Horses for Courses – picking the right tools for the fibre job!
April 25, 2012 in Adrian's tech blog
This blog started life on my Posterous page which I use for quick thoughts but the impacts have been troubling me so I decided to move it to my main page and add a little to it.
It started when I spotted this tweet from FiberNews, run by the excellent Marc Duchesne (If you don’t follow @mduchesn, then why not!):
“MikroTik RouterOS – Hardware suggestions for FTTH ISP bit.ly/IVK6v9“
Seeing it raised some big questions in my mind, and ones which I think are largely a UK specific issue and not one which may be of particular relevance to other countries beginning to fibre-up.
FttH is long-lasting national strategic infrastructure. At some point in the future there will be a copper switch-off and the fibre infrastructure left behind will become default telecommunications network in each country.
This isn’t a Mystic Meg prediction – I can’t say how long it will take but Regulators in some European countries are already starting to consider the conditions under which copper networks might be switched off.
When it does happen the fibre networks being built today become natural monopolies and will have the sole responsibility for delivering critical services and for supporting the market that relies on competition from wholesale operators through service providers to content organisations. All this will rest on the decisions being made today by the pioneering fibre operators.
Many countries around the world have active community-led broadband markets but there is a subtle difference in the UK. In my work with BDUK, I suggested three broad models for delivering community solutions:
- Partnership – where there is co-investment in the assets but the network is designed, built and operated by specialists
- Concession – where the community own all of the assets but a concession is offered to design, build and operate the network
- DIY – where the community own all the assets but also design, build and operate the network themselves
The UK, like many markets can find good examples of the first two, and this is the core focus of many international fibre markets, but I suspect the UK is alone in seeing the emergence of the third group.
In the context that any resulting fibre network will become the national strategic infrastructure, any undertaking by a community carries with it not inconsiderable responsibility and for any community considering a DIY approach this responsibility rests entirely on the shoulders of the community.
I should be clear here: There is absolutely nothing wrong with a community adopting such an approach and for a few this is something their communities will be willing and able to take-on, and I for one wouldn’t want to stop them – so long as they fully understand the responsibility they are taking on.
However, it was the tweet at the top of this blog that brought into sharp focus for me this sense of a community having a full understanding of these issues and what practical steps they need to take to assure themselves and their communities.
MikroTik is excellent kit. I’ve used it myself to build devices with features from pretty much all the Cisco catalogue from simple routers to deep packet inspection, intelligent traffic shaping and distributed load balancing devices.
MikroTik and similar kit formed the basis of the wireless network I built some years ago to deliver first generation broadband to homes and businesses in rural Oxfordshire. It allowed me to build features into that network that simply couldn’t have been cost-effective any other way, and in the same circumstances I’d do it exactly the same way again.
But I wouldn’t build a network the same way that will at some point in the future will become the network ultimately responsible for guaranteeing blue-light telephone calls or providing critical health-care services, or for sustaining the local leg of what has been identified as the largest and most vibrant on-line economy in the world.
There is a fundamental difference between deep-fibre based NGA networks and the previous generation of overlay broadband networks:
- Overlay networks largely have the luxury of choosing what traffic they carry and how;
- Deep fibre networks will become the national infrastructure and with that comes the same responsibility that today pretty much only the incumbent operators have to shoulder.
When you build a FttH network you are saying that you are prepared to take-over that responsibility at some point.
For that reason I would want to make sure the equipment I used to build such a network was designed to carry the burden, and that would rule out consumer grade network equipment and equipment that works brilliantly in overlay networks but isn’t designed for such a critical role.
I could still find a 100 and one uses for MikroTik hardware in my network but I don’t think I would sleep well at night using it for mission-critical network elements. The reality is that being able to meet these requirements necessitates carrier grade equipment with carrier grade processes and support systems. Its all about horses for courses – picking the right tools for the job.
Lessons from the US and Europe show that doesn’t necessarily mean gold-plated pricing or vast scale but it does require a level of understanding that few communities will easily find locally.
And it is this understanding that has typically led European and US communities to favour partnership and concession models – and deterred them from being more hands on.
I don’t want to send out the message that community-led broadband can’t work – it clearly can and I wholeheartedly support it. All I ask is that if your community is considering a DIY approach you weigh up the full implications alongside the benefits you have identified.
If you have any doubts, compare the outcomes and the risks with other models – with developing a partnership with a specialist or from offering a concession to run your network.