Abstract
Languages such as High Performance Fortran are used to implement parallel algorithms by distributing large data structures across a multicomputer system. To reduce the communication time for the redistribution of arrays, we proposes a new scheme, strip mining redistribution.
By using this scheme, the communication overhead is almost completely overlapped with the preceding or following computation. We have implemented a library for two-dimensional arrays using strip mining redistribution, and have achieved a speedup of 1.7 for a 2048×2048 ADI program.
Preview
Unable to display preview. Download preview PDF.
References
Seema Hiranandani, Ken Kennedy, and Chau-Wen Tseng. Compiling Fortran D for distributed-memory machines. Communications ACM, 35(8):66–80, August 1992.
High Performance Fortran Forum. High Performance Fortran Specification Version 1.0. Technical report, Rice University, Houston Texas, December 1992.
K. Kaneko et al. Processing element design for a parallel computer. IEEE MICRO, 10(2):26–38, 1990.
Michaell Noakes, Deborah Wallach, and William Dally. The J-Machine multicomputer: An architectural evaluation. In Intl. Symp. on Computer Architecture, pages 224–235, 1993.
William Press. Numerical Recipes in C, pages 665–666. Cambridge, 1988.
Michael Wolfe. Optimizaing Supercompilers for Supercomputers, pages 97–123. The MIT Press, 1989.
Akiyoshi Wakatani and Michael Wolfe. The effectiveness of message strip mining. In Seventh Int'l Conf. on Parallel and Distributed Computing Systems (submitted), 1994.
Akiyoshi Wakatani and Michael Wolfe. Optimization of the redistribution of arrays for distributed memory multicomputers. In Parallel Computing (submitted), 1994.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1994 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Wakatani, A., Wolfe, M. (1994). A new approach to array redistribution: Strip mining redistribution. In: Halatsis, C., Maritsas, D., Philokyprou, G., Theodoridis, S. (eds) PARLE'94 Parallel Architectures and Languages Europe. PARLE 1994. Lecture Notes in Computer Science, vol 817. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-58184-7_112
Download citation
DOI: https://doi.org/10.1007/3-540-58184-7_112
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-58184-0
Online ISBN: 978-3-540-48477-6
eBook Packages: Springer Book Archive