The Community for Technology Leaders
Proceedings of the 22nd International Conference on Parallel Architectures and Compilation Techniques (1998)
Paris, France
Oct. 12, 1998 to Oct. 18, 1998
ISSN: 1089-795X
ISBN: 0-8186-8591-3
pp: 358
Jing-Chiou Liou , AT&T Laboratories
Michael A. Palis , Rutgers University
ABSTRACT
In this paper, we present an efficient algorithm, called CASS-II, for task clustering without task duplication. Unlike the DSC algorithm, which is empirically the best known algorithm to date in terms of both speed and solution quality, CASS-II uses only limited ``global'' information and does not recompute the critical path in each refinement step. Therefore, the algorithm runs in $O(|E| + |V| lg |V|)$ which is faster than $O((|V|+|E|)lg |V|)$ of the DSC algorithm. Indeed, our experimental results show that CASS-II is between 3 to 5 times faster than DSC. (It is worth pointing out that we used the C code for DSC developed by the authors of the DSC algorithm. The C code for CASS-II was developed by the authors of this paper.) With respect to solution quality, experimental results show that CASS-II is virtually as good as DSC and, in fact, even outperforms DSC for very fine grain DAGs (granularity less than 0.6).
INDEX TERMS
Resource management, clustering, scheduling, distributed memory multiprocessor, compiler.
CITATION
Jing-Chiou Liou, Michael A. Palis, "A New Heuristic for Scheduling Parallel Programs on Multiprocessor", Proceedings of the 22nd International Conference on Parallel Architectures and Compilation Techniques, vol. 00, no. , pp. 358, 1998, doi:10.1109/PACT.1998.727277
87 ms
(Ver 3.3 (11022016))