Why the 10G Push Is Stuck in Neutral
New technologies, shifting standards leave cable operators reluctant to put the pedal to the metal
The cable industry is billing its big 10G initiative as the next big technology star. But given the current state of the cable access technology business, is it even ready for primetime?
Cable’s counter to the wireless business’s all-encompassing 5G hype bonanza, 10G actually incorporates many different technologies, all aiming to extend hybrid fiber coaxial (HFC) networks beyond speeds of 10 Gigabits per second and into lifespans stretching out another two or three decades.
But within that big 10 Gbps bucket is a lot of new, disruptive technology for cable operator engineers to digest, some of it competing with each other.
For starters, the shift to Distributed Access Architecture, in which headend and network functions are virtualized, has been described as the biggest technological change the cable industry has made since it migrated to HFC back in the early 1990s. Indeed, transitioning from a centralized Converged Cable Access Platform (CCAP) architecture to DAA is proving to be a highly complex process, affecting everything from core plant operation to the skill sets required from workers. “The migration from centralized access architecture to DAA has certainly proven to be more complex when compared to prior upgrade cycles,” said Sean Welch, vice president and general manager of Cisco’s Cable Access Business.
Related: Cisco Access Chief Welch: 'Don't Wait' to Transform Network Infrastructure
The shift to DAA is but one — admittedly huge — piece of the engineering puzzle. As chief technology officers and their teams reimagine their HFC networks for the futuristic era of autonomous cars, they’re also mulling pressure to move away from large proprietary hardware appliances, such as the cable modem termination system (CMTS), and move the work these devices do to off-the-shelf servers that will virtualize these processes.
As if that weren’t enough disruption, cable technologists are split as to which next iteration of DOCSIS (the standard that enables high-bandwidth data transfer over HFC) will best deliver 10G’s bandwidth promise. Led by Comcast, the industry had seemed to be lined up nice and neatly to adopt CableLabs’s Full Duplex DOCSIS standard (FDX). But due to a variety of issues, some operators are pushing FDX aside in favor of alternatives such as Extended Spectrum DOCSIS and Low Latency DOCSIS.
NEXT TV NEWSLETTER
The smarter way to stay on top of the streaming and OTT industry. Sign up below.
The level of complexity and disruption operators face is certainly “not trivial,” Todd Kessler, senior vice president of product at CommScope, conceded. Kessler’s company acquired Arris, cable’s biggest technology vendor, for $7.4 billion this year. At the SCTE/ISBE Cable-Tec Expo in New Orleans this week (Sept. 30-Oct. 3), Kessler said, CommScope will seek to position itself as a “strategic adviser” to its operator customers, “giving them all the facts they need to make good decisions.”
A No Good, Very Bad Year for Vendors
Certainly, for cable access vendors, some decisions would be nice.
Sales were down 38% sector-wide in the first quarter, according to research company Dell’Oro Group, which attributed much of the stalling to cable operators weighing DAA-related decision-making. And Dell’Oro on Sept. 25 published a report which found cable access equipment sales were down 40% in Q2. The demand just isn’t there right now, the firm said.
”I think we all have to keep in mind that many of the largest operators have been on a consistent, six-year cycle of purchasing and deploying capacity as they moved from DOCSIS 2.0 to 3.0 and 3.0 to 3.1,” Dell’Oro senior analyst Jeff Heynen said. “At some point, that cycle had to stop, even for a little bit, for the operators to deploy all that capacity.”
And stop the cycle has.
CommScope has let go most of Arris’s top-level management, including ex-CEO Bruce McClelland, with its network and cloud division reporting a 37% revenue slide to $344 million in the second quarter.
Cisco Systems, the second biggest cable-access supplier behind CommScope, doesn’t break out numbers for that sector. But it recently put a halt to the development of Full Duplex DOCSIS, and key executives, including former cable-access strategy chief John Holobinko, have left the company.
Revenue at Casa Systems did rebound 47% from a really tough Q1. But its $52 million in second-quarter sales were still off around 28% year-over-year.
Harmonic, which is disrupting the hegemony enjoyed by CommScope, Cisco and Casa with its virtual CCAP product, CableOS, saw revenue fall 14% in the second quarter. This was despite customer wins for CableOS that included a $175 million multiyear deal with Comcast, and another $55 million contract with an unnamed European operator widely believed to be Liberty Global. In the latest news, Harmonic said Comporium Communications in Rock Hill, South Carolina, had signed on to use CableOS in a new DAA configuration supporting DOCSIS 3.1 services.
Still, Kessler doesn’t see the overall cable access business perking up until next year. “We thought more operators would be past the evaluation phase by now,” he said.
Notes from rival cable access vendors jibe with this prediction.
“While we do expect MSOs to increasingly redirect investment into Distributed Access Architecture, we continue to believe that we haven’t yet reached an inflection point away from the industry-wide pause in cable spending,” Casa Systems CEO Jerry Guo said during a second-quarter earnings call in late July. Observing the industry from a more detached perch as CEO of Pineville, North Carolina’s Chamber of Commerce, Holobinko said he believes key issues regarding HFC aren’t being addressed by the new 10G tech.
“I agree there are too many technologies and all seem to be lacking the one thing that a mature industry needs — a solution to high operating costs,” the former Cisco technologist told Multichannel News. “The biggest challenge with cable access networks is, when compared to fiber and wireless networks, the day-to-day cost of operating and maintaining the cable plant is an order of magnitude higher than competitive access technologies.”
With this in mind, might cable operators start to seriously look at ditching HFC and going with fiber-to-the-home? There would certainly be operational efficiencies. The transition doesn’t come cheaply, though, as shown by Altice USA’s $5 billion FTTH initiative in its Optimum footprint.
While 2018’s Cable-Tec Expo in Atlanta was full of tour de force white-paper presentations and demonstrations of FDX, Extended Spectrum DOCSIS and Low Latency DOCSIS, vendors seem to be approaching this year’s New Orleans event with compromise solutions aimed at taking some of the pain points away.
For several years, the FDX standard from industry consortium CableLabs was widely viewed as the next iteration of DOCSIS, a standard that would someday delivery symmetrical 10 Gbps speeds over HFC networks.
But FDX presents certain inflexibilities. Notably, it requires an expensive “fiber deep” strategy, with no amplification permitted between the node and customer. Despite the support of Comcast, FDX has lost the customer volume needed for wide-scale vendor support, as evidenced by Cisco’s pullback over the summer.
CommScope will be demoing what it calls “Extended Soft Frequency Division Duplex,” which combines elements of both FDX and ESD. Rather than dictating symmetrical bandwidth, the technology allows operators to move bandwidth from downstream to upstream and vice versa, as desired.
Extended Soft FDD is more “amplifier friendly” than FDX, said Chris Busch, CommScope engineering fellow in the office of the CTO. “That’s a big boon to the operator,” he said, describing a technology being put into another new bucket of cable industry jargon, “DOCSIS 4.0.”
What is DOCSIS 4.0 to Busch? Like 10G, the term seems to be broad.
“I think it’s the next generation of DOCSIS performance,” he said.
Beyond easing pain points for emerging DOCSIS technologies, CommScope will also try to light the path for faster uptake of DAA.
Remote PHY refers to the technique of moving the physical layer of electronic circuitry (the “PHY”) out of Converged Cable Access Platform at the headend and putting that PHY circuit at the end of the network. Remote MAC-PHY means also moving the Media Access Control layer to the edge, as well.
Converting to this scheme represents a massive shakeup.
“Many operators are still in the evaluation phase, looking at ways to deploy Remote PHY or Remote MAC-PHY,” Kessler said. “Only a handful are in deployment. And that will likely stay consistent into next year.”
Rather than showing off some splashy new performance-driven DAA gear, CommScope’s product demos at Cable-Tec Expo seem to be focused on tools that drive down costs. DAA Aggregator, for example, lets operators connect up to eight HFC nodes to one Remote PHY device (RPD). This significantly increases the number of homes passed per RPD serving group — from a range of 20-40 to 150-300.
“It’s more capital-efficient,” Busch said. “It allows operators to gain the benefits of Distributed Access Architecture without as much investment.”
We’re Just Not There Yet
While the cable technology business’s recessionary trend certainly has a lot to do with the complexity associated with new standards and their adoption, Dell’Oro analyst Heynen believes that, despite all the hype, 5G isn’t — at least not yet — putting on the kind of pressure needed to make leading operators forget how much they just spent on upgrading to DOCSIS 3.1, while enabling 1 Gbps speeds that few consumers are using.
Is it too early to make 10G the star of the show?
“I don’t think spending improves significantly until [operators] feel there is a viable competitive threat to their 1-Gig DOCSIS 3.1 services, or an application or service [emerges] that challenges the capacity they have in their access networks now,” Heynen said. “That, plus slowing bandwidth consumption rates, are the reasons for the current trough in spending. I do see spending improving in the second half of the year, but certainly nowhere near the levels seen in 2018.”
Daniel Frankel is the managing editor of Next TV, an internet publishing vertical focused on the business of video streaming. A Los Angeles-based writer and editor who has covered the media and technology industries for more than two decades, Daniel has worked on staff for publications including E! Online, Electronic Media, Mediaweek, Variety, paidContent and GigaOm. You can start living a healthier life with greater wealth and prosperity by following Daniel on Twitter today!