Abstract
Alignment and distribution of array data should be managed by optimizing compilers for parallel computers, but current approaches to the distribution problem formulate it as an NP-complete graph optimization problem. The graphs arising in applications are large and difficult to optimize. In this paper, we improve some earlier results on methods that use graph contraction to reduce the size of a distribution problem. We report on an experiment using seven example programs that show these contraction operations to be effective in practice; we obtain from 70 to 99 percent reductions in problem size, the larger number being more typical, without loss of solution quality.
The work of these authors was supported by the NAS Systems Division via Contract NAS 2-13721 between NASA and the Universities Space Research Association (USRA).
Copyright 1995 by Xerox Corporation. All rights reserved.
The work of this author was supported by the NAS Systems Division via Contract NAS 2-13721 between NASA and the Universities Space Research Association (USRA) while he was a postdoctoral scientist at RIACS. Current affiliation: University of North Carolina, Chapel Hill.
Preview
Unable to display preview. Download preview PDF.
References
R. Bixby, K. Kennedy, and U. Kremer. Automatic data layout using 0–1 integer programming. Technical Report CRPC-TR93349-S, Center for Research on Parallel Computation, Rice University, Houston, TX, November 1993.
S. Chatterjee, J. R. Gilbert, and R. Schreiber. Mobile and replicated alignment of arrays in data-parallel programs. In Proceedings of Supercomputing'93, pages 420–429, Portland, OR, November 1993.
S. Chatterjee, J. R. Gilbert, R. Schreiber, and T. J. Sheffler. Array distribution in data-parallel programs. In K. Pingali, U. Banerjee, D. Gelernter, A. Nicolau, and D. Padua, editors, Proceedings of the Seventh Annual Workshop on Languages and Compilers for Parallel Computing, number 892 in Lecture Notes in Computer Science, pages 76–91, Ithaca, NY, August 1994. Springer-Verlag. Also available as RIACS Technical Report 94.09.
S. Chatterjee, J. R. Gilbert, R. Schreiber, and T. J. Sheffler. Modeling data-parallel programs with the alignment-distribution graph. Journal of Programming Languages, 2:227–258, 1994. Special issue on compiling and run-time issues for distributed address space machines.
J. R. Gilbert, C. Moler, and R. Schreiber. Sparse matrices in MATLAB: Design and implementation. SIAM J. Matrix Anal. Appl., 13(1):333–356, January 1992.
A. V. Goldberg and R. E. Tarjan. A new approach to the maximum-flow problem. J. ACM, 35(4):921–940, October 1988.
M. Gupta. Automatic Data Partitioning on Distributed Memory Multicomputers. PhD thesis, University of Illinois at Urbana-Champaign, Urbana, IL, Sept. 1992. Available as technical reports UILU-ENG-92-2237 and CRHC-92-19.
P. Hough and T. J. Sheffler. A performance analysis of collective communication on the CM-5. RIACS Technical Report in preparation.
U. Kremer. NP-completeness of dynamic remapping. Technical Report CRPC-TR93-330-S, Center for Research on Parallel Computation, Rice University, August 1993. Appears in the Proceedings of the Fourth Workshop on Compilers for Parallel Computers, Delft, The Netherlands, December 1993.
D. J. Palermo and P. Banerjee. Automatic selection of dynamic data partitioning schemes for distributed-memory multicomputers. Talk presented at the Workshop on Automatic Data Layout and Performance Prediction. Center for Research on Parallel Computing. April 19, 1995. Rice University.
T. J. Sheffler, R. Schreiber, J. R. Gilbert, and S. Chatterjee. Aligning parallel arrays to reduce communication. In Proceedings of Frontiers '95: The Fifth Symposium on the Frontiers of Massively Parallel Computation., pages 324–331, February 1995.
S. Wholey. Automatic Data Mapping for Distributed-Memory Parallel Computers. PhD thesis, School of Computer Science, Carnegie Mellon University, Pittsburgh, PA, May 1991. Available as Technical Report CMU-CS-91-121.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Sheffler, T.J., Schreiber, R., Pugh, W., Gilbert, J.R., Chatterjee, S. (1996). Efficient distribution analysis via graph contraction. In: Huang, CH., Sadayappan, P., Banerjee, U., Gelernter, D., Nicolau, A., Padua, D. (eds) Languages and Compilers for Parallel Computing. LCPC 1995. Lecture Notes in Computer Science, vol 1033. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0014212
Download citation
DOI: https://doi.org/10.1007/BFb0014212
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-60765-6
Online ISBN: 978-3-540-49446-1
eBook Packages: Springer Book Archive