Tracing a Large-Scale Peer to Peer System: An Hour in the Life of Gnutella

  • Authors:
  • Evangelos P. Markatos

  • Affiliations:
  • -

  • Venue:
  • CCGRID '02 Proceedings of the 2nd IEEE/ACM International Symposium on Cluster Computing and the Grid
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

Peer-to-peer computing and networking, an emergingmodel of communication and computation, has recentlystarted to gain significant acceptance.This model not onlyenables clients to take a more active role in the informationdissemination process, but also may significantly increasethe performance and reliability of the overall system, byeliminating the traditional notion of the "server" whichcould be a single point of failure, and a potential bottleneck.Although peer-to-peer systems enjoy significant and continuallyincreasing popularity, we still do not have a clear understanding of the magnitude, the traffic patterns, and the potential performance bottlenecks of the recent peer-to-peer networks.In this paper we study the traffic patterns of Gnutella, a popular large-scale peer-to-peer system, and show that traffic patterns are very bursty even over several time scales.We especially focus on the types of the queries submitted by Gnutella peers, and their associated replies. We show that the queries submitted exhibit significant amounts of locality, that is, queries tend to be frequently and repeatedly submitted.To capitalize on this locality, we propose simple Gnutella caching mechanisms that cache query responses.Using trace-driven simulation we evaluate the effectiveness of Gnutella caching and show that it improves performance by as much as a factor of two.