Peer-to-peer (p2p) content sharing and distribution has become of such widespread use that various user 22 communities, even quite large ones, have formed and may exhibit specific behaviors. The main aspects 23 characterizing the latter may stem from users’ interest groups, specific attitudes, or even from locality 24 reasons – e.g., lying behind a network access translation (NAT) protected Internet access, which is respon- 25 sible for the reduction in transparency of end-to-end applications. We investigate the traffic patterns of 26 one such community, adopting the eMule protocol, by means of an extensive data collection, which was 27 made possible by the deployment of a modified eMule client interface among the customers of a national 28 Internet service provider (ISP). We discuss the modifications introduced, and analyze the interaction 29 among peers within the community, as well as the interaction of community members with the rest 30 of the world, on the basis of the volumes exchanged. Our conclusions indicate that: (i) ‘‘client-side” anal- 31 ysis is effective to quantify the impact of file-sharing applications over the network; (ii) the presence of 32 NAT devices accounts for misbehaving of p2p applications; (iii) by performing modifications it is possible 33 to sensibly adjust the fairness of a file-sharing service; (iv) the eMule implementation suffers of some 34 design choices that impede the full exploitation of the available resources.

Traffic volume analysis of a nation-wide eMule community

DAVOLI, FRANCO
2008-01-01

Abstract

Peer-to-peer (p2p) content sharing and distribution has become of such widespread use that various user 22 communities, even quite large ones, have formed and may exhibit specific behaviors. The main aspects 23 characterizing the latter may stem from users’ interest groups, specific attitudes, or even from locality 24 reasons – e.g., lying behind a network access translation (NAT) protected Internet access, which is respon- 25 sible for the reduction in transparency of end-to-end applications. We investigate the traffic patterns of 26 one such community, adopting the eMule protocol, by means of an extensive data collection, which was 27 made possible by the deployment of a modified eMule client interface among the customers of a national 28 Internet service provider (ISP). We discuss the modifications introduced, and analyze the interaction 29 among peers within the community, as well as the interaction of community members with the rest 30 of the world, on the basis of the volumes exchanged. Our conclusions indicate that: (i) ‘‘client-side” anal- 31 ysis is effective to quantify the impact of file-sharing applications over the network; (ii) the presence of 32 NAT devices accounts for misbehaving of p2p applications; (iii) by performing modifications it is possible 33 to sensibly adjust the fairness of a file-sharing service; (iv) the eMule implementation suffers of some 34 design choices that impede the full exploitation of the available resources.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11567/217307
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 5
  • ???jsp.display-item.citation.isi??? 4
social impact