Impact of the implementation of MPI point-to-point communications on the performance of two general sparse solvers

Authors:
Patrick R. Amestoy;Iain S. Duff;Jean-Yves L'Excellent;Xiaoye S. Li
Affiliations:
ENSEEIHT-IRIT, 2 rue Camichel, F-31071 Toulouse, France;CERFACS, 42 Ave G Coriolis, F-31527 Toulouse Cedex 1, France;INRIA Rhone Alpes and Laboratoire de l'Informatique du Parallélisme, Ecole Normale Supérieure de Lyon, UMR CNRS-ENS Lyon-INRIA 5668, 46 allée d'Italie, F-69364 Lyon Cedex 07, France;NERSC, Lawrence Berkeley National Laboratory, MS 50F, 1 Cyclotron Rd., Berkeley, CA
Venue:
Parallel Computing
Year:
2003

Citing 6
Cited 2

Elimination structures for unsymmetric sparse LU factors

SIAM Journal on Matrix Analysis and Applications
A Supernodal Approach to Sparse Partial Pivoting

SIAM Journal on Matrix Analysis and Applications
Analysis and comparison of two general sparse solvers for distributed memory computers

ACM Transactions on Mathematical Software (TOMS)
MPI: The Complete Reference

MPI: The Complete Reference
A Fully Asynchronous Multifrontal Solver Using Distributed Dynamic Scheduling

SIAM Journal on Matrix Analysis and Applications
SuperLU_DIST: A scalable distributed-memory sparse direct solver for unsymmetric linear systems

ACM Transactions on Mathematical Software (TOMS)

Parallel sparse LU factorization on second-class message passing platforms

Proceedings of the 19th annual international conference on Supercomputing
Parallel sparse LU factorization on different message passing platforms

Journal of Parallel and Distributed Computing

Quantified Score

Hi-index	0.01

Visualization

Abstract

We examine the send and receive mechanisms of MPI and show how to implement message passing robustly so that performance is not significantly affected by changes to the MPI system. We discuss this within the context of two different parallel algorithms for sparse Gaussian elimination: a multifrontal solver (MUMPS), and a supernodal one (SuperLU). The performance of our initial strategies based on simple MPI point-to-point communication primitives is very sensitive to the MPI system, particularly the way MPI buffers are used. Using nonblocking communication primitives improves the performance and robustness, but at the cost of increased code complexity.