Parallel Performance of a Preconditioned CG Solver
for Unstructured Finite Element Applications
John N. Shadid, Scott A. Hutchinson and Harry K. Moffat
Sandia National Laboratories
Albuquerque, New Mexico 87185
Abstract
A parallel unstructured finite element (FE) implementation designed
for message passing MIMD machines is described. This implementation
employs automated problem partitioning algorithms for load balancing
unstructured grids, a distributed sparse matrix representation of the
global finite element equations and a parallel conjugate gradient (CG)
solver. In this paper a number of issues related to the efficient
implementation of parallel unstructured mesh applications are pre
sented. These include the differences between structured and
unstructured mesh parallel applica tions, major communication kernels
for unstructured CG solvers, automatic mesh partitioning algorithms,
and the influence of mesh partitioning metrics on parallel
performance. Initial results are presented for example finite element
(FE) heat transfer analysis applications on a 1024 proces sor nCUBE 2
hypercube. Results indicate over 95% scaled efficiencies are obtained
for some large problems despite the required unstructured data
communication.