基于BT对等方地理信息的权重tit-for-tat机制

Weighted tit-for-tat mechanism for BT based on geographical locations of peers

为了减少BitTorrent(BT)产生的大量跨ISP流量且同时保证用户的下载速率,提出了一种基于对等方地理位置的加权tit-for-tat机制(GLW-TFT).首先,根据对等方的地理位置信息将BT群集系统中的对等方分为4类,然后根据ISP收益情况给每一类对等方赋予权重,最后利用加权的下载速率来常规上载对等方集合.基于开源项目Vuze设计开发了GLW-TFT,它不依赖于网络基础设施也不需要ISP的协助.试验结果表明,相对于传统的BT,GLW-TFT能缩短约50%的下载时间,减少约30%的跨ISP流量.GLW-TFT算法较之其他相关算法具有更强的适应性和较好的性能.

In order to reduce cross-ISP（Internet service provider） traffic caused by BitTorrent（BT） and guarantee the downloading rate of BT users,a weighted tit-for-tat mechanism based on geographical locations of peers called GLW-TFT is put forward.First,peers in the BT swarm are divided into four categories according to their geographical locations.And then,a weight based on the ISPs＇ income is assigned to each category of peers.Finally,a metric based on weighted downloading rate is proposed to regular unchoke peers.GLW-TFT requires neither feeds from Internet infrastructure nor deployment on a large scale.GLW-TFT is implemented based on open source project Vuze.Experimental results on Internet show that GLW-TFT can not only reduce about 50% of the average downloading time but also save about 30% cross-ISPs traffic compared to traditional BT.GLW-TFT is a more general traffic optimizing algorithm,which has better adaptability and performance than other related algorithms.