P2P: From Internet Scourge to Savior

Continued from page 1

By Wade Roush

Friday, December 15, 2006

Even BitTorrent, an advanced P2P network long seen by movie and record executives as an irksome successor to Napster and Kazaa, is going mainstream. The company announced early this month that it had raised $20 million in a second round of venture-capital funding and acquired competitor μTorrent (pronounced "microtorrent"), maker of a compact version of the BitTorrent software meant to be suitable for set-top boxes and other non-PC devices. BitTorrent--which speeds up downloads by grabbing and reassembling file fragments from the most accessible peers on the network, rather than by transferring whole files from one peer to another--is still one of the best tools for locating and procuring Internet video. That's in part because it's free and in part because so many people use it and have built a worldwide archive of digital files.

Many other companies are jumping into the P2P video mix, including Peercast, Octoshape, Allcast, and Itiva. Dijjer, a project of Revver (which was founded by peer-to-peer pioneer Ian Clarke), reduces the load on individual users' computers by grabbing much of the content of a requested file from other users' computers. A team of researchers from three universities in the Netherlands has introduced Tribler, a BitTorrent-like program that adds useful features such as Amazon-like recommendations and real-time maps showing who is downloading the same content. And in the United Kingdom, the BBC is developing a peer-to-peer media player called iPlayer. In a trial conducted with 5,000 users from November 2005 to February 2006, users were able to download first-run BBC shows for seven days after their official TV broadcast; many participants used the service simply to catch up on their favorite programs, but interestingly, the network also found that study participants were disproportionately drawn to several new shows and "niche" shows that hadn't fared as well among regular broadcast viewers.

The Internet is already teeming with peer-to-peer traffic. In fact, P2P downloads may account for as much as 60 percent of network traffic--and as much as 60 percent of that traffic is video, according to CacheLogic, which has developed a proprietary system for accelerating P2P downloads. (The system enhances P2P distribution by giving peer-to-peer networks access to dedicated, high-capacity "edge servers" scattered around the Internet, in much the same way as the traditional content-distribution networks pioneered by companies like Akamai.)

So how could additional P2P traffic actually be a good thing for the Internet? Carnegie Mellon's Zhang points out that because peer-to-peer networks exploit both the downlink and uplink capacities of users' Internet connections, they distribute content more efficiently than centralized "unicast" technologies. Zhang also says it should be possible to label P2P traffic so that service providers can track it and decide how much of it to allow through their networks. He and colleagues from the University of California at Berkeley have founded a startup, Rinera, to develop software that will give service providers such control.

"The network itself needs to be informed about the types of traffic it's handling, and service providers need to participate by setting policies," says Zhang. "Otherwise, as applications like video downloading really take off, we will see a congested network, which will in turn impede the development of video-sharing technology."