Peer to Peer Traffic: A Catalyst in Media Evolution

Peer to Peer [P2P] traffic is the stowaway traffic of the net. It is latency and QoS insensitive, and happy to fill the unused capacity of broadband connections. The big growth rate on the chart below gives P2P the illusion of being important- the reality is it is just being efficient.

P2P is using the existing resources of the net more efficiently, and, unless it can be monetized, will never drive the deployment of more fiber to the home [FTTH], more transport bandwidth, and more switching hardware.

I’m convinced that the inflection point for FTTH deployment will come when a killer application requiring large upstream bandwidth arrives. On the surface, P2P fits this bill, as it is the only widely deployed consumer application that stresses upstream bandwidth. The problem is, P2P is a lunatic fringe application built on illegal activity. The nature of P2P’s popularity makes it impossible to monetize.

The frustrating reality for carriers is that as broadband connection speeds increase, the amount of peer to peer traffic grows disproportionately, an effect I wrote about while discussing network neutrality.

In short, when someone gets a faster connection, they are more likely to either start using P2P, or use a lot more than they did before. Traffic increases geometrically with bandwidth, driving up the costs of back haul connectivity and reducing the carrier's ability to do statistical multiplexing. It costs real money to transport.

However, P2P is currently playing a vital role by forcing the media companies to evolve. They must adjust their business models to sell content electronically or die. Without Napster (NAPS) for music or BitTorrent for movies, there would not have been a catalyst for these companies to start selling media electronically.

Media distribution will evolve, and then P2P will fade into the background as an inexpensive best-effort way to deliver content, and a domain for those who still refuse to pay.

Conclusion: P2P isn’t a business. It’s a catalyst. Keep that in mind.

Footnote: Mary Meeker, ex-Queen of the Net, is back in the swing of things at Morgan Stanley. She recently gave a presentation on “The State of the ‘Net” that is filled with good data, including the above chart.

Comments

[Hiten Parmar:]

If P2P were forced, couldn't it be leveraged? Comcast could stick larger harddrives into its cable modems or DVRs and offer P2P delivery services. Microsoft could do the same thing with Xboxs.

2006 Dec 11 03:00 PM

[Andrew Schmitt:]

P2P is an awful technology for Cablecos because it stresses upstream bandwidth, which is the Achilles heel of their plant. Comcast would rather pay for caching than deal with the hit on their upstream bandwidth.

P2P is the best way to distribute the anti-long tail media, but such media is also the easiest to cache. I just don't see where P2P fits in as a mainstream distribution technology.

2006 Dec 11 05:28 PM