Server load based prefetching strategy for P2P VoD streaming

In a peer-to-peer (P2P) video-on-demand (VoD) system, a peer can prefetch some movie data for its future needs when it is watching a movie. These prefetched movie data can also offload the servers, which is in the best interest of VoD service providers. To offload the servers effectively, which segment of a movie a peer should prefetch is an important problem. In this paper, we propose a server load based prefetching strategy to solve this problem, which aims at reducing server load. In this strategy, we propose to use the EWMA (Exponentially Weighted Moving Average) server load to determine which segment of a movie a peer should prefetch. Furthermore, we integrate this strategy into our previously proposed P2P VoD scheme. Through simulations, we demonstrate that this strategy can effectively reduce server load.