Fueling the Distributed CCAP Debate

Even though the industry has just started installing Converged Cable Access Platform (CCAP) devices in their headends, cable engineers are heating up the debate over how to split up that equipment and move at least some of its functions to the access network. And that debate shows no signs of cooling off soon.

At the SCTE Cable-Tec Expo show in New Orleans last month, this topic, now most commonly known as Distributed Access Architecture (DAA) or Distributed CCAP (DCA), dominated both the agenda in the technical workshops and the discussion on the exhibit floor. As mentioned in the previous post in this series last month, such prominent players as Casa Systems Inc. , Cisco Systems Inc. (Nasdaq: CSCO), Gainspeed , Harmonic Inc. (Nasdaq: HLIT) and Huawei Technologies Co. Ltd. all demonstrated their versions of distributed CCAP systems with DOCSIS 3.1 capabilities. (See Sorting Out Cable's Gigabit Strategy.)

For instance, Huawei showcased its D-CCAP solution at its "Gigaband Network" booth on one end of the convention floor. A bit more radical than most of the solutions on display, this approach calls for shifting all of the CCAP equipment and functions from the headend to the network node, placing optical line terminals (OLTs) in the headend and using GPON/10GPON technology to transmit signals over the fiber portions of the cable hybrid-fiber coax (HFC) network. It also calls for cable operators to switch out their existing analog fiber nodes for new digital fiber nodes. (See Huawei Plugs Distributed CCAP.)

In two video interviews with Light Reading at Cable-Tec Expo, senior Huawei executives Allen Wang and Jack Moran argued that the company's D-CCAP approach offers the best way to maximize the gigabit potential of the new DOCSIS 3.1 spec and migrate eventually to an FTTH network architecture. Wang noted that Huawei is now in D-CCAP trials with an undisclosed cable operator in North America, with plans to make the new platform generally available next year. That move comes as Huawei also enters the DOCSIS 3.1 market, with the first public deployment of its D3.1 equipment just announced by Vodafone New Zealand . (See Progress of DOCSIS 3.1 in the Cable Industry and The Challenges Facing Deployment of DOCSIS3.1.)

Huawei's numerous competitors, though, are just as passionate about their own distributed access approaches. For example, Casa Systems executives are now pushing hard for their more incremental Remote PHY approach, contending that it makes the most sense as the first distributed step for cable providers.

The three leading DAA options right now include: Remote PHY, which calls for separating the PHY circuitry or chip from the CCAP core and moving it to the optical node in the access network; Remote MAC-PHY, which calls for moving both the PHY circuitry and the CCAP device's media access controller (MAC) to the network node; and Split MAC, which calls for moving all the PHY circuitry and some of the MAC functions to the node but retaining the rest in the headend. And while Huawei's approach is a version of the Remote MAC-PHY option, it's not even the only version in that camp.

Although each of the Distributed CCAP options and vendor solutions is different, they all share many of the same benefits, which include reducing the space and power requirements in the headend and supporting more wavelengths on digital fiber to ease cable's move to Fiber Deep deployments and more fiber nodes. The shared benefits also include boosting the critical signal-to-noise ratio on the HFC plant, thereby improving data transport speeds, which is especially key for DOCSIS 3.1.

With so many options available, many cable operators are now scratching their heads in confusion over which approach to try. Not surprisingly, then, no industry consensus has yet emerged as to which option would be the best one to take. Anybody want to be a referee?

"It's too early to tell which one will ultimately win out," said Tom Cloonan, CTO of Network Solutions for Arris, noting that all three leading DAA options have their distinct pros and cons. "It's a confusing problem for MSOs to look at."

Seeking to clear up that industry confusion, Arris engineers recently conducted a study comparing and contrasting the three approaches, coming up with the 26 features most important to cable operators and then scoring the competing architectures on each attribute. But even this detailed study did not produce truly definitive results, concluding that all the options are workable solutions. Accordingly, Arris is now pursuing the development of both Remote PHY and Remote MAC-PHY products for deployment in late 2016 nor early 2017.

"We're kind of leaving it open to go either way," Cloonan said. "We're sitting on the fence a little bit."

So cable's Great D-CCAP Debate will likely only intensify over the next year. Stay tuned for more.

— Alan Breznick, Cable/Video Practice Leader, Light Reading

This blog is sponsored by Huawei.

COMMENTS Add Comment
inkstainedwretch 11/13/2015 | 2:34:09 PM
Confuzzlement A couple thoughts on the subject I've heard recently:
  • From a pure IP standpoint, it makes absolutely no sense to NOT put the MAC out in the field.
  • From an operations standpoint, you'd have to be nuts to put the MAC out in the field where just about anybody, from inept technicians to vandals, could bungle it. 

-- Brian Santo
Sign In