• DocumentCode
    2028350
  • Title

    Nearby Neighbor Selection in P2P Systems to Localize Traffic

  • Author

    Sheng, Lijie ; Wen, HaoYu

  • Author_Institution
    Sch. of Comput. Sci. & Technol., Xidian Univ., Xi´´an
  • fYear
    2009
  • fDate
    24-28 May 2009
  • Firstpage
    68
  • Lastpage
    73
  • Abstract
    Peer-to-peer (P2P) software as major Internet applications bring convenience and pleasure to users, while it becomes a nightmare for Internet service providers (ISPs) because they produce huge traffic on the network. The enormous traffic come from the fact that when P2P systems construct overlay networks, they ignore the underlying network topology, so most of the connections are between remote peers and most of the traffic are cross-network traffic. To overcome this, we present a simple mechanism named LHF (least hops first) to construct overlay network mostly with nearby peers. With analysis we show that P2P network constructed by LHF is ldquosmall worldrdquo network, in which most of the connections are local ones, while a few remote connections accelerate the system dramatically. Simulation experiments demonstrate that LHF can decrease cross-network traffic remarkably, and can improve the performance of P2P application at the same time.
  • Keywords
    Internet; peer-to-peer computing; telecommunication network topology; telecommunication traffic; Internet; P2P systems; cross-network traffic; least hops first; nearby neighbor selection; network topology; overlay networks; peer-to-peer software; Application software; Computer network management; Computer science; Conference management; Electronic mail; IP networks; Network topology; Shape measurement; Telecommunication traffic; Web and internet services; BitTorrent; LHF; P2P; neighbor selection; overlay network; peer-to-peer;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Internet and Web Applications and Services, 2009. ICIW '09. Fourth International Conference on
  • Conference_Location
    Venice/Mestre
  • Print_ISBN
    978-1-4244-3851-8
  • Electronic_ISBN
    978-0-7695-3613-2
  • Type

    conf

  • DOI
    10.1109/ICIW.2009.18
  • Filename
    5072499