摘要
对P2P集群的测量与分析,有助于理解其现状及演变规律,从而对其进行优化和管理.根据赠券收集模型的一般原理并进行了优化,设计实现了一种主动测量方法及平台,可用较少的主机对全球BitTorrent集群开展持续的主动测量.该方法能在短时间内,获取每个集群中超过90%的在线节点,同时对网络本身的影响较小.在近9个月的实验中,分别测量了从Torrentz获取的3万多个BT集群.结果表明BT集群中的节点分布具有明显的地域性,而且国内外用户在客户端和端口使用上具有明显的偏好.
Detecting the P2P swarm,analyzing their distribution is a challenging task,which has not received the deserved attention.We used active measurement methodology to continuously trace the real-world BitTorrent swarms over the Internet from a stub network for a long period of time.Our measurements achieved the ability of real-time scanning the online peers,avoiding the potential copyright infringement and heavy traffic on Internet.During the over eight-month measurements,from July 15 2010 to March 31 2011,we crawled 33,008 swarms,collected 8,963,776 distinct peers.Then we analyzed the distributions and preferences of peers.We find that the geographical distribution of peers in all swarms span quite universally.And there are some quite differences in the preferences between the users in and outside China.
出处
《微电子学与计算机》
CSCD
北大核心
2012年第11期134-137,共4页
Microelectronics & Computer
基金
国家自然科学基金项目(61070170)
江苏省高校自然科学基础研究项目(11KJB520017)
苏州大学大学生创新创业训练计划项目(111028525)
苏州大学大学生课外学术科研基金项目(KY2012073A)
关键词
P2P集群
主动测量
赠券收集模型
地域分布
用户偏好
peer-to-peer swarm
active measurement
coupon collector problem
geographical distribution
user preferences
