Exploring proximity based peer clustering in BitTorrent-like Peer-to-Peer file sharing systems

摘要

A hierarchical clustered BitTorrent (CBT) system is proposed to improve the file sharing performance of the BitTorrent system, in which peers are grouped into clusters in a large-scale BitTorrent-like underlying overlay network in such a way that clusters are evenly distributed and that the peers within the cluster are relatively close to each other. A fluid model is developed to compare the performance of the proposed CBT system with the BitTorrent system, and the result shows that the CBT system can effectively improve the performance of the system. Simulation results also demonstrate that the CBT system improves the system scalability and efficiency while retaining the robustness and incentives of the original BitTorrent paradigm.