Hi, we are pleased to announce a new report on BitTorrent locality. This work in mainly on how much can we keep BitTorrent traffic local to ISPs, and what would be the global impact (in terms of inter-ISP traffic reduction and peers download completion time) of a locality policy deployed in the Internet.
In addition, we believe that this work gives several additional interesting hints on BitTorrent (not the focus of the paper, so we don't really stress those results, but the results are in the figures and should be interesting to the community) -with experiments with up to 10 000 real BT clients we found that BT is only 40% slower than an ideal over provisioned client server scenario (by ideal, we mean a server with a capacity equivalent to 10 000 peers, thus 1.6Gbit/s in our experiments, and no congestion on the access links). Even if it does not come as a surprise that BT is extremely efficient, it is, to the best of our knowledge, the first large scale controlled experiments that show and quantify this efficiency. -we have interesting results on the distribution of the peers per AS for a very large number of real torrents and ASes. Stevens Le Blond, Arnaud Legout, Walid Dabbous. Pushing BitTorrent Locality to the Limit. Technical Report (inria-00343822, version 2 - 12 May 2009), INRIA, Sophia Antipolis, May 2009. http://hal.inria.fr/inria-00343822/en/ Regards, Arnaud. -- Arnaud Legout, Ph.D. INRIA Sophia Antipolis - Planète Phone : 00.33.4.92.38.78.15 2004 route des lucioles - BP 93 Fax : 00.33.4.92.38.79.78 06902 Sophia Antipolis CEDEX E-mail: [email protected] FRANCE Web : http://www-sop.inria.fr/planete/Arnaud.Legout/index.html _______________________________________________ p2p-hackers mailing list [email protected] http://lists.zooko.com/mailman/listinfo/p2p-hackers
