Building resilience for the big event: Part 1 - The network

27 July 2012 | by Josh R

So, it starts today. That thing we can't really talk about. But we'll probably be sat at home watching it on telly (unless you're Chris and off to watch some movie that came out AGES ago).

It won't surprise you that, like the rest of the country, we've been preparing for this for some time. Even before it hit the front pages, we knew that Internet connectivity in the UK would come under unprecedented strain over the next 17 days. There's no way that we wanted to be caught with our metaphorical pants around our ankles.

In addition, as businesses across the UK set out their stall in front of a global audience, hosting providers and technical support teams will be under even greater pressure to keep servers online while being hammered from across the world.

Through the efforts of our network, engineering and support teams, we're ready to handle this. In this three-part series, we'll take you through the extensive preparation we've taken to ensure your hosting is as reliable as ever and that we can continue doing 'business as usual'.

Part 1: Building a resilient network

Until recently, over 80% of the UK's Internet traffic went through LINX - the London Internet Exchange in East London. At its peak, LINX has handled traffic of 1.2 terabits per second, "pushing the capability of even the most bleeding-edge equipment".

As one of the few hosting providers in the north west in complete control of our infrastructure, we can be confident in the readiness and reslience of our network.

Every piece of hardware that is critical to the network has a hot-standby available. And our network transit is based on a blend of routes that leave the UK without going through London.

That's also why we supported the creation of IX Manchester - a "regional peering exchange" in the north west which allows hosting providers, ISPs and others like the BBC to send traffic to each other without needing to go to London. Simply put, traffic which travels to London simply to come back up north can now take a new bypass to cross the road.

And that bypass has been incredibly popular. Already, IX Manchester is regularly sustaining 600 megabits per second - taking a hugely significant edge off the pressure in London. The recent levelling-out of LINX's London traffic is evidence that this strategy is working. And this confirms Manchester as the biggest Internet hub in the UK outside of London.

And, if LINX does go down, our self-healing network will automatically find an alternative route for traffic around the country and overseas - keeping your servers globally accessible on the Internet.

In conclusion, with the BBC broadcasting 24 HD channels of sport and 27,000 journalists across the world converging on East London each demanding live connections back to their home base, it could get brutal. But by "baking in" resilience to every part of our network infrastructure and traffic routes, we're confident that we can handle what promises to be an incredible global event for the Internet.

Coming up next...

In Part 2, we'll talk about our reliable server hardware and the disaster-recovery mechanisms  we offer to get your business back up and running in the unlikely event of a server failure.

PS talking of baking, check out what's cooking at Melbourne this afternoon!

[googleplusauthor] @ Melbourne

Like what you've read?