Chinese Academy of Sciences
Peer-To-Peer (P2P) techniques are broadly adopted in modern applications such as Xunlei and Private Tracker. To address the problem of service availability, techniques such as bundling and implicit uploading are suggested to increase the swarm lifespan, i.e., the duration between the birth and the death of a swarm, by motivating or even forcing peers to make more contributions. In these systems, it is common for a peer to join a swarm repeatedly, which can introduce substantial bias for lifespan modeling and prediction. In this paper, the authors present a mathematical model to study the lifespan of a P2P swarming system in the presence of multi-participation. They perform evaluations on three traces and a well-known simulator.