International Workshop on Collaobration and Cognition in Next Generation Networks

Research Article

Avoiding Greediness in Cooperative Peer-to-Peer Networks

Download
534 downloads
  • @INPROCEEDINGS{10.1007/978-3-642-03354-4_27,
        author={Matthias Brust and Carlos Ribeiro and Jaruwan Mesit},
        title={Avoiding Greediness in Cooperative Peer-to-Peer Networks},
        proceedings={International Workshop on Collaobration and Cognition in Next Generation Networks},
        proceedings_a={CONGN},
        year={2012},
        month={5},
        keywords={Peer-to-Peer Networks Topology Control Cooperation},
        doi={10.1007/978-3-642-03354-4_27}
    }
    
  • Matthias Brust
    Carlos Ribeiro
    Jaruwan Mesit
    Year: 2012
    Avoiding Greediness in Cooperative Peer-to-Peer Networks
    CONGN
    Springer
    DOI: 10.1007/978-3-642-03354-4_27
Matthias Brust1,*, Carlos Ribeiro1,*, Jaruwan Mesit2,*
  • 1: Technological Institute of Aeronautics, Computer Science Division
  • 2: University of Central Florida
*Contact email: matthias.brust@ita.br, carlos.ribeiro@ita.br, jmesit@cs.ucf.edu

Abstract

In peer-to-peer networks, peers simultaneously play the role of client and server. Since the introduction of the first file-sharing protocols, peer-to-peer networking currently causes more than 35% of all internet network traffic— with an ever increasing tendency. A common file-sharing protocol that occupies most of the peer-to-peer traffic is the BitTorrent protocol. Although based on cooperative principles, in practice it is doomed to fail if peers behave greedily. In this work-in-progress paper, we model the protocol by introducing the game named (T4TNT) that gives an interesting access to the greediness problem of the BitTorrent protocol. Simulations conducted under this model indicate that greediness can be reduced by solely manipulating the underlying peer-to-peer topology.