• DocumentCode
    2016983
  • Title

    Dynamic and Distributed Multipath Routing Policy for High-Speed Cluster Networks

  • Author

    Lugones, D. ; Franco, D. ; Luque, E.

  • Author_Institution
    Dept. of Comput. Archit. & Oper. Syst., Univ. Autonoma of Barcelona, Barcelona
  • fYear
    2009
  • fDate
    18-21 May 2009
  • Firstpage
    396
  • Lastpage
    403
  • Abstract
    The increasing demand of parallel applications in cluster computing requires the use of interconnection networks to provide low and bounded communication delays. However, message congestion appears when communication load between nodes is not fairly distributed over the network. Congestion spreading increases latency and reduces network throughput causing important performance degradation. In this paper we present dynamic routing balancing with multipath distribution (DRB-MD), a new method developed to control network congestion based on a uniform balancing of communication load. DRB-MD distributes the traffic load according to a gradual and load-controlled path expansion. It monitors message latency in network switches, makes decisions about how many alternative paths should be used, and finally decides which path (or paths) to use between each source-destination pair. Experiments with permutation patterns and hotspot traffic were conducted to evaluate DRB-MD performance under conditions commonly created by parallel scientific applications.
  • Keywords
    parallel processing; telecommunication congestion control; telecommunication network routing; telecommunication switching; telecommunication traffic; workstation clusters; cluster computing; distributed multipath routing policy; dynamic multipath routing policy; dynamic routing balancing; high-speed cluster networks; interconnection networks; message congestion; message latency monitoring; multipath distribution; network congestion control; network switches; network throughput; parallel scientific applications; traffic load; Communication system control; Computer applications; Computer networks; Concurrent computing; Degradation; Delay; Multiprocessor interconnection networks; Routing; Telecommunication traffic; Throughput; Adaptive routing; High performance networks; communication load balancing; congestion control;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Cluster Computing and the Grid, 2009. CCGRID '09. 9th IEEE/ACM International Symposium on
  • Conference_Location
    Shanghai
  • Print_ISBN
    978-1-4244-3935-5
  • Electronic_ISBN
    978-0-7695-3622-4
  • Type

    conf

  • DOI
    10.1109/CCGRID.2009.13
  • Filename
    5071897