7 Comments

Summary:

It should come as no surprise: the big and fat video files are one of the main reason why the bandwidth glut created by overbuilding in the telecom boom of 1990s is evaporating. The good news is that the demand for bandwidth is not going to […]

It should come as no surprise: the big and fat video files are one of the main reason why the bandwidth glut created by overbuilding in the telecom boom of 1990s is evaporating. The good news is that the demand for bandwidth is not going to end any time soon.

A report released by Cisco Systems (CSCO) predicts that consumer-related traffic running over IP networks is going to grow at a compound annual rate of 58 percent from 2006 to 2011, and will end up totaling 17 exabytes per month by 2011.

Cisco sees peer-to-peer traffic as a major component of the consumer internet traffic, even though it will decline from current 50% to 39% of the total. According to Network World:

Internet video streaming and downloads will grow from 9 percent of all consumer Internet traffic in 2006 to 30 percent in 2011.

Internet video-to-TV will increase by more than a factor of 10 from 2007 to 2011.

Internet video-to-PC will increase by a factor of four. Internet video-to-TV will exceed Internet video-to-PC by 2009.

The reports’ findings are consistent with predictions that bandwidth to the home will increase manifold, with 20 megabit per second connections becoming common place by end of the decade. It also indicates that the world might finally get to see HD video over the web.

Just to keep things in context, you should discount some of these forecasts because Cisco makes a living by selling gear such a router, switches and storage systems that benefit from this growth in traffic. Such a report could boost the demand for its products.

  1. After the BBC iPlayer affair, Cisco report will raise again the “Net neutrality” issue : do you believe that the telcos will accept to invest zillion dollars in Cisco products and backbone infrastructures to empower the Youtube access for their customers, without any extra revenue ?

    Share
  2. I remember reading one of your posts in which you showed profound skepticism with Cisco’s second time coming. With the strong financial results a couple weeks back and a multitude of such video explosion reports (not just from Cisco), do you still feel skeptical? Or are you convinced now?

    Share
  3. That prediction of future p2p traffic clearly makes assumptions that might be dead wrong. If mainstream video begins to use p2p architectures, which I believe they must, there will be no decline in such traffic.

    One of these days I would like to see a scorecard of the predictions of so-called “futurists” to see how often they are wrong, and how seldom they are correct.

    Share
  4. Disclaimer! I work for Cisco and was involved in the study. But I am posting as an individual who has developed a personal interest in ip traffic… and who just realized how sad that sounds. Anyway opinions below do not represent Cisco in any way.

    I have found it convenient to use the numbers in the report to help evaluate various claims about traffic growth, for instance the claim that BBC iPlayer traffic has the potential to tax service provider networks. You can estimate the potential traffic and then compare it to the 2007 estimate for overall internet traffic in Europe as published in the report. In an aggressive scenario, iPlayer traffic from one million users could generate 13 PB per month, which is just about a quarter of total UK consumer traffic today. That’s quite a bit, especially for those ISPs who are paying by the bit. A less aggressive scenario would have it at 7-12% of traffic a year from now. Today traffic from iPlayer (assuming 480K hours per month) is just a fraction of a percent of total traffic.

    Traffic growth seems to be at the center of a number of industry debates, so I think it’s important to do this kind of evaluation, using the Cisco report or any other reference.

    Side-note: on the credibility of the forecast, the growth rates of the forecast are pretty reasonable, perhaps even on the conservative side – really amounts to a little under 50% growth per year. The Telegeography figures, which are for international traffic but are a good indicator of overall growth, were about 75% for 2005-2006, I believe. They haven’t yet published 2006-2007 numbers.

    And note to Michael Gersh – you’re right. The report acknowledges that P2P TV is a wildcard scenario not worked into the assumptions of the forecast. Not sure it’ll happen before 2011 though.

    Share
  5. Joe,

    If you read all my previous posts, it has been since late 2005 I have been writing that video will lift the fortunes of the telecom/networking business. and that is exactly what has happened. I remain a little skeptical of reports that show that traffic is doubling in 2 years. Remember, ten years ago people talked about traffic doubling every 90 days or whatever, and that cause irrational investments. It was a problem then, and that is why I remain skeptical. Not a non-believer, just more cautious.

    Share
  6. Online video growth will increase not only in raw transaction terms but with the increasing resolutions supported by PCs and TVs. I’d venture to say that even if you kept streaming transcations flat until 2011, the move to larger resolutons, i.e 480p and 720p, will account for a sizable increase in IP bandwidth.

    Share
  7. [...] month Om wrote about the boom in online video, pointing to a report released by Cisco Systems (CSCO) that predicts video will grow to thirty [...]

    Share

Comments have been disabled for this post