Optimal partitioning and granularity of uniform task graphs
- 20 Downloads
Task partitioning is an important technique in parallel processing. In this paper, we investigate the optimal partitioning strategies and granularities of tasks with communications based on several models of parallel computer systems. Different from the usual approach, we study the optimal partitioning strategies and granularities from the viewpoint of minimizingT as well as minimizingNT 2, whereN is the number of processors used andT is the program execution time usingN processors. Our results show that the optimal partitioning strategies for all cases discussed in this paper are the same—either to assign all tasks to one processor or to distribute them among the processors as equally as possible depending only on the functions of ratio of running time to communication timeR/C.
KeywordsExecution Time Parallel Processing Communication Overhead Processor Utilization Optimal Partitioning
Unable to display preview. Download preview PDF.
- H. S. Stone, High-performance computer architecture. Addison-Wesley Publishing Company, 1987.Google Scholar
- Li Guojie and B. W. Wah, Optimal granularity of parallel evaluation of And trees. 1986 Procedings Fall Joint Computer Conference, 297–306, Nov. 1986.Google Scholar
- Boontee Kruatrachue and Ted Lewis, Grain size determination for parallel processing.IEEE Software, 23–32, Jan. 1988.Google Scholar
- Xia Peisuet al., The architecture of BJ parallel computer series (invited paper). Int’l. Conference on Computational Physics, June 1988.Google Scholar