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.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
G. Bandera. A novel approach for the compilation of parallel sparse matrix updating. Technical report, Computer Architecture Dept., Univ. of Malaga, 1998.
G. Bandera, M. Ujaldón, M. A. Trenas and E. L. Zapata. The sparse cyclic distribution against its dense counterparts. Proceedings 11th Int'l Parallel Processing Symposium, Geneve (Switzerland), April 1997.
G. Bandera and E. L. Zapata. Sparse matriz block-cyclic redistribution. Proceedings 13th Int'l Parallel Processing Symposium. San Juan (Puerto Rico), April, 1999.
R. Barret, M. Berry, T. Chan, J. Demmel, J. Donato, J. Dongarra, V. Eijkhout, R. Pozo and C. Romine y Henk van der Vorst. Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods. Ed. SIAM, 1994.
R. Barriuso and A. Knies. SHMEM User's Guide for C. Cray Research Inc, 1994.
D. R. Chakrabarti, N. Shenoy, A. Choudhary and P. Banerjee. An efficient uniform run-time scheme for mixed regular-irregular applications. In'l Conference on Supercomputing, 1998.
I. S. Duff, A. M. Erisman and J. K. Reid. Direct Methods for Sparse Matrices. Clarendon Press, Oxford, 1986.
I. S. Duff, R. G. Grimes and J. G. Lewis. User's Guide for the Harwell-Boeing Sparse Matrix Collection. Research and Technology Division, Boeing Computer Services, Seattle, Washington, USA, 1992.
B. B. Fraguela, R. Doallo and E. L. Zapata. Modeling set associative caches behaviour for irregular computations. ACM Int'l Conf. on Measurement and Modeling of Computer Systems (SIGMETRICS'98), Madison, Wisconsin, 1998.
S. Pizzanetzky. Sparse Matrix Technology. Academic Press Inc., 1984.
L. F. Romero and E. L. Zapata. Data distributions for sparse matrix vector multiplication solvers. Journal Parallel Computing, 21:583-605, 1995.
J. Saltz, R. Ponnusamy, S. Sharma, B. Moon, Y. Hwang, M. Uysal and R. Das. A manual for the chaos run-time library. Technical report CS-TR-3437. Computer Science Dept, Univ. of Maryland, 1995.
G. P. Trabado, O. Plata and E. L. Zapata. HPF directives and run-time support for molecular dynamic problems. Proceedings 7th Workshop on Compilers for Parallel Computers, pages 43-56. Linkping (Sweden). June, 1998.
M. Ujaldón, E. L. Zapata, B. Chapman and H. P. Zima. Vienna-Fortran/HPF extensions for sparse and irregular problems and their compilation. IEEE Trans. on Parallel and Distributed Systems, 8(10):1068-1083, 1997.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Bandera, G., Ujaldón, M. & Zapata, E.L. Compile and Run-Time Support for the Parallelization of Sparse Matrix Updating Algorithms. The Journal of Supercomputing 17, 263–276 (2000). https://doi.org/10.1023/A:1026563323328
Issue Date:
DOI: https://doi.org/10.1023/A:1026563323328