• DocumentCode
    593639
  • Title

    An alternative domain decomposition technique for CUDA-based 3D FDTD methods

  • Author

    Livesey, M. ; Stack, James F. ; Costen, Fumie ; Nanri, T. ; Fujino, Shunya ; Nakashima, Norihiro

  • Author_Institution
    Accenture, Manchester, UK
  • fYear
    2012
  • fDate
    Oct. 31 2012-Nov. 2 2012
  • Firstpage
    469
  • Lastpage
    472
  • Abstract
    The use of General Purpose computing on a GPU is an effective way to accelerate the FDTD method. This paper introduces flexibility to the theoretically best available approach and examines the performance on both Tesla and Fermi architecture GPUs and identifies the best way to determine the GPU parameters for the proposed method.
  • Keywords
    finite difference time-domain analysis; general purpose computers; graphics processing units; parallel architectures; CUDA-based 3D FDTD methods; Fermi architecture GPU; GPU parameters; Tesla architecture GPU; alternative domain decomposition technique; general purpose computing; Bandwidth; Computer architecture; Finite difference methods; Graphics processing units; Instruction sets; Kernel; Time domain analysis; Computational electromagnetics; Electromagnetic fields; Electromagnetic modeling; FDTD; Finite difference methods; GPU; Numerical simulation;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Radar Conference (EuRAD), 2012 9th European
  • Conference_Location
    Amsterdam
  • Print_ISBN
    978-1-4673-2471-7
  • Type

    conf

  • Filename
    6450761