A grouping algorithm of helpers in peer-to-peer video-on-demand systems

Peer-to-peer (P2P) content distribution effectively solves the problem of large-scale streaming media delivery. But the performance of a P2P system is primarily bottlenecked, because the upload bandwidth of the participating peers is limited. Using helpers in the P2P system utilizes upload bandwidth of those idle peers uninterested in the delivered content, which further improves the performance of P2P systems, and the feasibility and validity have been testified. Aiming at the p2p video-on-demand system, this thesis designs a novel grouping algorithm of helpers. In order to make more rational and effective use of helpers, the algorithm divides the helpers into different groups according to the number of each video´s users, thus reducing the average view delay of whole system and improving the utilization rate of the helpers, reinforcing watch experience for users of the system. Analysis and simulation results corroborate the effectiveness of the proposed algorithm in the paper.