Huffpost UK Tech uk
The Blog

Featuring fresh takes and real-time analysis from HuffPost's signature lineup of contributors

Thierry Grenot Headshot

What Can Major Events Like the US Election Tell Us About Our Digital Infrastructure?

Posted: Updated:

The US presidential election night was a whopper by most measurements, especially where media network traffic was concerned. Foxnews.com delivered 3,039,989 video streams to the public, and Twitter's website hit a peak of 327,453 tweets a minute the moment the race was called.

Whilst Obama and Romney battled it out for the key to the White House, company networks would have been battling to cope with the additional strain placed on them. Imagine a media site like CNN or the BBC going down because its network resources were unable to cope. It would have been more than a little embarrassing. Past instances of poor network management stand out. In April 2011, the entire BBC website crashed, due to Royal Wedding fever. With thousands of viewers attempting to watch Kate Middleton walk down the aisle, their networks couldn't cope, leaving many viewers staring at a static screen instead of watching the nuptials.

For businesses, the dangers of increased and unmanaged traffic levels over corporate networks are immense. Each application running over a network (like Twitter, or videos, or Facebook), drains the network's resources a little. If one app is greedily sapping more than its fair share of the bandwidth due to heavy usage, then the other apps suffer from poor performance and users end up frustrated. If business-critical apps (such as video conferencing) are being side-lined due to employees frenziedly tweeting or streaming election result videos, then the company has a problem.

Losing five minutes per day from poor application performance creates a 1% drop in productivity, which can limit profitability by 10%. This is never acceptable in today's economic environment. These peaks in traffic are not just a phenomenon at election time - natural peaks (such as those occurring when companies run system back-ups at specific times of the day) can produce similar spikes in demand.

So how can businesses prevent this? Regardless of the company's size, they need to have greater visibility over the network, and to be able to see exactly what's running over it. If employees are streaming election videos from certain news sites, then the person in charge of IT absolutely needs to be aware of it, and acting where necessary.

We're certainly moving past an age where one 'pipe' is equal for all applications using it. The network itself and its managers need to be able to understand the traffic it is carrying and to make more intelligent decisions about how resources are allocated.

Not all apps are created equal

Once visibility has been established, the next step is gaining control. It could be that certain apps are particularly demanding, and need to be reined in for the sake of wider network performance. Take video streaming as an example. With so many employees streaming the news to keep up-to-date on the election results, you can bet that other (potentially business-critical) apps were being side-lined as a result. The network needs to be managed in such a way that business-critical traffic is given priority.

This would mean that the streaming of elections would not have any negative impact on the other applications that are needed by the business. So there comes a need to 'rank' traffic, and decide which flow of data should take precedence over others if network resources are ever strained.

Just how far the IT manager wants to go in terms of 'controlling' usage is of course up to them. In a recent survey of CIOs and IT managers across Europe, we found that 67% said they currently block Facebook, 56% do the same for on-line video, and 49% block Twitter. This is interesting, and sparks a natural debate over whether it's fair to curtail an employee's social media use so heavily.

Events as significant as the presidential elections don't crop up every week, and this is something which CIOs and network managers can perhaps be grateful for. But it's important to know that when they do arise, either within the business itself or more generally, the company's networked tills can continue processing orders, its websites remain live and its logistics systems continue to function.

In addition to here at the Huffington Post, we talk about these topics over on our blog, www.AppsSoluteblog.com