Analysis of BT network based on active measurement

ICNDS), 2010 2nd International Conference(2010)

引用 0|浏览6
暂无评分
摘要
Peer-to-peer (P2P) systems enable a wide range of important, scalable and reliable services such as file sharing, voice-over-IP (VoIP) and video streaming. And they also generate most traffic in all regions. Excessive P2P traffic can cause significant network management problems because these systems are notoriously greedy with regard to bandwidth consumption. As a result, the ISPs and all network administrators have a mutual interest in shaping, blocking or otherwise limiting P2P traffic. Before doing these, understanding, analyzing and identifying P2P traffic is necessary. In this paper, we inside analyze BitTorrent (BT) traffic with a special modified Vuze client in a stub network. Preliminary results indicate that among all forty-nine BT clients we have detected, the most popular two are μTorrent and Azureus. They have over 50%-86% market share. And about one-third of all clients are the legacy or proprietary client, which have no capacity of supporting the Message Stream Encryption (MSE) protocol. Further, the BT clients relied excessively on trackers, because near half peers' information come from trackers, especially in the beginning of new mission. Moreover, Distributed Hash Table (DHT) did not play the proper role as it is expected. Finally, we unexpectedly find that the most popular used port is still 6881, but it is only 2.24%.
更多
查看译文
关键词
computer network management,computer network security,cryptographic protocols,peer-to-peer computing,telecommunication traffic,μtorrent,azureus,bt network,bittorrent,isp,p2p traffic,vuze client,active measurement,distributed hash table,message stream encryption protocol,network administrator,network management problem,peer-to-peer system,stub network,dumb client,peer-to-peer,traffic analysis,market share,p2p,cryptography,voice over ip,computer networks,network management,protocols,internet telephony,file sharing,bandwidth,information analysis
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要