WSS ROADMs provide the flexibility of the future-today
by Ronen Mikdashi
Reconfigurable optical add/drop multiplexers (ROADMs) have been deployed since 2005, mainly by North American carriers. Recently, though, an increasing number of European carriers have recognised the direct and indirect cost benefits of deploying this advanced technology.
After recovering from the 2000 downturn of the market, carriers and service providers are once again rushing to upgrade their optical transmission infrastructure to be better poised for today’s competitive environment. New high-bandwidth video services, business and storage connectivity, and mobile 3G backhaul networks contribute to this race.
However, unlike the network build-outs so popular during the telecom bubble, carriers today are looking for an evolutionary approach to introduce new services, while maximising revenues from legacy services and existing equipment and lowering the life cycle cost of their networks. This position calls for technologies that offer scalability, flexibility, and lower total cost-of-ownership.
Traditional fixed DWDM networks address the need for increased capacity and fibre relief but bring no flexibility. Provisioning, maintaining, and deploying such networks are complex affairs that require multiple manual operations and frequent re-engineering of the network, which results in high total cost-of-ownership. New technologies targeted at reducing overall costs and increasing flexibility in the network have been developed, primarily automatically switched optical network (ASON) platforms and ROADMs.
First-generation ROADMs were based on a demultiplexer-switch-multiplexer technology or on liquid-crystal wavelength blockers. Second-generation ROADMs were based on planar lightwave circuit technology, while today’s third-generation ROADMs are based on wavelength-selective switch (WSS) technology. The table presents a comparison of the different generations of ROADM technology.
WSS ROADM is now considered to be the technology of choice for carriers worldwide, bringing more flexibility and lower total cost-of-ownership in various network scenarios. WSS ROADMs allow carriers to route any wavelength (or any combination of wavelengths) to any node without the need to predefine traffic demands or install additional devices, thereby reducing time-to-market for new services. Furthermore, WSS ROADMs offer “colourless” ports, enabling the operator to select and then re-select the specific wavelength to add/drop at the node. This is of particular benefit when traffic is hard to predict and/or if it is expected to change often.
WSS ROADMs’ multidegree architecture means that they can be deployed not only in rings and chains as in previous generations but also in the multiring and mesh topologies so common in metro core and regional WDM networks. It is this multiple topology support that elevates ROADMs from a niche to a mainstream optical technology.
For example, in a recent deployment a new regional WDM network for a major European carrier was designed twice: once based on WSS ROADMs and again based on traditional fixed DWDM. The network was mesh topology with a nodal degree from 2 to 4. Detailed capital expenditure (capex) comparison revealed a striking and counter-intuitive result. Although at first the ROADM-based network was 10% more expensive than a traditional fixed network, once populated with even less than five wavelengths per link, it became more cost-effective (see Fig. 1).
ROADMs can also be converged with packet switching and TDM to form an agile multiservice transport platform. Such a converged platform can provide both data and wavelength services, along with traditional circuit services, all from the same system-typically no more than one-third of a telecom bay in size. This approach can bring considerable capex and operational expenditure (opex) savings. Moreover, when coupled with Optical Transport Network (OTN) functionality as defined by ITU-T, mainly under G.709, these next-generation systems can also offer longer reach, increased service transparency, improved resilience, better quality of service (QoS) through unified monitoring of all wavelength and subwavelength services, and much needed multivendor interoperability.
Figure 2 shows the concept of a multidimensional converged transport system. Also shown is an example of how lower-speed business data services are provided by the Ethernet/MPLS core, then groomed via SDH, mapped to OTN, and then optically switched by the ROADM. The same system can provide a high-speed 2.5 Gbit/s STM-16 wholesale service directly from the OTN layer while performing subwavelength switching, to combine high bandwidth utilisation with full service and timing transparency, all switched by the ROADM.
Taken together, the features of the WSS ROADM provide capex savings in several areas:
- Elimination of regenerators makes the ROADM-based network more cost-effective, typically beyond the fifth wavelength in a meshed network.
- Elimination of back-to-back connections between separate systems typically results in up to 45% savings.
Meanwhile, opex savings accrue because:
- Remote provisioning and reconfiguration eliminates the need for forklift upgrades whenever additional wavelength services are provisioned.
- The use of one converged system instead of three or more different systems reduces the cost of spare parts.
ROADMs also improve QoS and overall competitiveness. The use of WSS ROADM with tunable lasers enables any-to-any connectivity, allowing the operator to react to any service opportunity far more quickly than ever before. The use of OTN for managing all wavelengths as well as for subwavelength grooming ensures service transparency and better manageability, resulting in improved overall QoS.
Until not so long ago, industry analysts speculated whether ROADM technology would proliferate in considerable numbers beyond the North American cable-telco “war”. Recently, though, Infonetics Research forecast dramatic growth for ROADMs in the global market, specifically in EMEA. According to this research, total WDM ROADM switch (metro and long-haul) hardware will reach over $400 million, representing 37% of the total WDM revenue in 2010, from 29% at the end of 2006, as represented in Fig. 3.
Analysis of multiple RFPs issued by European carriers during the last 12 months shows that the adoption of ROADM technology is expected to come not only from wireline carriers but also from other providers, including wireless, carriers of carriers, utility telcos, and government and defense networks. Infonetics Research also foresees high traction for converged ROADM-TDM-Ethernet systems, referred to as “hybrid migration”, making up for some of the expected decline in traditional SDH/SONET multiservice provisioning platforms. This trend is expected to apply to the global market, including Europe.
Both system and subsystem vendors are working on further enhancing ROADM platforms to provide more benefits to carriers. Major areas include:
• Edge ROADM: Today, ROADMs are deployed mostly in metro core, regional, and long-haul networks. Nevertheless, carriers do want to introduce flexibility and automation in the metro edge layer as well. To support this, various new ROADM technologies are being developed that are cost- and performance-optimised for ring applications with low channel counts.
• ASON-based optical networks: ASON aims at making optical networks more intelligent and automatic, thereby simplifying installation and provisioning and introducing efficient protection and restoration schemes. To date, ASON has been primarily applied to SDH/SONET networks. The challenge for the near future is to find ways to seamlessly integrate ASON with WSS ROADMs to achieve intelligent photonic networks.
• Deeper packet and optical convergence: As explained before, converged hybrid platforms bring efficiency, flexibility, and operational simplicity. This trend is bound to continue, as vendors focus on converged architecture in their products and solutions.
Ronen Mikdashi is senior product marketing manager, Transport Networking Division, at ECI Telecom (www.ecitele.com).