Research > Faculty Projects

Flow-based Peer-to-peer Identification

Principal Investigator
Jenq-Neng Hwang, Alan Lippman

Sponsor(s)
Redback Networks, Inc.

Award Period
09/16/2007 - 09/15/2008

Abstract
In order to utilize the network bandwidth, many companies have products for detecting peer-to-peer (P2P) traffic, since today the P2P flows consume more than 80% of the bandwidth. Most approaches focus on port detection or signature detection, also called Deep Packet Inspection (DPI). However, these approaches hardly detect new P2P protocols, since the protocols may change in new applications. To overcome this problem, a new approach called flow-based identification has been proposed. This approach tries to identify the P2P flows by statistically retrieving the specific characteristics of P2P flows. In this project, we will develop a streaming media (both unicast and P2P) traffic identifier and controller to enable networks to identify which flows contain streaming media. This identifier will have the following features. For the traffic identifier, our methods will identify streaming media traffic on a Flow by Flow Basis, using Port, DPI and Flow information. For the controller part, we will implement a control model to update and manage new signatures for both DPI and Flows. Finally, we will analyze the accuracy of these methods on a real world network that heavily uses streaming media applications.

For more information, visit http://allison.ee.washington.edu/p2p/index.html

Updates or corrections to this page should be sent to gheaton@u.washington.edu.