Modeling and Analysis Traffic Flows of Peer-to-Peer Application

Recently, peer-to-peer (P2P) applications have generated the majority of Internet traffic. In this study, we model and analyze characteristics of P2P traffic flows by measuring real P2P application data captured from operational network. With lognormal mixture distribution and lognormal distribution, we get accurate statistical models for flow size and flow duration of Maze application traffic. We also find that moderate correlation exists between flow size and flow duration. With scatter plots, we analyze the reason for correlation formed.