• DocumentCode
    2804521
  • Title

    Benchmarking Performance of an MPI-based Solver for 3D Elasticity Problems

  • Author

    Lirkov, Ivan ; Magenov, Svetozar ; Paprzycki, Marcin

  • Author_Institution
    Inst. for Parallel Process., Bulgarian Acad. of Sci., Sofia
  • fYear
    2006
  • fDate
    3-6 Oct. 2006
  • Firstpage
    260
  • Lastpage
    265
  • Abstract
    Numerical solution of 3D linear elasticity equations is considered. Problem is described by a coupled system of second order elliptic partial differential equations. This system is discretized by trilinear parallelepipedal finite elements. Preconditioned conjugate gradient iterative method is used for solving large-scale linear algebraic systems arising after the finite element method (FEM) discretization of the problem. The displacement decomposition technique is applied at the first step to construct a preconditioner using the decoupled block-diagonal part of the original matrix. Then circulant block-factorization is used to precondition thus obtained block-diagonal matrix. Since both preconditioning techniques, displacement decomposition and circulant block-factorization, are highly parallelizable, a portable parallel FEM code based on MPI is developed. Results of numerical tests performed on a number of modern parallel computers using real-life engineering problems from the geomechanics in geosciences are reported and discussed
  • Keywords
    application program interfaces; finite element analysis; matrix algebra; message passing; partial differential equations; 3D elasticity problems; 3D linear elasticity equations; MPI-based solver; benchmarking performance; block-diagonal matrix; circulant block-factorization; decoupled block-diagonal part; displacement decomposition; finite element method; large-scale linear algebraic systems; numerical solution; parallel computers; portable parallel FEM code; preconditioned conjugate gradient iterative method; problem discretization; second order elliptic partial differential equations; trilinear parallelepipedal finite elements; Concurrent computing; Elasticity; Finite element methods; Geology; Iterative methods; Large-scale systems; Matrix decomposition; Partial differential equations; Performance evaluation; Testing;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Modern Computing, 2006. JVA '06. IEEE John Vincent Atanasoff 2006 International Symposium on
  • Conference_Location
    Sofia
  • Print_ISBN
    0-7695-2643-8
  • Type

    conf

  • DOI
    10.1109/JVA.2006.9
  • Filename
    4022071