• DocumentCode
    3464216
  • Title

    Arrival and departure processes of nodes in P2P systems

  • Author

    Shibusawa, Susumu ; Kobayashi, Mamoru ; Ohno, Hiroshi ; Yonekura, Tatsuhiro

  • Author_Institution
    Dept. of Comput. & Inf. Sci., Ibaraki Univ., Hitachi, Japan
  • fYear
    2011
  • fDate
    5-9 Dec. 2011
  • Firstpage
    609
  • Lastpage
    614
  • Abstract
    Since the departure of nodes has a large effect on the integrity of a P2P system, there is a strong need for methods that can mitigate the effects of node departure. In this paper we model the departure process of nodes in a P2P system with a renewal process to determine the node departure time for several node departure distributions, and we present connection conditions for active nodes that compensate for node departures. The fact that the departure interval of nodes in a P2P system can be expressed in terms of residual lifetimes is used to derive the expected values of node departure time for several lifetime distributions of nodes. It is also shown that the Laplace transform of the number of departure nodes can be expressed in terms of the node lifetime distribution function. Furthermore, we present several stochastic conditions for connecting active nodes in response to node departures.
  • Keywords
    Laplace transforms; peer-to-peer computing; Laplace transform; P2P systems; active node connection; connection conditions; departure interval; node arrival process; node departure distributions; node departure process; node departure time; node lifetime distribution function; residual lifetimes; stochastic conditions; Conferences; P2P system; distributed stochastic processing; file sharing; node arrival; node departure; renewal process;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    GLOBECOM Workshops (GC Wkshps), 2011 IEEE
  • Conference_Location
    Houston, TX
  • Print_ISBN
    978-1-4673-0039-1
  • Electronic_ISBN
    978-1-4673-0038-4
  • Type

    conf

  • DOI
    10.1109/GLOCOMW.2011.6162523
  • Filename
    6162523