• DocumentCode
    2628177
  • Title

    Solving Markov chains using bounded aggregation on a massively parallel processor

  • Author

    Mattingly, R.B.

  • Author_Institution
    Dept. of Math., Youngstown State Univ., OH, USA
  • fYear
    1993
  • fDate
    1-4 Dec 1993
  • Firstpage
    128
  • Lastpage
    133
  • Abstract
    An SIMD implementation of a method for approximating the stationary distribution vector of a Markov chain is presented. A key feature of the implementation is the simultaneous computation of several matrix inverses. Computational results from a MasPar MP-1 system are discussed
  • Keywords
    Markov processes; aggregation; matrix inversion; parallel processing; Markov chains; MasPar MP-1 system; SIMD implementation; bounded aggregation; massively parallel processor; matrix inverses; stationary distribution vector; Application software; Arithmetic; Computer architecture; Computer networks; Heart; Mathematics; Nearest neighbor searches; Queueing analysis; Registers; Steady-state;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Parallel and Distributed Processing, 1993. Proceedings of the Fifth IEEE Symposium on
  • Conference_Location
    Dallas, TX
  • Print_ISBN
    0-8186-4222-X
  • Type

    conf

  • DOI
    10.1109/SPDP.1993.395541
  • Filename
    395541