Characterizing the network behavior of P2P traffic


Nowadays the majority of Internet traffic is generated by peer-to-peer (P2P) file sharing applications. As the popularity of these applications has been increasing dramatically over the past few years, it becomes increasingly important to analyze their behavior and to understand their effects on the network. The ability to quantify their impact on the network is fundamental to a number of network operations, including traffic engineering, capacity planning, quality of service, forecasting for long-term provisioning, etc.

We present here a measurement study on the characteristics of the traffic associated with two different P2P applications. Our aim is to provide useful insight into the nature of P2P traffic from the point of view of the network. To achieve this, we introduce a novel measurement, Content Transfer Index (CTI), to distinguish two classes of behavior associated with P2P traffic: the download and the signaling traffic profile. Next we apply the CTI to our data sets and show that it effectively offers a general characterization of P2P traffic. Finally, we present a number of statistical measurements that are significantly unbiased due to having considered the distinction between the two classes. To the best of our knowledge, this is the first study to follow this approach.

We believe such a study will help researchers better understand the impact of P2P applications on the network and how to improve their performance.

Proceedings of the Fourth International Telecommunication Networking Workshop on QoS in Multiservice IP Networks (IT-NEWS'08)