Myth of the Bandwidth Hog

Internet Service Providers (ISPs) have been quick to blame problems with service quality on so-called “bandwidth hogs.” According to AT&T, the top 5% of their Digital Subscriber Line (DSL) subscribers use 46% of the consumed bandwidth, and the top 1% of subscribers use 21%. But it is unclear what these figures mean, and if congestion problems could even be caused by those who use the network the most. These figures would seem to be describing the bandwidth consumption totals at the end of some designated time period (day, week, month). If this is the case, then 5% of subscribers using 46% of bandwidth consumed is not necessarily cause for alarm.

Excessive bandwidth usage is only a problem when it degrades the quality of service for other users of the network. Service is only likely to be degraded for other users during peak usage hours because that is when the most people are on the network. In order for a user to fall within the top 5% of bandwidth users for a given time period they would have to be downloading content all day, not just during heavy usage times.

For example, lets assume we study two users on the network, lets call them Alice and Bob. Both are on the same service plan and network, which allow them 5mbps normally, but the network gets congested during peak hours (5pm - 11pm) allowing a user to use at most 2mbps. Alice gets online at 2 am and starts downloading large files, maxing out her internet connection until the next day at 2 pm. Bob gets home from work at 5 and hops online to watch streaming videos until he goes to sleep at 11pm, using only 1mbps. At the end of a month Alice will have downloaded a huge amount of data and doubtless be tagged as bandwidth hog, yet contributed nothing to the congestion problem. Bob will be considered a normal user at the end of the marking period - but has a far more significant impact on the lower quality of service during peak hours than Alice does.

Targeting these alleged bandwidth hogs for having large total usage by restricting network access or raising rates doesn't get at the real problem, which is congestion during times of heavy usage. These penalties are essentially punishing consumers for gaining extra utility from the service, particularly at non-peak usage hours, at no harm to the ISP or other users.

Hogging implies taking much of some resource for one's self and preventing others from gaining access to that resource. If a customer is able to do this, then the network is not being managed properly. If the network is so constrained such that 5% of users can cause congestion, even during off peak times, then the infrastructure is clearly due for an update or the ISP has promised much more bandwidth than it can deliver.

Throttling everyone during peak usage is the only sensible solution to this problem (besides upgrading the network). Getting online and finding speeds to be slow during peak hours is akin to trying to place a cell call during especially heavy usage and getting denied service, it is an unavoidable degradation in quality of service which should not pose a problem if it happens infrequently.

Bandwidth caps won’t help solve these congestion issues, the problem is not with aggregate bandwidth usage over all times, but rather congestion which occurs occasionally. Any form of time agnostic capping or throttling doesn't target the real issue. Bandwidth caps simply encourage anticompetitive behavior, especially in the video entertainment sector. Most uses of the internet which would lead to going over a cap on bandwidth usage are internet video and the same companies which are imposing these new caps are selling cable TV services to their customers. Clearly it is in their interest to steer consumers towards their products and away from the competition.

There are many ways to manage traffic and increase performance, and companies should be free to make use of them, so long as they are agnostic to user, protocol and destination/source of data.

The truth is ISPs need to find long terms solutions to deal with heavy users on their networks, especially since today's bandwidth hog will be tomorrow's average user. New technologies are on the horizon which should provide an abundance of bandwidth to the home with room to expand capacity when demand grows, yet it seems that at some point all new networks grow to be old and there will inevitably be constraints on bandwidth availability.

Artificial suppression of bandwidth demand won't solve the problems networks face today and are likely to slow growth in the information sector and delay much needed investment to upgrade the networks.

The Latest