A Novel Peer Selection Algorithm to Reduce BitTorrent-like P2P Traffic between Networks

In a peer-to-peer (P2P) file sharing system, the problem of selecting peers to download from, affects greatly the performance, efficiency, and traffic distribution of the whole system. At present, Bit Torrent(BT) is most popular P2P software used by large-scale people, and BT traffic is main component in P2P traffic. In order to control the BT-like P2P traffic and reduce their bandwidth occupancy at some key location such as gateway, in this paper, a peer selection algorithm based on file piece convergence degree (FPCD) is brought forward, which takes BT as an example. Applying this algorithm, BT-like P2P traffic between networks can be controlled and reduced heavily, which was testified by modeling and analyzing the peer selection algorithm based on FPCD.