Getting From Here to CCAP

BTR's CCAP panel at the Cable Show, (L-R): Tom Cloonan, ARRIS; Mike Kelsen, TWC; Jeff Finkelstein, Cox; Jorge Salinger, Comcast; and Carl Weinschenk, BTRThe cable industry's engineering community always has done a great job of producing acronyms. Perhaps one that it should think about is CCAP GPS.The GPS element of the facetious acronym is that the road to CCAP - when, via what route and how long the trip will take - is unique for each MSO. The many decisions operators face CCAP were discussed at a June 11 breakfast panel at The Cable Show in Washington, DC. The event was presented by Broadband Technology Report and sponsored by ARRIS (NASDAQ:ARRS).The cable industry, in its efforts to please its customers and stave off competition, has spent decades piling technologies atop technologies, from video to telephony to high-speed data. In addition to creating these layers, the technology driving each has changed drastically, but often requiring legacy platforms to remain. It all works - but the results are inefficient Rube Goldberg-esque platforms and headends that are as densely packed as New York City subway trains at rush hour.Comcast: The Transition is Under WayCCAP - the Converged Cable Access Platform - will change all that. That, at least, is the plan. And it is a plan that is in action, as the bigger players in the industry are rolling it out. Jorge Salinger, Comcast's (NASDAQ:CMCSA) vice president of network architecture and the panel’s moderator, said the MSO's transition is underway. Currently, he said, there are three ongoing tests of integrated CCAP and three of modular CCAP.As the names imply, modular CCAP is a staged approach in which different elements are logically stitched together. In this approach, full CCAP operation in which video edge QAMs transition to CCAP come later. The integrated approach involves a single CCAP chassis, though the actual video elements can be added later.The challenges of CCAP aren't just technical. Salinger said a key difference is that formerly discrete data and video teams would in some way be combined in a CCAP world. This, and related operational and management changes, seem as big a part of what Comcast is doing now as working through the fundamental technology. "We realized more things change than remain the same," Salinger said.Time Warner Cable: Eight is EnoughThe significant reordering of how data and video is distributed makes timing a very important element. It was addressed by Mike Kelsen, the senior director of network architecture for Time Warner Cable (NYSE:TWC). Kelsen first laid out the fundamental deployment scenarios which, similar to the choice envisioned by Salinger, focused on a phasing in of CCAP video functions after data or a jump from current to full CCAP, or at least the introduction of the underlying chassis.Kelsen quite specifically tackled the vital issue of when it makes sense to move to CCAP, at least from the data perspective. He suggested that the transition begins to make financial sense when more than eight channels in a serving group must be dedicated to support DOCSIS functions. That crossover point at which CCAP makes sense is fairly low because vendors are offering aggressively favorable licensing terms in order to jumpstart the market. A complementary driver of early crossover is the natural inclination of cable operators not to purchase equipment they know soon will be antiquated.Kelsen presented important insights into some of the operational issues in CCAP. For instance, there tend to be more CCAP serving groups than in CMTS-based environments. This could lead to a bit of mixing and matching in the combing network. The sense from Kelsen's remarks is that proper planning will keep this from being a significant issue. A somewhat more important decision is whether to employ a non-routing approach. In this approach, IP routing tasks are handled by routers elsewhere. This leads to somewhat smaller CCAP boxes.Cox: Pitching CCAP to C-Level ExecsTop MSO engineers such as Salinger, Kelsen and Jeff Finkelstein, the executive director of network architecture for Cox Communications – are happiest when discussing bits, bytes, modulation schemes and prevention of packet loss. In the real world, however, they reached the upper levels of their respective companies by finding a way to convey the meaning of all the technology to C-level executives whose responsibility is the bottom line.Finkelstein’s presentation was structured in this context. Cox has opted for a modular approach to CCAP migration. He said this pushes the number of channels dedicated to DOCSIS functions to 16 or 24 before CCAP becomes an obvious choice. He presented CCAP as a potent answer to questions the check writers are sure to ask. As cable companies add functionality, power-hungry equipment - in the form of more lasers and amplifiers - must be added. This leads to realignment of service groups. The challenge is that such moves require more substantial headends that have greater HVAC and space demands. In some cases, new facilities are required.This, needless to say, isn't want the financial folks want to hear. It is, however, the perfect opening for the CCAP pitch. The very thing that the platform is intended to provide - a consolidation of boxes - can alleviate the worst impact of the growth. "[This can be a way] of pitching this to the check writers, who can see head space savings, increasing capability and potentially reduced node splits in the short term .... It can reduce the heating, cooling and all the HVAC issues in the headend,” Finkelstein said.ARRIS: Features and RoadmapsBreakfast host ARRIS was represented on the panel by Dr. Tom Cloonan, the chief technical officer for the company's networking group. He suggested that the path for an operator depends on the point from which it starts and that the companies making the gear have to be responsive to their clients' needs.It seems that the vendor battle will come down to three essential issues. Vendors must create gear that is extensible to standards and protocols that CCAP will grow to encompass, such as DOCSIS 3.1, Ethernet Passive Optical Networks (EPON) and inclusion of the actual optics, which will remain separate in early iterations of CCAP. A second differentiator for vendors will be will be the order in which functions are added to the CCAP chassis. Finally, equipment makers will have to decide which non-CCAP features, such as redundancy, are offered.The road to CCAP is a long one. But, like rest stops along the highway, incremental opex reduction is possible. The bottom line consistently running through the presentation was that many roads lead to CCAP. Said Cloonan: “In most cases, the path depends on the starting point.”Carl Weinschenk is the Senior Editor of Broadband Technology Report. Reach him at [email protected].