Measuring and Modeling Message-Level Characteristics of BitTorrent

BitTorrent (BT) applications are very popular on the Internet. Measuring the impact it has on the network and understanding their behavior is an important field. From the perspective of network management, a thorough understanding of the stochastic properties of BT traffic is thus crucial to designing better traffic and resource control algorithms. However, previous works on measuring and analyzing BT application mostly focus on both data and message traffic. In order to gain insights into BT systems especially the message characteristics, we have undertaken an in-depth measurement and modeling study of BT message. Our results show that the Message model can well describe the BT incentive mechanism and the statistics collected on different files can fit equally well into distributions.