Yesterday, the NFL announced that it would be streaming the Super Bowl live to iPads, Verizon mobile phones and the NBC/NFL websites. In my mind, this marks a significant moment in the annals of streaming video as the potential turning point in the transition from traditional delayed broadcast video to live streaming. Sure, there have been plenty of live streams across the Internet (and not just pirated streams of sports events!). In fact, the NFL streams games every week and the past two Olympics have been streamed online. The NFL claims that 20,000-30,000 viewers watch their regular Sunday night streams, compared to 21,000,000 viewers of the broadcast version. However, the Super Bowl was watched by 111,000,000 viewers last year. NBC has upped the ante by promising excusive content and analysis for the streaming version, which is likely to result in people having both the streaming and broadcast versions active in their household at the same time – the big consumer most likely will be the iPad version.
Why is this significant for the Internet?
1) The NFL and NBC have confidence in their ability to stream content to millions of simultaneous consumers around the world. Although the NFL is not a worldwide sport like soccer (football to the rest of the world!), people watch the Super Bowl worldwide. If this goes well, expect the World Cup final to get the same treatment in 2014.
2) Broadband networks are expected to be able to handle the streaming load (To Be Determined!) across millions of subscribers. I expect broadband operators to be scrambling to ensure that their peering links with the CDN for this event is up to the task.
3) The CDN delivering the live streaming is confident that it is up to the task for a live event. This is not a case where you can pre-position content to meet the expected demand.
4) Latency issues between the live telecast and the streaming event are expected to be minimal. This one is interesting to me as I have seen many cases where a score is reported via Twitter before it happens on my TV screen – for many sports.
If this event succeeds and the NFL is able to claim that they are the first to stream such a significant event live, it will start a rush towards content owners for live events to monetize their properties. It will also potentially signal an aggravation of the problem for broadband operators, as it will give new meaning to “peak” hours. Peak will now be whenever the event occurs – and not during the standard peak hour times. Some broadband networks will struggle with network congestion in this scenario, so it will be interesting to see which operators can deliver high-quality streams during this event.
Here at Procera, we have been seeing stress building on the Intelligent Policy Enforcement infrastructure; live video streaming will add a new dimension to it. In addition to the bandwidth stress, I would expect to see services pop up in this area – like zero-rated streaming, priority streaming (for a fee) – all of which will not only stress the infrastructure, but also the management and OSS layer for policy management and charging. With live events, this will be an even more acute problem for PCRF/PCEF/OCS/OFCS systems, as the sessions will start within a very short time, creating almost a “network outage”-type flood on the network.
We will be watching this event very closely. Stay tuned for our report after the event.
