【摘要】 随着以BitTorrent为代表的P2P文件共享系统应用的迅速增长,对BitTorrent流量进行有效的识别和监管已经成为ISP网络管理中越来越迫切的需求。本文通过高速捕包、应用层协议还原进而深包分析,实现了可以适应大流量环境下的Bit-Torrent流量识别系统。在ISP骨干网路由器实际测量结果显示本系统可以实现在大流量下进行准确、高效的识别。更多还原

【Abstract】 The use of Peer-to-Peer (P2P) applications is growing dramatically,especially for BitTorrent file sharing system. It is therefore become ugerent to classify and monitor the traffic generated by BitTorrent from the ISP’s perspective. In this paper,we design a BitTorret traffic classification system across a large ISP network through packets captuiring and application-level reassembly. The measurement of the backbone router of a larsge ISP shows that our system is able to classify the traffic accuratly and efficiently under the gigantic network traffic load.更多还原