BT demonstrates 10Gbps on normal fibre


news British telco BT has demonstrated that it is possible to deliver broadband speeds of up to 10Gbps over its normal fibre infrastructure extending to some homes and businesses; the same Fibre to the Home infrastructure which is being deployed in Australia as part of the Federal Government’s National Broadband Network project.

BT, which is the UK’s former monopoly telco (similar to Telstra in Australia) is currently rolling out fibre to the node infrastructure in the UK as part of plans to significantly upgrade the country’s existing copper network to deliver speeds of up to 80Mbps. In addition, the telco recently announced that it would be allowing customers to choose to have fibre fully extended to their premises, delivering a large speed upgrade to 330Mbps. However, overnight the telco revealed the results of a trial which promised even higher broadband speeds that would be possible for widespread use in the future.

In a statement, BT noted that it had conducted a live trial of 10Gbps speeds on its normal fibre network, using Cornwall-based engineering firm Arcol UK as the testsite for the new technology. The 10Gbps service actually runs in tandem with the company’s existing 330Mbps service on the same fibre, and merely requires different networking equipment connected to the customers’ premises. The trial uses new fibre technology called XGPON (Tens of Gigabits on a Passive Optical Network) and was supplied by Chinese networking giant ZTE, in partnership with BT’s Openreach wholesale infrastructure division.

Ranulf Scarbrough, Programme Director for the Cornwall SuperFast Broadband Programme, said: “What is exciting about this trial is that these hyper-fast speeds have been obtained over the exactly the same fibre that carries BT’s fibre broadband services today. All we are doing is changing the electronics at either end. This trial shows we are thinking and ready for the future even though there are no current plans to deploy this technology. A lot of this project is about future proofing – making sure that it’s not just the fastest speeds today but that we can continue to be at the cutting edge for five, ten, twenty years.”

The 10Gbps trial runs over high speed fibre optic network established by the Superfast Cornwall Programme, a broadband partnership between the EU, BT and Cornwall Council, which has made fibre optic broadband available to over 100,000 Cornish homes and businesses.

Until recently, Arcol’s 40 staff had shared a 1.5Mbps internet connection, according to BT. The Superfast Cornwall Programme has delivered high speed fibre-to-the-premise to the business park enabling Arcol to connect at 330Mbps. Alun Morgan, technical director at Arcol, said the ability to connect at such fast speeds was “opening the door” for the company to achieve much more.

“We are still only just discovering the sorts of things we can do with these speeds, such as taking advantage of services like videoconferencing and using a cloud-based ERP system so we can access this information elsewhere, and it has enabled us to be much more efficient and aggressive,” Morgan said.

BT’s trial does much to demonstrate how the future of next-generation broadband in Australia could evolve. In Australia, the Federal Government is currently deploying fibre to the premises as part of its National Broadband Network project; a more ambitious style of deployment than BT as the fibre is being extended all the way to most premises nationally. The company deploying the infrastructure, NBN Co, has said that speeds of 1Gbps will be achievable over the network in the short-term future, but it is likely that speeds of 10Gbps will eventually be possible over the network, as the BT trial demonstrates.

The Coalition currently has a policy featuring a very similar network rollout style to BT, featuring a Fibre to the Node-style deployment, with the possibility that such a network would be extended in a similar fashion to the way BT is currently deploying fibre all the way to the premise in some areas. It is likely that any FTTN-style deployment in Australia would eventually be upgraded to fibre in many areas over the next several decades, should demand for Internet bandwidth continue to increase at the same speed it is currently.

What BT’s trial shows is the future of both the NBN policies espoused by the Coalition and Labor. If Labor’s NBN vision continues to be rolled out following the next Federal Election, it is likely that we will eventually see the NBN’s networking equipment upgraded to support speeds of 10Gbps, especially in areas where business and government is concentrated. If the Coalition’s FTTN policy goes ahead, it seems inevitable that fibre will eventually be extended all the way to premises in many areas, again, especially in areas of high demand where business and government exists, and that the same 10Gbps speeds will eventually be in the offing, but perhaps much further down the track.

Will we need these speeds? The past several decades of trends in networking and bandwidth demand shows that it is very likely. And in any case, as the networking equipment to provide these speeds becomes increasingly commoditised, there will eventually be few reasons not to deliver these speeds. We saw what happened with copper networks around the world; they were eventually built to provide telephone calls with manual – manual! – switching of connections. Eventually the end point equipment was upgraded to provide unthinkable bandwidth speeds; up to 24Mbps as is common under the ADSL2+ standard. It is inevitable that this same upgrade cycle will occur with fibre – whether it starts off as fibre to the home or fibre to the node.

Image credit: Clix, royalty free


  1. But… but… I thought this technology was going to be obsolete by the time it was built? Do you mean the Libs have been lying to me? *boggle*

  2. Technically not just lasers, Frikkin’ Lasers, attached to the head of a miniature shark.

  3. You didnt emphasize that if we do have to use those speeds we MUST to go from FTTn(Coalition policy) to FTTH(Labor policy). FTTn will just not be able to deliver more than 80mbps. So this just prooves that Labor’s policy is much better for our future needs. Also if we do go to FTTh from FTTn, all those nodes will be a big waste of taxpayer money.

    This is all assuming that the Liberals dont scrap the NBN like Mr Abbot says.

    • Yes with the FttN plan, the curbs and nature strips will be dug up for the next five, ten and twenty years. Every time someone decides they need more speed now, they have to replace their copper with fibre, with who footing the bill for the works carried out in the adhoc single premesis manner?

    • porn and illegal movie & music downloads most likely.

      that’s all the internet is used for anyway.

      • Don’t forget gambling. That’s the only other thing high-speed internet is used for according to that greatest of all tech-heads*, Mr Richard Alston. They were probably downloading gigs of smut while gambling at 10 different casinos at the same time.

        *Did I say tech-head? I meant freaking luddite.

        • Gambling as consumer of high-speed internet? What sort of gambling was he thinking of? Maybe he was confusing it with high-frequency stock trading…

  4. “The 10Gbps trial runs over high speed fibre optic network established by the Superfast Cornwall Programme, a broadband partnership between the EU, BT and Cornwall Council, which has made fibre optic broadband available to over 100,000 Cornish homes and businesses. ”

    BT noted that it had conducted a live trial of 10Gbps speeds on its normal fibre network, using Cornwall-based engineering firm Arcol UK as the testsite for the new technology. The 10Gbps service actually runs in tandem with the company’s existing 330Mbps service on the same fibre, and merely requires different networking equipment connected to the customers’ premises. The trial uses new fibre technology called XGPON (Tens of Gigabits on a Passive Optical Network)

    FTTH service, using GPON. Not fibre from a FTTN Node

  5. Wait, what???

    BT hasn’t even finished it’s FttN roll-out and it’s already offering FttH AND looking at how to upgrade that?

    I’m starting to think FttN might have been a good idea if Telstra it’s self had done it many years ago using it’s own copper, but even they didn’t seem to think so in more recent times ( and they were planning that pre-NBN…)

    Seriously, is Malcolm trying to sell us a white elephant even before his NBN even starts??? Most other countries/companies seem to think so :/

      • Some folks won’t even see it now, after having it spelled out to them NBNAccuracy ;o)

        I love a good irony, and the fact that Malcolms poster child FttN roll-out is (even as he points to it as one of _the_ great examples of FttN success), switching to FttH while he does so is just golden…

  6. I love that last point. As technology progresses and the equipment gets cheaper we stop asking why and instead ask “why not?”.

    I remember almost a decade ago when 80GB HDDs were just starting to appear and I wondered what you’d do with that much space. Then they got cheaper and now pretty much everyone has a couple of TBs floating around somewhere. Same with flash, ten years ago the idea of a 240GB SSD that gets reasonably close to the speed of SDRAM would have seemed ridiculous. Now some of us wonder how we managed without it.

    ………… that said I’m not going to get too excited about the idea of 10Gbps ‘nets when most of us still have <100Mbps between us and the router :)

    • My first modem was 1200/75 baud (thats 1200 characters per second down. A good touch typist could actually type faster than the thing would send the 75 characters per second up!!!).

      • Actually, that was 1200/75 _bits_ per second (not characters). With a start bit and a stop bit, it could download at 120 characters per second and upload at 7.5. Note that at this speed it could take 17 seconds to refresh a text-only screen, or several minutes if it included many colours (ANSI escape codes). My first modem was 2400bps: I remember some BBSs were painful on that near the end, but it was magical to connect to another machine.

        If you wanted to be super-technical it was 600 baud (since baud is a measure of symbols per second and if you had 16 possible symbols you could transfer 4 bits per symbol) I’ve just been reading and your modem probably was V.23 which actually was 1200 baud: AFSK encoding and not my modem’s QAM.

  7. Point of order Renai. Turnbull has admitted that FTTN wouldn’t be updated for a number of years.

    FTTN is limited to the last mile speed. Thus 10Gbps is pointless when the tail is likely expected to meet a target speed of 12Mbit (via ADSL2+, VDSL isn’t certified for broad deployment, iirc).

    Even if Turnbull’s ideas take flight, expectations around delivered speed are still very mich low to mid two digit (MBit) figures.

    This is the thing to remember – these sorts of awesome ludicrous speeds are irrelevant to an FTTN network.

  8. < Insert fact-backed claims that “wireless is superior to fibre; besides, who’d want that old slow stuff anyway?” here >

  9. Im really not sure why BT think this is a big deal???

    When I was Customer Delivery Manager at Adam Internet I managed multiple 10Gbps Fibre deployments for customers using industry standard of the shelf components and single pairs of fibre – this is not that expensive or difficult (relatively speaking).

    • I suspect its fancy because it’s based on GPON, not straight fibre deployment.

      GPON being what NBNCo are rolling out. (and by the sounds of things, this was tested in BT’s GPON FTTH-only area of Cornwall, that sounds like it was rolled out with financial assistance from the Cornish local government.
      This sounds very much like the exact opposite of the coalition plan, and completely unrelated to BTs FTTN (and fibre extension) build.

      But, I mean I just got that impression from the article.

    • I think the point is that they’re running it on a (probably 32 split) PON, while an older PON network is still active on the same fibre.

      This isn’t a huge deal, but it does show that fibre delivers exactly what it promises.

  10. Daily Telegraph takeaway: faster pr0n!

    Alan Jones takeaway: faster trolls!

    Tony Abbott takeaway: I’m no Bill Gates…

  11. Hang on guys – with XGPON, that 10 Gbps is shared – shared across all the 32 – 64 endpoints on the PON.
    We don’t want no stinking shared bandwidth network that will bog down like cable does now – we need dedicated bandwidth like DSL provides (I’ve seen this argument, I kid you not! – apparently the backplane of a DSLAM doesn’t count).

    This XG-PON stuff is just a stepping stone to WDM-PON – a dedicated optical wavelength to each endpoint. 10 Gbps symmetric per user, although by then 40 Gbps and 100 Gbps optics might be cheap enough too.

    40 Gbps dedicated lambda per home – now THATS what I’m talking about.

    • I wouldn’t worry about it. NBN is using a split of around 32, that’s if everyone gets a network connection. Even then 32:1 is way better that any consumer grade ISP runs of the rest of it’s network anyway. It isn’t the bottleneck.

  12. Any chance you’ll change the title of the article to: “BT demonstrates 10gbps over GPON fibre” rather than ordinary fibre?

    This is not “ordinary” fibre, this is GPON optically split fibre. Ordinary fibre has been doing 10gbps+ for years. 10gbps over GPON is MUCH more interesting than 10gbps fibre.

    Just saying.

    • J-wireless is great, but it is the fantastic new technology of IP over Avian Carriers that is the most cost-effective and future-proof broadband implementation. As long as you don’t mind the latency… :D

  13. The trouble with this Renei, is that Coalition will continue with it’s FTTN/FTTC policy and will never upgrade again.

    The reason why they are doing something now is because of politics – nothing to do with Helping Australia getting off it’s feet.

    Oh in regards to the XGPON – that is down the upgrade path of the NBN (see the upgrade path in the original corporate plan – see Page 75).

    How much money would it cost the Goverment of the day if Coalition were voted in and rolled out FTTC?

    I’d say alot more money than NBN and then upgrade to XGPON technology.

    Perhaps Renei could have added this Corporate Plan graph to the original article?

  14. 10Gbps is an insane amount of bandwidth that no one will probably need for some time. Most corporate server networks aren’t even using 10Gbps switching yet, let alone the desktop LAN or internet connection. So don’t get too excited about this trial, It will be quite a while before 10Gbps becomes a cheap commodity. Though what it does mean is the FTTH is future proof. Maybe 20 years from now you might see 10Gbps being offered.

    • It is 10Gbps “shared” between up to 32 premises. That word in quotes because it’s not shared like a HFC network is shared, but PON networks have splits that put multiple users on a single optical path. NBN are currently rolling out 2.5GPON so 10GPON doesn’t really sound like a massive upgrade? I guess if it uses its own wavelengths some of those 32 will be on 2.5 and others on the 10, and they won’t affect each other.

    • Agreed. If FTTN gets rolled out, we’ll certainly be regretting it in 20 years… (OK we’ll be regretting it straight away, but it will be orders of magnitude more stifling in 2 decades).

    • Ron, 10Gbps fibre WAN links are already commonplace, I managed the rollout of 3 of them in 2011 and that was just in little old Adelaide!

      • I wouldn’t say it’s common place. Yes it’s there, but really only large enterprise can afford them. Once medium-large business can afford them then pretty much you can say it’s common place. By then I’m sure large enterprise markets will have something faster.

  15. Enter Janet6, the next generation of the network. This new incarnation will take advantage of the latest optical transmission technology called “coherent optical transport” which will deploy a 100G backbone, and will be able to support the exponential growth in bandwidth needed by applications discussed here.

    A 100G network, at present, is the benchmark. Yet this network is scalable to 400G and beyond. It’s a dedicated network on par or surpassing many national telecommunications networks in terms of sophistication and speed, capable of transporting massive amounts of data long distances to multiple recipients economically and efficiently.
    These developments are made possible by the constant search for better, faster and more efficient methods of transporting large volumes of R&E information across the globe. Driven by commercial requirements, rather than scientific needs, the world’s communications providers will be able to take advantage of these innovations when needed because of the boundary-pushing work of research and education networks. One thing is certain: the future of mainstream broadband networks is already here. You just need to know where to look for it.

    • Transdimensional Rosen bridge tachyon timetravel FTW.
      I could be there in person before you finish reading this and smack you with a fish I pick up from the Atlantic on the way, and be back before you could blink.

  16. “Until recently, Arcol’s 40 staff had shared a 1.5Mbps internet connection, according to BT. ”

    This is the kicker that the ordinary Joe seems to ignore.
    “Oh, I don’t need 100Mbps internet. What ever would I do with it?”

    Yes the average home consumer may well be happy with 12Mbps or 24Mpbs connections, but Business cannot.
    Where your average business needs are increasing to depending much more on internet based commerce. From Banking, superannuation, logistics, accounting to name just the necessities for any business. Some of these increasingly moving to the Cloud layer.

    The problem is and has always been, that business are forced to pay so much more for the connection, that at times, are no better than the home consumer in terms of performance.
    This is where Australia will take a productivity boost, simply from shifting many of these businesses from unreliable copper connections, to fibre.

    • “Oh, I don’t need 100Mbps internet. What ever would I do with it?”

      One issue with “those” people, is they most likely live alone and only use “the internetz” for email.

      I share my home connection with two other people,over a cable connection which at times faces it’s own congestions issues. Even at 50Mbps, the connection quality at my place (to put it kindly) is “wildly variable”…

      • I dislike this argument a lot. “Those People” whose use of the internet is a little bit of email and not much more. Their use is just as valid as anyone elses, and should not be criticized in this manner I think. Not everyone is hipster enough to require fast toobs to the net. I do think this tends to be forgotten a lot in the whole NBN discussion.

        I certainly don’t think this is where someone like Alan “Frothing by 6am” Jones fits in, but there are a genuine proportion of the Aus populace who couldn’t give a rats about the NBN, and that is actually a valid viewpoint.

        Anyway, just randomly piping up :)

        • No, it is quite right to be critical of their argument. The argument that it is not needed because they don’t need it is a very selfish one. Some of those who have made the argument, most strenuously, that no one needs that sort of speed end up being a single person who already has a 100Mb connection.

        • I agree that some people will only ever use the internet for email. However, time and time again I have seen people who have said they never wanted something, when given the option, but appreciated it later down the track.

          One example is my dad told me off for “connecting my computer to the phone line” when I had access to a bulletin board via an old 2400 baud modem back in ’97. I used it for downloading some MIDI files, etc. However, the next year when my brother and sister went overseas to Cambodia, we got a 56k modem and signed up to an ISP for an internet connection. All of a sudden, my father was the one bragging to all his friends about getting emails that had been written just minutes earlier in another country.

          I’m sure if my dad was still alive today, he would be one of those people who only used the internet for email, and wouldn’t see the point of the NBN. But the NBN is a similar leap further a few years later. Chances are, he’d discover more uses as time went on.

          Also without the support of these light uses, the NBN would not be viable. After all, the money collected from all users, heavy and light, (which as it turns out, is comparable to what they are paying today,) goes mostly towards improving broadband (through NBN), rather than making one company (and their investors) rich.

        • ferretzor
          Quite agree, however the needs of the current “occupants” of a particular premises may well change over the coming decades, let alone the needs of the future “occupants”.
          The NBN is NOT being built for current or even the next decades needs only, but to be the Nations Communications platform for many decades to come. Ad Hoc upgrades on demand is extremely wasteful, inefficient and expensive, not only that but over time has the potential to create all sorts of network issues

  17. Actually what people fail to factor in also, was how much bandwidth they needed 10 years ago versus today and yet think they will never need more! Also with a growing number of devices becoming internet aware/connected the need for bandwidth to a household is increasing. In my house I’ve already got close to 10 devices on a network via wired and wireless. If just one of those devices is using the internet nothing else can access the internet at a reasonable speed.

  18. “it is likely that we will eventually see the NBN’s networking equipment upgraded to support speeds of 10Gbps”

    It will only require replacement of the terminal equipment for FTTP.

    For FTTN, it will require scraping of the node and everything beyond it, installing a GPON system, plowing up the streets and installing a FTTP system.
    This is of course assuming that the existing nodes are located where required for FTTP rollout. This is unlikely to be the case for every upgrade given their positioning based on teh copper network, so in many cases it will make more sense to scrap the entire FTTP system and rollout a new FTTP replacement.

Comments are closed.