held in Hamilton on Tuesday, February 5, 2008. <br /><br />While this is hardly newsworthy, here's the interesting part: The meeting's theme is fiber-to-the-home/premises, etc. <br /><br />Call me crazy, but I kinda think that's BIG news. <br /><br />It seems much has changed since <i>The Wall Street Journal</i>'s oft-cited August 17, 2006, article, "Cable industry may need to spend heavily on upgrades," which touched off the will-they/won't-they debate regarding cable MSOs and FTTH technology. <br /><br />The article cited a Cable Labs report that argued that the MSOs would have to spend billions to ward off the potential threat posed by Verizon's FTTP-based FiOS initiative and AT&T's FTTN-based Project Lightspeed. At the time, many of the largest MSOs balked at the report's findings, arguing that their current networks and the emerging DOCSIS 3.0 would be more than sufficient for their long-term needs.<br /><br />But now we have the SCTE, <B>THE</B> professional association of the cable MSOs, publicly announcing its interest in FTTH.<br /><br />According to the call, "possible topics of interest" include: <br /><br />* FTTX topologies; <br />* FTTX technologies;<br />* FTTX advanced services/content to be delivered; <br />* FTTX cost models; <br />* FTTX challenges in deployment; and <br />* FTTX state of the industry, current and future deployments.<br /><br />As someone who was only recently married after 30 (ahem) years of being single, I know first-hand there's a big difference between "interest" and commitment. For their part, the FTTH system vendors seem determined to woo the cable MSOs. (Look for my article on this subject in <i>Lightwave</i>'s February issue.) But now it seems the MSOs have decided FTTH is, at the very least, worth getting to know a little better. <br /><br />In the world of optical communications, this is definitely a Page Six-worthy development.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2008/01/msos-go-public-with-ftth-interest.html2008-01-08T21:54:00.000ZMeghan's Top Trends to Watch in 2008noemail@noemail.orgMeghan Fuller HannaPosted by <a>Meghan Fuller Hanna</a><br /><br />While snowed in yesterday, I read the most recent issue of <i>Time Magazine</i>, which was chock full of end-of-the-year Top 10 Lists. (They even published a Top 10 List of Reasons Why We Love Top 10 Lists. I kid you not.)<br /><br />Here at <i>Lightwave</i>, we compile our own version of the Top 10 List. Okay, so it's actually a Top 5, and it's not really an end-of-year list. Instead, we publish our Top 5 systems vendors, component vendors, and companies and trends to watch to coincide with OFC/NFOEC, that annual bellwether of a conference that unofficially marks the end of one year in optical communications and the beginning of the next. (For a complete list of last year's rankings, click <a href=’s-Top-5-choices-got-down-to-business/>here</a>). <br /><br />I'm afraid you'll have to wait until the February OFC/NFOEC issue to see which companies made the list this year (in part because we have yet to finalize them!)<br /><br />But, this being the gift-giving season and all, I couldn't leave you empty handed.<br /><br />So without further adieu, here are my Top 5 Trends to Watch in 2008 (in no particular order): <br /><br />&#8226 <b>Packet optical network platforms</b> (PONPs): When I first wrote about the PONP in our September issue (see <a href=>"PONPs optimized for legacy and packet transport"</a>), there was still some debate about the class of equipment to which it belonged. Is it an MSPP? An MSTP? Does it belong in the category of next-gen SONET/SDH? Well, it seems the industry has decided it's a bona fide equipment category all on its own. We know this because the analysts have begun to chart its progress! <a href=>Infonetics Research</a> forecasts the worldwide PONP market to account for $1.7 billion in annual sales by 2010. Not a bad haul for such a nascent market.<br /><br />&#8226 <b>Silicon photonics</b>: In its recent report, "New markets for telecom & datacom lasers: 2007 to 2012," <a href=>CIR</a> analysts argue that, "Silicon Photonics could be the single most important technology in the near term, gradually pushing short-wave optics and multimode fiber into niche oblivion." <a href=>Luxtera</a>, one of <i>Lightwave</i>'s Companies to Watch in '07, is now spearheading the OIDA's Silicon Photonics Alliance, which also includes Corning, Kotura, Molex, and US Conec. Look for additional coverage on this trend in <i>Lightwave</i> going forward.<br /><br />&#8226 <b>The North American cable multiple-systems operators (MSOs)</b>: Oh, those cable MSOs. Look for them to continue their strong push into the business services market in '08, and keep a close eye on potential FTTH-like activity as well. My sources tell me the cable MSOs are moving beyond the tire-kicking phase and into serious lab trials with emerging cable-centric PON technologies. There's a reason why "RFOG" (RF over Glass) is one of the hottest acronyms in the optical communications space today.<br /><br />&#8226 <b>The European FTTH market</b>: Some wonder whether the flurry of activity this year in and around the city of Paris will serve as a catalyst for the rest of the region. Competitive carrier Free jump-started the market with a sizeable investment in FTTH infrastructure around the city, prompting incumbent France Telecom to commit to its own FTTH plans. Alternative operator Neuf Cegetel also expanded its FTTH infrastructure in the city via the acquisitions of Erenis and Mediafibre. France's ARCEP is thus far the only European telecom regulator to champion FTTH, but will others follow? (See Kurt Ruderman's editorial, <a href=>"Solving the infrastructure puzzle,"</a> in the Q4 2007 issue of <i>Lightwave Europe</i>.) <br /><br />&#8226 <b>Consolidation</b>: At the risk of sounding like a broken record, to borrow a phrase from my mom, I'm putting consolidation on the list again this year. (It made our official Top 5 Trends to Watch list last year too. Come to think of it, I think it made the cut in '06 as well.) The last few years have seen consolidation at the top of the food chain, but everyone is waiting with bated breath for notable M&A activity among the component vendors. Thus far, any consolidation among the component guys has occurred on an opportunistic basis--and it's mostly involved profitable companies. Will '08 finally bring the consolidation many have predicted since the bubble burst?<br /><br />That's all you get for now. As I mentioned earlier, you'll have to wait until the February issue to learn which companies made our coveted Top 5 Lists this year. <br /><br />In the meantime, which trends will you be following over the next 12 months? Which companies impressed you this year, and who do you think seems poised to break out in '08? We'd love to hear from you!http://localhost:4503/content/lw/en/blogs/lightwave-blog/2007/12/meghan-s-top-trends-to-watch-in-2008.html2007-12-21T18:51:00.000ZNight night FlexLight?noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br /> reported on December 6 that GPON pioneer FlexLight Networks has filed for bankruptcy. (See the story <a href=>here</a>.) Haaretz's Guy Griml quotes an unidentified FlexLight executive as suggesting the company came out with a product too early. However, the executive also blamed a lack of marketing and development focus.<br /><br />"At some point it became apparent that FlexLight would be sold or closed down. The investors were impatient, and the big players, Alcatel, Siemens, and Telrad, released similar GPON products," Griml quotes the executive as saying. "As far as I'm concerned, this company is a huge miss."http://localhost:4503/content/lw/en/blogs/lightwave-blog/2007/12/night-night-flexlight-.html2007-12-18T21:36:00.000ZPOF: Still searching for a home?noemail@noemail.orgMeghan Fuller HannaBy <a>Meghan Fuller Hanna</a> <br /><br />Yesterday, the Universal Serial Bus (USB) 3.0 Promoter Group announced that it is looking to enlist additional contributors for its first industry-wide specification review, to be held in Las Vegas on January 14-15, 2008 (See <a href=></a>).<br /><br />Initially announced in September at the Intel Developers Forum, USB 3.0 is a 'superspeed' personal interconnect that promises to deliver a ten-fold increase in data transfer rates from the 480-Mbit/sec rate of USB 2.0 to 4.8 Gbits/sec from USB 3.0. This increased speed is accomplished via a fiber-optic link that will run alongside the traditional copper connection. (USB 3.0 will be backwards compatible with USB 2.0.)<br /><br />The USB 3.0 spec will target peripheral devices--including digital cameras, video cameras, and optical drives like DVD and Blu-Ray--to enable the more rapid transfer of digital content. <br /><br />While some believe that USB 3.0 could challenge the IEEE 1394 interconnect standard known as FireWire, from a <i>Lightwave</i> perspective, the more important implication is that all consumer devices will have some kind of fiber interface.<br /><br />I think that's worth repeating: All consumer devices will have some kind of fiber interface. <br /><br />And this makes me wonder: Have we finally found that 'killer app' for plastic optical fiber?<br /><br /><i>Lightwave</i> has covered plastic optical fiber (POF) from time to time (See Stephen's 2006 article, <a href=>"Proponents hope home is where the plastic is"</a> as an example), and it's always seemed to me like one of those perennially on-the-verge-but-never-quite-ready-yet technologies. Despite its potential usefulness as an alternative to existing in-home wiring, POF has yet to catch on with communications network architects (though it's doing fairly well in the automotive sector, from what I understand.)<br /><br />That said, the technology has made some impressive strides of late. This summer, Siemens' R&D arm, Siemens Corporate Technology, announced what it claimed was a significant milestone in the evolution of POF technology: Researchers successfully transmitted 1 Gbit/sec over a 100-meter-long test route in the company's laboratory--long enough for application in the home network environment and certainly long enough to interconnect peripheral devices to a PC, for example.<br /><br />The USB 3.0 specification is expected sometime next year, with commercial products appearing in late 2009, early 2010. With a bit more tweaking to increase its capacity, could POF finally find its way into the home networking space via the consumer electronics market?http://localhost:4503/content/lw/en/blogs/lightwave-blog/2007/11/pof-still-searching-for-a-home-.html2007-11-30T20:36:00.000ZGoogle wants you (maybe)noemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Here's a belated update to the last post on SFP+ optics for Google's DIY 10GbE switch. (No, I'm not going to link to it -- just scroll one item down!)<br /><br />As I mentioned in the post, I shot an email to Google to ask if the stories about the existence of the switch were accurate and, if so, what they could tell me about the optics. I didn't have a lot of hope of getting a response, particularly when my query, directed at the media relations department, elicited a canned email from the Help Desk in reply.<br /><br />However, lo and behold, I received an email from Google's Sonya Boralv (who is a frequently quoted Google media relations person) the very next day. Her response:<br /><br /><i>We can't share details on our infrastructure but would encourage any interested engineers to check out our job listings at We currently have job openings for hardware and software engineers with networking backgrounds that might be of interest to your readers.</i><br /><br />So I've struck out so far on the optics story -- but I may be the gateway to your next career. 10% of your first paycheck there will do just fine as thanks.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2007/11/google-wants-you-maybe-.html2007-11-29T21:07:00.000ZGoogle 10GbE switch opticsnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />As you've no doubt read elsewhere, Andrew Schmitt of Nyquist Capital broke a story on his blog (and if his blog isn't on your regular visit list, it should be) about Google building it's own 10GbE switch. (Read it <a href=>here</a>.)<br /><br />Andrew and I have been going back and forth over the past couple of days about his description of the optics used in the switch, which he says are nonstandard implementations of SFP+ transceivers:<br /><br /><em>What is interesting about Google's approach is that it has eschewed traditional 10GBASE optical standards and instead adopted off-standard solutions that better suit its needs for time-to-market, power and port density, and cost. While Google makes use of the SFP+ cage format, it does not use the receive dispersion compensation (EDC) function typically associated with SFP+. Instead Google is looking to employ a combination of twinax cabling for short reach (<10m) intra-rack cabling and a motley 850nm SR-like standard. Off the shelf SR optical modules appear to work well up to 100m over without receive equalization. Ironically, Finisar (FNSR) proposed such a solution several years ago.</em><br /><br />It sounded to me that either the assumption is that Google is using SR transceivers with 62.5-micron cabling (which indeed would be nonstandard -- and curious) or that SFP+ has gotten a bit too closely associated with LRM, to the point that SR versions might be considered nonstandard. Andrew tells me that "(t)hey are using standard pre-emph/receive EQ but not the full LRM EDC spec," so that's not the explanation.<br /><br />Unfortunately, Andrew has been too busy to continue our discussion beyond that last bit of information. I have a query into Google on the subject, but while I was holding my breath and turning blue waiting for them to respond, I thought I'd throw the question out to you: What am I missing here?http://localhost:4503/content/lw/en/blogs/lightwave-blog/2007/11/google-10gbe-switch-optics.html2007-11-20T20:18:00.000ZTake me in to the ballgamenoemail@noemail.orgStephen HardyPosted by <a href="">Stephen Hardy</a><br /><br />Motorola just announced the results of a survey that suggests less than a third of sports fans would rather watch a football game in person than see it at home in high-def. That compares to 45 percent who said they'd rather watch the contest on HDTV. (There must have been a fair amount of undecideds; you can check out details of the survey <a href="">here</a>.)<br /><br />In some quarters this news would elicit another round of lamentation about the increasing slothfulness of Americans. But not here. I agree that there are sporting events that are much better viewed on TV than in person. And as fiber-fed broadband services increase, so too will the reasons to stay home.<br /><br />The fact that the survey involved football games isn't surprising, because cold or otherwise inclement weather would be one factor that would routinely make people wish they were on the couch. I saw a playoff game at Gillette Stadium in 2004 where the <a href="">Patriots beat the Titans</a> with the wind chill at kickoff of -10 degrees. My buddy and I sort of dealt with it as a badge of honor -- but I wouldn't want to be a fan in Green Bay or some other northern climate (pro or college) and face such misery on a regular basis.<br /><br />Then there are sports like hockey, where it's almost impossible to follow the puck live, particularly if you wear glasses made from planetarium-strength telescope lenses like I do.<br /><br />Regardless of the sport, however, optically enabled improvements in broadband services will make staying at home even more tempting in the future. In particular, I eagerly await the day when broadcasters will give the home viewer access to the video feeds from all the cameras in the stadium. Can you imagine the remote-control wizardry you could perform once you got the hang of it, changing camera angles two or three times through the course of a play?<br /><br />The day this technology is first made available will be marked by two rounds of popping noises across America. Champagne corks will cause the first round. The second will erupt when our spouses' heads start exploding.<br /><br />All right, so maybe we should approach some broadband services with caution. But I'm getting my thumb in shape, just in case.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2007/11/take-me-in-to-the-ballgame.html2007-11-16T16:56:00.000ZMeghan's back -- and she has pictures!noemail@noemail.orgMeghan Fuller Hanna<a href=""><img style="FLOAT: left; MARGIN: 0px 10px 10px 0px; CURSOR: hand" alt="" src="" border="0" /></a><br /><div><a href=""><img style="FLOAT: left; MARGIN: 0px 10px 10px 0px; CURSOR: hand" alt="" src="" border="0" /></a><br /><br /><div><a href=""><img style="FLOAT: left; MARGIN: 0px 10px 10px 0px; CURSOR: hand" alt="" src="" border="0" /></a><br /><br /><br /><div>Posted by <a href="">Meghan Fuller Hanna</a><br /><br /><br /><br /><br /><div><br /><br /><br /><div><br /><br /><br /><div>Well, I'm back from my honeymoon, and I first want to thank all of you who called or sent email messages of congratulations. I know I have been talking about my wedding and the Red Sox ad nauseam lately (sometimes not even in that order), so I also wanted to say thanks to all who graciously endured my ceaseless chatter! </div><div><br /></div><div>I promised I would post a few photos from the big day, so here goes. The first is one of my favorites of the two of us, taken just before the reception. A few months before the wedding, Stephen's wife, Kristin Lewotsky, gave us a great piece of advice: Take a moment just for the two of you, just to be together and let it all sink in before the hoopla begins, she suggested, and we're very glad now that we took her advice. </div><div><br /></div><div></div><div>You'll no doubt recognize some folks in the second photo; the Lightwave sales and editorial staff was there en masse to cheer me on! From left to right, Lightwave's Editorial Director Stephen Hardy; National Account Manager Greg Goulski; our Publisher, Tim Pritchard; me (in the Sox hat) with my new husband, John; National Strategic Account Manager Kathleen Skelton; and Managing Editor Carrie Meadows.</div><div></div><div><br />And finally, I have included a photo of our guest book. To commemorate the anniversary of the Red Sox winning the World Series in 2004 (little did we know they would do it again the next day!) John and I asked our guests to sign a Major League Baseball-issue home plate in lieu of the traditional guest book. </div><div><br /></div><div></div><div></div><div>So thanks again for indulging me over the past few months. I promise to once again turn my attention to fiber optics (although the Patriots are working on a record-breaking season, and the resurgent Celtics are 3-0 . . . . )</div><div></div><div></div><div><br /></div><div>--Mrs. Meghan Fuller Hanna<br /></div><div> </div><div></div></div></div></div></div></div>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2007/11/meghan-s-back----and-she-has-pictures-.html2007-11-09T15:53:00.000ZLast train to FTTHnoemail@noemail.orgStephen Hardy<p>Posted by <a href="">Stephen Hardy</a><br /> <br /> So I was killing time on the phone with Christy Batts, telecom marketing manager at the Clarksville Department of Energy, before the briefing on which today's FTTH story is based, when I said something like, &quot;So, I suppose you get all kinds of 'last train' comments.&quot;<br /> <br /> And she said, &quot;Yes, we do. And that's because Clarksville is the town in the song.&quot;<br /> <br /> For those of you totally clueless when it comes to the history of bubblegum pop (which puts you in the same company as Meghan and our managing editor, Carrie Meadows, when I brought this up in a meeting this morning), Clarksville was the inspiration for the <a href="" target="_blank">Monkees</a>' hit, &quot;Last Train to Clarksville.&quot;<br /> <br /> Clarksville is, among other things, near Ft. Campbell (<a href="" target="_blank">Home of the Screaming Eagles</a>). Batts says the song is meant to <a href="" target="_blank">tell the poignant story</a> of a GI about to be shipped to Vietnam who is trying to hook up with his girlfriend one more time before he leaves.<br /> <br /> Alas, Batts says that the town has so far failed to erect a statue to the Prefab Four in recognition of <a href="" target="_blank">their contribution to Clarksville's notoriety</a>.</p>http://localhost:4503/content/lw/en/blogs/lightwave-blog/2007/11/last-train-to-ftth.html2007-11-05T21:32:00.000ZShe's getting married in the morningnoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />Tomorrow will be a sad day for single men everywhere. That's because Meghan Fuller is getting married. (Yes, that's right -- <a href=>you blew your chance</a>.)<br /><br />John Hanna is the lucky guy. (No, not <a href=>that one</a>.) He's an actuary. Believe it or not, they got to know each other while playing in a kickball league.<br /><br />The date of the wedding, October 27, was not selected at random. As every card-carrying member of Red Sox Nation knows (as well as the non-card carrying members who roll their eyes at <a href=>yet another ploy to squeeze money out of the fan base</a>), October 27 is the date <a href=>the Sawx beat the Cardinals to win the 2004 World Series</a>. Meghan planned a morning wedding to avoid conflict with Game 3 of this year's Series, just in case.<br /><br />So with the Sox in the Series, Meghan was quite concerned Wednesday when the weather looked threatening for Game 1 in Boston. If they had to postpone the first game, would they move all the games back? I told her that would mean there wouldn't be a game on her wedding day, which would pose one less distraction for all concerned.<br /><br />"But what would I do after the wedding?" she asked.<br /><br />No, I'm not making that up.<br /><br />And, yes, I believe John knows what he's getting himself into.<br /><br />So raise a glass or a coffee mug or whatever to Meghan and John (and, of course, to the Red Sox), and feel free to post your best wishes in the comment area below. Also, feel free to vote on whether I should follow through on my threat to post a <a href=>Bill Simmons-style running diary</a> of tomorrow's festivities.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2007/10/she-s-getting-married-in-the-morning.html2007-10-26T15:18:00.000ZInside the numbersnoemail@noemail.orgStephen HardyPosted by <a href="">Stephen Hardy</a><br /><br />If you subscribe to the excellent newsletter from <a href=""></a> or are a regular visitor to <a href=""><i>Telephony</a></i>'s website, you're probably aware of the controversy surrounding the most recent North American FTTH subscriber numbers the North American FTTH Council released early this month at its conference in Orlando.<br /><br />For those new to the story, Principal Analyst Teresa Mastrangelo broke with the usual market research rules of decorum by calling the 21.4 million FTTH subscriber number derived by <a href="">RVA LLC</a> "too aggressive" and suggesting that RVA and/or its council client "continue to be evasive as to the actual data that makes up these numbers." (The statements were contained in's October 18 newsletter, which you can <a href="">find here</a>.)<br /><br />As I point out in my editorial for the upcoming November issue of <i>Lightwave</i>, the implication that something slippery is going on is incendiary because of RVA's relationship with the North American arm of the FTTH Council, which uses the company as its primary source of market data. The council has repeatedly released studies conducted by RVA as if they were official FTTH Council data; as you'll notice, Mastrangelo attributed to the council a figure that she printed in her newsletter that I would guess RVA created.<br /><br />Clearly the potential controversy surrounding these market figures doesn't do the FTTH Council any good, since it implies that the council is promoting data that over-hypes the FTTH market. Certainly the last thing the optical communications space needs is another round of market hyperbole -- and the council doesn't need to have its credibility questioned, particularly as it tries to lobby Congress for more fiber-friendly policies.<br /><br />Mastrangelo caps her commentary by challenging the council and RVA to "show me the data" that backs up the 2.14 million number. And, lo and behold, Mike Render, the "R" in "RVA," has stepped up to the task. In a letter to <i>Telephony</i> (see it <a href="">here</a>), Render outlines how he derived his figures and why his numbers might not match those collected by Mastrangelo and others.<br /><br />I applaud Render for speaking up -- not only was his credibility called into question, but so was that of his client, the FTTH Council. One can debate Render's assumptions, but the spotlight is now squarely on him and off the council.<br /><br />The FTTH Council is lucky Render stepped up. This episode points to an important aspect of using market research to advance your aims: Regardless of where research numbers come from, once data appears in public next to your name, those numbers are yours. Know where those numbers came from and how they were derived, and be prepared to discuss these points.<br /><br />Market research is opinion -- and everyone has a different opinion.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2007/10/inside-the-numbers.html2007-10-23T19:44:00.000ZThoughts on 40/100G, posted by the bride-to-benoemail@noemail.orgMeghan Fuller HannaPosted by <a>Meghan Fuller</a><br /><br />In his last post, Stephen talked about the mounting evidence to suggest that the 40G market is real and viable. I just finished an article about tunable optical dispersion compensation (TODC) for our November issue, and if ever there was a market that would reflect an upsurge of activity in 40G deployment, it's the TODC market. By all accounts, the market is more active today than it was just six months ago, and the vendors say they are readying, if not already shipping, their 40G devices. <br /><br />Guy Martin, vice president of technology and business development at Teraxion, told me that of the 2,000 TODC units Teraxion has shipped this year, 80% are being used for 40G applications. <br /><br />Civcom, meanwhile, has a single-channel 40G TODC device in the final stages of development. Yair Itzhar, Civcom's vice president of sales and marketing told me that the product has already been shipped to Tier 1 and Tier 2 customers for evaluation. <br /><br />ANDevices has a 40G TODC based on PLC technology, but Wenhua Lin, vice president of technology and new product development, noted that her company's customers are now making more stringent demands in terms of the tuning range. Two years ago, 200-300 psec/nm would have been sufficient, but today they are looking for something closer to 400 psec/nm and higher. Lin confirmed that ANDevices is now working on an AWG-based device with a wider pass band and dispersion window. <br /><br />From my perspective, this round of interviews yielded far more concrete answers to questions about 40G than a series of interviews on the same topic elicited back in March (See <a href=>TODC vendors ready inline and integrated devices</a> from our April issue), lending credence to Stephen's assertion that there will be a market for 40G.<br /><br />Incidentally, I also asked the TODC vendors about the market for 100G and was surprised to hear Giovanni Barbarossa, CTO of Avanex, admit that he does not see much activity in this realm. That's news to me, as I feel like four out of every five phone calls I field has something to do with 100-GbE. But as Barbarossa pointed out, there's a big difference between 100G <i>speeds</i> and 100G <i>capacity</i>. If, as most people believe, the first generation of 100-GbE uses a parallel architecture, then the highest bit rate would be either 10-, 20-, 40-, or possibly 50-Gbits/sec, multiplexed for an aggregate of 100 GbE. And, as Barbarossa reminded me, Chromatic Dispersion becomes more problematic as bit rates scale, not capacity.<br /><br />While that's probably an obvious distinction, I've been a little slow on the uptake lately. A few too many long nights spent watching the Red Sox come up with new and innovative ways to lose to the Indians? Perhaps. Or maybe--just maybe--I might be a little preoccupied because I'M GETTING MARRIED IN NINE DAYS. <br /><br />Speaking of which, Stephen has threatened to bring his laptop to the wedding to provide <i>Lightwave</i> readers with a running commentary on the festivities. He may or may not have been kidding, so stay tuned!<br /><br />--MJFhttp://localhost:4503/content/lw/en/blogs/lightwave-blog/2007/10/thoughts-on-40-100g-posted-by-the-bride-to-be.html2007-10-18T20:28:00.000ZThere is a market for 40G -- reallynoemail@noemail.orgStephen HardyPosted by <a>Stephen Hardy</a><br /><br />In light of <a href=",-Mintera-link-for-40G-product-development/" target="_blank">today's announcement regarding the partnership between JDSU and Mintera</a>, it's clear JDSU believes the 40G market is worth pursuing. This belief comes from the company's insight into the RFPs their system house customers are fielding, according to Director of Product Marketing Craig Iwata.<br /><br />"JDSU is involved in pretty much every transport RFP that's out there. And they're all coming through with requirements for 40-gig," he told me.<br /><br />While there's a difference between requiring 40G capability and actually fielding it, Iwata also told me that the market research community is backing this viewpoint as well. As part of the presentation he and Mintera's Niall Robinson delivered to me last week, Iwata said that when you look at the 40G line-side interface forecasts from CIR, Infonetics, and Ovum-RHK, the average unit CAGR for 2008 through 2011 is 100%, with their respective hockey sticks all taking a distinctly upward turn beginning in 2009.<br /><br />Neither Ovum-RHK nor Infonetics have much on the public parts of their websites covering their 40G forecasts. You can find a little bit about CIR's view <a href="" target="_blank">here</a> and <a href="" target="_blank">here</a>.<br /><br />As usual, the big issue will be cost, Daryl Innis of Ovum-RHK told me. "There's still a serious price problem. The cost of a 40-gig line card is something like five to six times the cost of a 10-gig line card. Until that cost comes down, that's going to limit the growth rate of 40-gig," he said.<br /><br />I think there's mounting evidence that there will indeed be a market for 40G technology, particularly in the long haul and ultra long haul. (Which probably translates into a "good" but not "great" market.) Certainly 100G will eventually overtake 40G in terms of popularity, but that's going to be a fairly long time coming. The standards have to be put in place first -- and if 40G technology currently is 5X to 6X your average 10G system, one can only imaging the price differentials of the first and even second generations of 100G platforms.http://localhost:4503/content/lw/en/blogs/lightwave-blog/2007/10/there-is-a-market-for-40g----really.html2007-10-16T20:29:00.000ZThe Lightwave Blog: True QAMnoemail@noemail.orgStephen HardyPosted by Stephen Hardy<br /><br />Yes, at long last, the Lightwave editors step into the 21st Century with the launch of The Lightwave Blog. Senior Editor Meghan Fuller and I plan to share this space, which we'll use as a forum to offer opinions, report on events, comment on trends, pass judgment on rumors, and maybe show a few sides of ourselves that don't come through in print.<br /><br />This being the Lightwave site, you can expect a lot of what we'll write about will pertain to optical communications. We'll likely report on the visits of company executives to the office, our experiences at trade shows, and maybe even our reaction to relevant television commercials, like the Verizon one referenced in this post's title. (Do they run the "20-dB hot" commercial with the Verizon technician and that scary-looking kid everywhere in the U.S., or just the Northeast?)<br /><br />But we also want to give you insight into what goes on behind the scenes here at the Lightwave offices. That means bringing you into some of the conversations that normally would never leave the cubicles here in our Nashua, NH, offices. So you can expect a few posts on truly important topics such as baseball (Red Sox fans to George Steinbrenner: Fire Joe Torre? Go ahead, make our day), pop culture, and the mystical properties of chocolate.<br /><br />Naturally, we're hoping to make this blog as interactive as possible, so we welcome your comments, including a heads up on any topics you'd like to see treated or any questions you'd like answered.<br /><br />Meghan and I think this blog is going to be a lot of fun. We hope you'll bookmark this page and check us out whenever you have a few minutes.<br /><br />-- Stephenhttp://localhost:4503/content/lw/en/blogs/lightwave-blog/2007/10/the-lightwave-blog-true-qam.html2007-10-10T22:07:00.000Z 500

Cannot serve request to /content/lw/en/blogs/lightwave-blog/_jcr_content.feed on this server

ApacheSling/2.2 (Day-Servlet-Engine/4.1.52, Java HotSpot(TM) 64-Bit Server VM 1.7.0_51, Windows Server 2012 6.2 amd64)