The first certified G.fast products are due to hit the market in the next few months, but in the meantime there's more work required just to ensure that different G.fast products can all work together in the same network. That work includes an imminent plugfest designed to verify some of the nitty-gritty details of interoperability among different G.fast chipset implementations.
The plugfest is being conducted under the auspices of the Broadband ForumBroadband Forum the week of January 23 through 27. The University of New Hampshire InterOperability Laboratory's G.fast Consortium, which is centered on University of New Hampshire InterOperability Laboratory (IOL)'s testbed, will host.
Different pieces of equipment based on the different chipsets have been demonstrated to be able to communicate with each other, but there are a number of G.fast features that have to be implemented similarly, so that G.fast can be confidently mixed-and-matched when installed in commercial G.fast deployments.
One example, said Lincoln LaVoie, senior engineer at UNH-IOL, is fast rate adaption. FRA enables rapid reconfiguration of data rates in response to abrupt changes in channel or to compensate for the introduction of noise.
If the various chipset vendors have interpretations of FRA, or implementations of FRA, that vary significantly from each other, that could lead to unexpected difficulties in a G.fast network that uses equipment from multiple vendors.
The plugfests are not only helping the vendors all get on the same page, they're helping UNH-IOL prepare for its certification duties. The certification process was scheduled to start at the beginning of this year. LaVoie said no products have been officially certified yet, but the official process should begin shortly. Everything about the certification process will be confidential -- including the identities of the participants -- until actual certifications are announced, he said.
The list of participants in the upcoming plugfest includes the G.fast chipset vendors Sckipio Technologies, Broadcom Corp., Triductor Technology and Ubilinx. Intel Corp. has also signed up; it is an investor in, and partner with, Sckipio.
Ubilinx is Realtek Semiconductor Inc. , a Taiwanese semiconductor company that has participated in previous plugfests with its G.fast chipsets. Triductor, headquartered in China, is an IC design house that has supplied VDSL chips and other communications devices mostly to OEMs in its home market.
Arris Group Inc. and Technicolor (Euronext Paris: TCH; NYSE: TCH), both suppliers of customer premise equipment and other networking systems, are also participating.
Want to know more about the companies, people and organizations driving developments in the test, monitoring and assurance sector? Check out Testapedia, the most comprehensive online resource covering the telecom test and measurement industry.
Many of the companies that signed up for this plugfest also participated in an interoperability demo at the Broadband World Forum in London late last year. (See Interop Demo Pushes G.fast Forward.)
Companies with ongoing participation in the series of plugfests that won't be at this particular event include chipset vendor Metanoia and test companies Telebyte and Digital Lightwave (VeEX).
Keith Kelley, SVP and GM of telco and satellite CPE at Arris, told Light Reading in an email exchange: "We are looking forward to collaborating with companies throughout the networking industry to make G.fast interoperable across various equipment, which is a vital step in making G.fast technology truly mainstream. In participating in UNH-IOL's plugfest, we are investing in providing our customers with high-speed, high-quality connections, regardless of the vendor on the other end of the wire."
Chris Dunford, product manager of the access division at Exfo, provided Light Reading with the following comment: "As the World's first handheld G.fast test set vendor, Exfo strongly supports the initiative and opportunity put forth by the Broadband Forum's G.fast Chipset & System Integrator Interoperability Plugfest at the University of New Hampshire's Interoperability Lab (UNH-IOL). It's events like the G.fast Plugfest that propel the G.fast technology towards the goal of a globally ubiquitous technology. The event proves a collective commitment towards interoperability from the industry with a goal that benefits operators and most importantly end subscribers who continually demand ultra bandwidth services delivered to their premises."
— Brian Santo, Senior
Editor, Components, T&M, Light
Reading
CALLING ALL TEST, ASSURANCE AND MONITORING COMPANIES: Make sure your company and services are listed free of charge at Testapedia, the comprehensive set of searchable databases covering the companies, products, industry organizations and people that are directly involved in defining and shaping the telecom test and measurement industry.
brooks7, User Rank: Light Sabre 1/17/2017 | 4:08:39 PM
Re: G.fast in the USA @Brian Thanks for Clarifying.
And that has added to the reason that the more advanced versions of DSL have gotten to be smaller and smaller markets. ADSL (and 2 and 2+) were designed to operate within the cable lengths that the Telco networks used in their POTS design. Once you need shorter and shorter loops you have to add more and more fiber to the network. As you do that, suddenly a full FTTP build becomes more and more palitable. Think about AT&T and U-verse. If they did g.fast and eventually FTTP, they will have done 4 buildouts in access over that period of time (and I got one of the first ADSL lines from them in 1999). That could mean 4 generations of technology in one depreciation cycle.
There is an MDU problem as well. Now we would require DSLAMs in MDU basements and that the building owner allows telco access and power for them. They are not mandated to allow such access.
All of that leads to these more advanced technologies having smaller markets that they can address. Tier 1 Telcos want property wide solutions. Or at least something that is deployable across many lines. The cost to approve, deploy and maintain these sub-scale technologies just makes no sense - outside of high density housing in Asia.
The telcos have latched on to Wireless because they can sell the same bits per second for a lot more money AND they get to charge per person instead of per household. Now they are just running into families not being able to spend more on their services.
Re: G.fast in the USA Why would you replace FTTP with G.fast? You wouldn't even consider it.
That's what I meant -- even if VZ wanted to upgrade wireline (which it doesn't), it's been doing FTTP, which is why it wouldn't likely even consider G.fast. Sorry if that wasn't clear.
brooks7, User Rank: Light Sabre 1/17/2017 | 2:32:15 PM
Re: G.fast in the USA @Brian Why would you replace FTTP with G.fast? It would make 0 sense. FTTP is faster. The whole point of G.fast is to not do the drop cable with fiber but instead use the existing copper. But you have to bring the fiber into the local 8 home (or so) crossconnects. That is the fiber rebuild required. Then there is the whole power thing. People have been talking about back powering the DSLAM from the homes (I just worry about somebody unplugging their modem and crashing the entire DSLAM). Other than that, you now have to place a LOT of power nodes or extend it out from the CO. That is why FTTC (which is what g.fast is) has only happened in 3 places - East Germany, BellSouth and Las Vegas. The East German network got overbuilt by standard DSL. BellSouth and Vegas are the old Reltec - Marconi units and the carriers have had difficulty with replacements.
@Kbode - Tom Nolle declared it an absolute loss. I don't believe it. I think that they see DSL upgrades as a bad deal compared to other investments. Unless we force upgrades legally, I would expect them about never.
Re: G.fast in the USA No, it's roughly 300 Gbps at 300 meters -- and that's if you aggregate multiple lines.
Nokia, for example, says explicitly that you can't do G.fast without also doing a fiber rebuild because you need to move your DSLAMs closer to the prem (presumably you'd need more DSLAMs then, too?).
https://networks.nokia.com/solutions/g.fast
There's a chart on that page that shows how G.fast can in fact do 1 gig at very short distances, but that the data rate falls off on a fairly steep slope measured against distance (a well-known phenomenon with DSL).
Neither AT&T nor Verizon have any stomach for such massive rebuilds. And even if Verizon were to change its mind and spend a lot more money on wireline, it would have to consider FiOS, which is FTTP. As it is, both seem to be legitimately vastly more enthusiastic about wireless broadband.
Which leaves Centurylink, which, yes, is mostly isolated in the upper left-hand corner of the country, which, for all intents and purposes, can be adequately described as "Idaho" (speaking as someone who resides in a nominally non-Idaho part of that corner).
So in short, mostly Europe, some parts of Asia,and Idaho right now.
KBode, User Rank: Light Sabre 1/17/2017 | 12:59:34 PM
Re: G.fast in the USA Makes since it's more common overseas given the dominance of DSL in other markets. But outside of CenturyLink (in I think Idaho) it's still curious it's not seeing more adoption here. I had read I thought that they're at the point where they can deliver 300 Mbps at a thousand meters, which still sounds pretty promising for a lot of these MDUs?
KBode, User Rank: Light Sabre 1/17/2017 | 12:56:03 PM
Re: Broader deployment Did he me mean that growth in residential broadband is flat so ISPs don't think it's profitable enough (and are therefore refocusing on media, ads and enterprise)? Or did he say broadband simply isn't profitable? I'll go try to read the article...
inkstainedwretch, User Rank: Light Sabre 1/17/2017 | 12:08:48 PM
G.fast in the USA G.fast offers something like 300 Mbps up to 300 meters. A decade ago, loops that short were rare in the US. I haven't seen the stats in years, but I assume connections that short are still rare.
What I've been hearing consistently since the technology was first developed was that G.fast would start out being most attractive in markets where physical access is restricted either by statute or by the physical limitations of the premises, or both, which describes markets in Europe.
Not to say G.fast won't be used in the US. It's just that most vendors still seem to think the biggest early markets will be elsewhere.
brooks7, User Rank: Light Sabre 1/17/2017 | 11:45:04 AM
Re: Broader deployment We had an article last week here quoting Tom Nolle that Broadband Internet loses money (I don't believe it). Why would people spend money on the service given that?
KBode, User Rank: Light Sabre 1/17/2017 | 9:12:50 AM
Broader deployment I would really love to see broader deployment of this. I've heard about technology like this for a decade, (crosstalk reduction, etc) but it never seems to see any substantive deployment here in the States where it would be the most useful in shoring up lagging networks.
Triductor Triductor was founded in 2006, and has been supplying VDSL and G.hn circuitry to Chinese OEMs. (Thnx to Dave Burstein for pointing me to this info).
You might think Amazon's Unlimited is just another me-too streaming music service. You'd be wrong. If successful, it will be a critical tool to help slice off a fat, juicy chunk of Google's $75 billion ad business.
To save this item to your list of favorite Light Reading content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
(10) | 
Digg
Del.icio.us
Reddit
Tweet This