Evaluating Peer Churn Effects on P2P-Based Video-on-Demand Services

This paper investigates the effects of peer churns in peer-to-peer networks when used to provide video-on-demand services. To evaluate the impact of user behavior in terms of server load and peer satisfaction, three system models have been developed, which are based on: the Gilbert-Elliot chain, the fluidic representation of the user behavior and a queuing analysis of the system. The models are compared evaluating the additional resources that can increase the reliability of P2P network. Simulations show important relationships between playback buffer length, peer request rate, peer average lifetime, and server upload rate.