A measurement study supporting P2P file-sharing community models

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.comnet.2008.11.007 Byline: R. Bolla (a), R. Gaeta (b), A. Magnetto (b), M. Sciuto (a), M. Sereno (b) Keywords: Peer-to-peer communities; File sharing; Measurement; Overlay topology; Gnutella Abstract: Knowledge of emergent properties of existing peer-to-peer file-sharing communities can be helpful for the design and implementation of innovative peer-to-peer protocols/services that exploit autonomicity, self-configuration, resilience, scalability, and performance. It is well known that the performance of this class of peer-to-peer applications depends on several parameters that represent the topological structure of the overlay network, the users' behavior, and resource dynamics. Estimation of these parameters is difficult, but it is crucial for analytical models as well as for realistic simulation of peer-to-peer applications. In this paper, we present an active measurement-based study designed to glean insights on the above parameters within the Gnutella network. The measurement software that we developed is able to collect topological information about the overlay network topology in a few minutes; in a second step, it contacts the users discovered during the topological measurement in order to acquire a novel dataset regarding the shared resources. Author Affiliation: (a) Dipartimento di Informatica, Sistemistica e Telematica, Universita di Genova, Via Opera Pia 13, 16145 Genova, Italy (b) Dipartimento di Informatica, Universita di Torino, Corso Svizzera 185, 10149 Torino, Italy