• DocumentCode
    470727
  • Title

    Parallisation of the MLFMM for Distributed Memory Systems

  • Author

    van Tonder, J.J. ; Jakobus, Ulrich

  • Author_Institution
    EM Software & Syst., S.A. (Pty) Ltd., Stellenbosch
  • fYear
    2007
  • fDate
    11-16 Nov. 2007
  • Firstpage
    1
  • Lastpage
    4
  • Abstract
    This paper describes the efficient parallisation of the multilevel fast multipole method (MLFMM) for distributed memory systems. Data is distributed between the processes to obtain a good load balance and to minimise the number of communications. The total runtime efficiency is highly dependent on the efficiency of the matrix-vector-product, as well as the computation of the preconditioner.
  • Keywords
    computational electromagnetics; digital simulation; distributed memory systems; matrix algebra; distributed memory systems; load balance; matrix-vector product; multilevel fast multipole method; Method of moments; fast multipole method; parallel efficiency; preconditioner;
  • fLanguage
    English
  • Publisher
    iet
  • Conference_Titel
    Antennas and Propagation, 2007. EuCAP 2007. The Second European Conference on
  • Conference_Location
    Edinburgh
  • Print_ISBN
    978-0-86341-842-6
  • Type

    conf

  • Filename
    4458409