• DocumentCode
    2539502
  • Title

    Multi-core and multiprocessor implementation of numerical integration in Finite Element Method

  • Author

    Mamza, J. ; Makyla, P. ; Dziekonski, A. ; Lamecki, A. ; Mrozowski, M.

  • Author_Institution
    Dept. of Metrol. & Optoelectron., Gdansk Univ. of Technol., Gdansk, Poland
  • Volume
    2
  • fYear
    2012
  • fDate
    21-23 May 2012
  • Firstpage
    457
  • Lastpage
    461
  • Abstract
    The paper presents techniques for accelerating a numerical integration process which appears in the Finite Element Method. The acceleration is achieved by taking advantages of multi-core and multiprocessor devices. It is shown that using multi-core implementation with OpenMP and a GPU acceleration using CUDA architecture allows one to achieve the speedups by a factor of 5 and 10 on a CPU and GPUs, respectively.
  • Keywords
    finite element analysis; graphics processing units; multiprocessing systems; parallel architectures; CUDA architecture; GPU acceleration; finite element method; multicore implementation; multiprocessor implementation; numerical integration; Decision support systems; CUDA; FEM; GPU; OpenMP; numerical integration;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Microwave Radar and Wireless Communications (MIKON), 2012 19th International Conference on
  • Conference_Location
    Warsaw
  • Print_ISBN
    978-1-4577-1435-1
  • Type

    conf

  • DOI
    10.1109/MIKON.2012.6233633
  • Filename
    6233633