• DocumentCode
    151882
  • Title

    LU decomposition method implementation using graphics processing units - GPU

  • Author

    Gomez, Yensy ; Osorio, John ; Perez, Lina

  • Author_Institution
    Sirius HPC, Univ. Tecnol. de Pereira, Pereira, Colombia
  • fYear
    2014
  • fDate
    3-5 Sept. 2014
  • Firstpage
    184
  • Lastpage
    191
  • Abstract
    This article pretends to show how you can use GPU´s to make the calculations of an LU decomposition and take advantage of the vectorial processors. At the first section we make an introduction of matrix factorization, in the second one we show some features of actual graphics cards, the third one we propose the problem to solve, in the fifth one we expose the implementation and finally the results of the research.
  • Keywords
    graphics processing units; matrix decomposition; parallel processing; GPU; LU decomposition method implementation; graphics cards; graphics processing units; matrix factorization; vectorial processors; Algebra; Computational modeling; Engines; Graphics processing units; Hardware; Presses; Random access memory; ATI; CUDA; GPU; HPC; LU; Matricial; OpenCL; nVidia;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computing Colombian Conference (9CCC), 2014 9th
  • Conference_Location
    Pereira
  • Type

    conf

  • DOI
    10.1109/ColumbianCC.2014.6955342
  • Filename
    6955342