A Dynamic Popularity-based Cache Replication Strategy for P2P VoD Systems

Peer-to-Peer Video-on-Demand (P2P VoD) over Internet has become an effective and popular way to provide video information to users. The knowledge of video popularity is very important for system operation, such as video caching on peer-to-peer network. The video popularity distribution at a given time is quite well understood. However, in this paper, we study how the video popularity changes with time, classifying them into different types, and apply the results to design a novel popularity prediction model, which is derived by analyzing the video’s user rating information as well as time-variation. Based on this model, we propose a dynamic popularity-based cache replication strategy (DPCR). The simulation results demonstrate that our proposed algorithm can greatly reduce servers' workload and improve the streaming quality.