Ann Bednarz’s latest article for Network World wasn’t that surprising in the broad sense – yes, social networks are popular at the workplace, yes, they take up a bunch of bandwidth – but I’m not sure it makes the case that it tries to.
The key to the article is an analysis from managed security services vendor Network Box, which tracked 19 billion URLs and ranked the top five Web sites visited from business addresses by volume of traffic. It recorded that 5.8% of all Web traffic went to Facebook, with Google at 4.1%. That’s by visit – by bandwidth usage, YouTube consumed 7.8% of that bandwidth, with Facebook at 4.4% and Windows Update at 3.8%.
But what gets me is that while these numbers are interesting, they’re not particularly useful. As a metric of what Web sites are popular with people at businesses, it’s okay, but it doesn’t give you any idea about the actual impact of Facebook or YouTube on network performance – in short, it doesn’t tell you whether there’s a problem with Facebook or YouTube in your company, or even in companies overall. Plus, it only looks at Web traffic. Without knowing what percentage of overall enterprise traffic is Web traffic, there’s no way to deduce the overall impact on the network. Even then, some networks might have a very little amount, percentage-wise, of Facebook traffic, and it might cause a problem, or conversely, another network might have a large percentage of their traffic going to Facebook, but their network is able to handle the demand.
As such, it’s hard to use that data in the article to make the case that “Social networks take a bite out of corporate bandwidth.” Does that mean that social networking isn’t taking a lot of bandwidth, or isn’t a concern? No – just that it’s not the right evidence to support the conclusion.
That said, gathering evidence that lets you know whether Web traffic (or any other traffic, really,) is affecting your business critical applications is still a vital part of a nutritious network breakfast.
No comments yet.