Minimum distance: a method for partitioning recurrences for multiprocessors
- 1 January 1989
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 38 (8), 1203-1211
- https://doi.org/10.1109/12.30873
Abstract
Parallel execution of nonvectorizable uniform recurrences is considered. When naively scheduled, such recurrences could create unacceptable communication and synchronization on a multiprocessor. The minimum-distance method partitions such recurrences into totally independent computations without increasing redundancy or perturbing numerical stability. The independent computations are well suited for execution on a multiprocessor, but they may not utilize all available processors. How extra processors can be applied to the independent computations is addressed. The methods are especially attractive for multiprocessors comprised of clusters.Keywords
This publication has 7 references indexed in Scilit:
- Independent partitioning of algorithms with uniform dependenciesIEEE Transactions on Computers, 1992
- Minimum distance: a method for partitioning recurrences for multiprocessorsIEEE Transactions on Computers, 1989
- Partitioning and Mapping Algorithms into Fixed Size Systolic ArraysIEEE Transactions on Computers, 1986
- The IBM 3090 system: An overviewIBM Systems Journal, 1986
- Parallelism detection and transformation techniques useful for VLSI algorithmsJournal of Parallel and Distributed Computing, 1985
- Improved Time and Parallel Processor Bounds for Fortran-Like LoopsIEEE Transactions on Computers, 1982
- Time and Parallel Processor Bounds for Fortran-Like LoopsIEEE Transactions on Computers, 1979