• DocumentCode
    2634465
  • Title

    Efficient algorithms for global data communication on the multidimensional torus network

  • Author

    Fragopoulou, Paraskevi ; Akl, Selim G.

  • Author_Institution
    Dept. of Comput. & Inf. Sci., Queen´´s Univ., Kingston, Ont., Canada
  • fYear
    1995
  • fDate
    25-28 Apr 1995
  • Firstpage
    324
  • Lastpage
    330
  • Abstract
    Efficient interprocessor communication is crucial to increasing the performance of parallel multiprocessors. In this paper, a special framework is developed on the multidimensional torus, a network that is currently receiving considerable attention. Using this framework as the basic tool, or spanning graph with special properties, to fit various communication needs, is constructed on the network. The importance of this graph is demonstrated with the development of optimal algorithms for three fundamental communication problems, namely the multinode broadcasting and the single-node and multinode scattering under the store-and-forward, all-port communication model
  • Keywords
    multiprocessing systems; parallel algorithms; performance evaluation; algorithms; all-port communication model; global data communication; interprocessor communication; multidimensional torus network; multinode broadcasting; optimal algorithms; parallel multiprocessors; performance; scattering; spanning graph; store-and-forward; Broadcasting; Computer networks; Concurrent computing; Costs; Data communication; Independent component analysis; Information science; Multidimensional systems; Scattering;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Parallel Processing Symposium, 1995. Proceedings., 9th International
  • Conference_Location
    Santa Barbara, CA
  • Print_ISBN
    0-8186-7074-6
  • Type

    conf

  • DOI
    10.1109/IPPS.1995.395952
  • Filename
    395952