Performance improvement of sparse matrix vector product on vector machines

Authors:
Sunil R. Tiyyagura;Uwe Küster;Stefan Borowski
Affiliations:
High Performance Computing Center Stuttgart, Stuttgart, Germany;High Performance Computing Center Stuttgart, Stuttgart, Germany;NEC HPCE GmbH, Stuttgart, Germany
Venue:
ICCS'06 Proceedings of the 6th international conference on Computational Science - Volume Part I
Year:
2006

Citing 3
Cited 1

An optimal storage format for sparse matrices

Information Processing Letters
Scientific Computations on Modern Parallel Vector Systems

Proceedings of the 2004 ACM/IEEE conference on Supercomputing
Vectorized sparse matrix multiply for compressed row storage format

ICCS'05 Proceedings of the 5th international conference on Computational Science - Volume Part I

Block-Based Approach to Solving Linear Systems

ICCS '07 Proceedings of the 7th international conference on Computational Science, Part I: ICCS 2007

Quantified Score

Hi-index	0.00

Visualization

Abstract

Many applications based on finite element and finite difference methods include the solution of large sparse linear systems using preconditioned iterative methods. Matrix vector multiplication is one of the key operations that has a significant impact on the performance of any iterative solver. In this paper, recent developments in sparse storage formats on vector machines are reviewed. Then, several improvements to memory access in the sparse matrix vector product are suggested. Particularly, algorithms based on dense blocks are discussed and reasons for their superior performance are explained. Finally, the performance gain by the presented modifications is demonstrated.