Compile and Run-Time Support for the Parallelization of Sparse Matrix Updating Algorithms

Authors:
Gerardo Bandera;Manuel Ujaldón;Emilio L. Zapata
Affiliations:
Computer Architecture Department, University of Málaga;Computer Architecture Department, University of Málaga;Computer Architecture Department, University of Málaga
Venue:
The Journal of Supercomputing
Year:
2000

Citing 8
Cited 1

Direct methods for sparse matrices

Direct methods for sparse matrices
Data distributions for sparse matrix vector multiplication

Parallel Computing
A manual for the CHAOS runtime library

A manual for the CHAOS runtime library
Vienna-Fortran/HPF Extensions for Sparse and Irregular Problems and Their Compilation

IEEE Transactions on Parallel and Distributed Systems
An efficient uniform run-time scheme for mixed regular-irregular applications

ICS '98 Proceedings of the 12th international conference on Supercomputing
Modeling set associative caches behavior for irregular computations

SIGMETRICS '98/PERFORMANCE '98 Proceedings of the 1998 ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems
The Sparse Cyclic Distribution against its Dense Counterparts

IPPS '97 Proceedings of the 11th International Symposium on Parallel Processing
Sparse Matrix Block-Cyclic Redistribution

IPPS '99/SPDP '99 Proceedings of the 13th International Symposium on Parallel Processing and the 10th Symposium on Parallel and Distributed Processing

Data Locality Exploitation in Algorithms including Sparse Communications

IPDPS '01 Proceedings of the 15th International Parallel & Distributed Processing Symposium

Quantified Score

Hi-index	0.00

Visualization

Abstract

This work presents a survey of the capabilities that the sparse computation offers for improving performance when parallelized, either automatically or through a data-parallel compiler. The characterization of a sparse code gets more complicated as code length increases: Access patterns change from loop to loop, thus making necessary to redefine the parallelization strategy. While dense computation solely offers the possibility of redistributing data structures, several other factors influence the performance of a code excerpt in the sparse field, like source data representation on file, compressed data storage in memory, the creation of new nonzeroes at run-time (fill-in) or the number of processors available. We analize the alternatives that arise from each issue, providing a guideline for the underlying compilation work and illustrating our techniques with examples on the Cray T3E.