• DocumentCode
    1915027
  • Title

    A Low Level Component Model Enabling Performance Portability of HPC Applications

  • Author

    Bigot, Julien ; Zhengxiong Hou ; Perez, C. ; Pichon, Vincent

  • Author_Institution
    Maison de la Simulation, CEA, Gif-sur-Yvette, France
  • fYear
    2012
  • fDate
    10-16 Nov. 2012
  • Firstpage
    701
  • Lastpage
    710
  • Abstract
    Scientific applications are increasingly getting complex, e.g. to improve their accuracy by taking into account more phenomena. Meanwhile, computing infrastructures are continuing their fast evolution. Thus, software engineering is becoming a major issue to achieve easiness of development, portability, maintenance, while achieving high performance. Software component model is a promising approach, which enables to manipulate the software architecture of an application. However, existing models do not provide enough support for performance portability. This paper proposes a low level component model (L2C) that supports directly native connectors for typical scenarios of high performance computing, such as MPI, shared memory and method invocation. We walk through a typical example of stencil computation, i.e. a 2-D Jacobi application with domain decomposition. The experimental results show that L2C can achieve the equivalent performance as native implementations, while gaining benefits such as performance portability on the basis of the software component model.
  • Keywords
    object-oriented programming; parallel processing; software architecture; software maintenance; 2D Jacobi application; HPC application; L2C; MPI; computing infrastructure; domain decomposition; high performance computing; low level component model; performance portability; scientific application; shared memory; software architecture; software component model; software development; software engineering; software maintenance; software portability; stencil computation;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    High Performance Computing, Networking, Storage and Analysis (SCC), 2012 SC Companion:
  • Conference_Location
    Salt Lake City, UT
  • Print_ISBN
    978-1-4673-6218-4
  • Type

    conf

  • DOI
    10.1109/SC.Companion.2012.96
  • Filename
    6495878