Automatic generation of efficient array redistribution routines for distributed memory multicomputers
- 19 November 2002
- proceedings article
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 342-349
- https://doi.org/10.1109/fmpc.1995.380436
Abstract
Appropriate data distribution has been found to be critical for obtaining good performance on Distributed Memory Multicomputers like the CM-5, Intel Paragon and IBM SP-1. It has also been found that some programs need to change their distributions during execution for better performance (redistribution). This work focuses on automatically generating efficient routines for redistribution. We present a new mathematical representation for regular distributions called PITFALLS and then discuss algorithms for redistribution based on this representation. A significant contribution of this work is the ability to handle arbitrary source and target processor sets while performing redistribution; another is the ability to handle arbitrary dimensionality for the array being redistributed in a sealable manner. The results presented show low overheads for our redistribution algorithm as compared to naive runtime methods.Keywords
This publication has 6 references indexed in Scilit:
- A compilation system that integrates High Performance Fortran and Fortran MPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Runtime array redistribution in HPF programsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- A Convex Programming Approach for Exploiting Data and Functional Parallelism on Distributed Memory MulticomputersPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1994
- Communication optimization and code generation for distributed memory machinesPublished by Association for Computing Machinery (ACM) ,1993
- Compiler and runtime support for structured and block structured applicationsPublished by Association for Computing Machinery (ACM) ,1993
- Compiling Fortran D for MIMD distributed-memory machinesCommunications of the ACM, 1992