BSU: A Biased Seed Unchoking Algorithm for P2P Systems

Abstract

P2P architecture is one of the two predominant architectural paradigms in modern network applications, and will still be an important technology for building the content delivery applications in the future Internet. As the source of the content in P2P systems, the main objective of the seed is to help deliver the content to all peers as soon as possible and to ensure the quality of service of the peers as well as to stimulate them to contribute resources to the system. Recent researches mostly concentrate on either the unchoking algorithms of the leechers or the scheduling algorithms of the seed. In contrast, the unchoking algorithms of the seeds have drawn little attention. Experimental results have shown that the unchoking algorithm of the seeds may have a remarkable impact on the performance of the system, especially when there are a large amount of free-riders in the system. Firstly, a P2P content delivery model based on the fluid model is established in this paper. Secondly, a biased seed unchoking algorithm (BSU) is put forward. Moreover, a random seed scheduling (RSS) algorithm is proposed to further improve the efficiency of the system. Finally, a series of experiments in both homogeneous and heterogeneous environments are conducted to investigate the performance of BSU and RSS algorithm. Experimental results show that BSU and RSS algorithm can achieve better performance than existing algorithms used in BitTorrent-like systems remarkably.