Scalable CFD Computations Using Message-Passing and Distributed Shared Memory Algorithms

  • Authors:
  • Joanna Pazek;Krzysztof Banas;Jacek Kitowski

  • Affiliations:
  • -;-;-

  • Venue:
  • Proceedings of the 7th European PVM/MPI Users' Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

Two parallel programming models applied to an adaptive finite element code for solving nonlinear simulation problems on unstructured grids are compared. As a test case we used a compressible fluid flow simulation where sequences of finite element solutions form time approximations to the Euler equations. In the first model (explicit) the domain decomposition of unstructured grid is adopted, while the second (implicit) uses the functional decomposition - both applied to the preconditioned GMRES method that solves iteratively the finite element system of linear equations. Results for HP SPP1600, HP S2000 and SGI Origin2000 are reported.