Abstract
Proposes a novel methodology to develop new parallel loop scheduling algorithms. First, existing algorithms are parameterized into a general framework. Then, a genetic algorithm is used with a shared-memory multiprocessor simulator to »evolve» better strategies. Two new algorithms (called CS-2 and FS-alt, due to their similarity to chunk scheduling and factoring, respectively) are found that perform well compared to existing algorithms.
Original language | English (US) |
---|---|
Title of host publication | Proceedings of the 28th Annual Hawaii International Conference on System Sciences, HICSS 1995 |
Publisher | IEEE Computer Society |
Pages | 187-188 |
Number of pages | 2 |
ISBN (Electronic) | 0818669306 |
DOIs | |
State | Published - 1995 |
Event | 28th Annual Hawaii International Conference on System Sciences, HICSS 1995 - Wailea, United States Duration: Jan 3 1995 → Jan 6 1995 |
Publication series
Name | Proceedings of the Annual Hawaii International Conference on System Sciences |
---|---|
Volume | 2 |
ISSN (Print) | 1530-1605 |
Conference
Conference | 28th Annual Hawaii International Conference on System Sciences, HICSS 1995 |
---|---|
Country/Territory | United States |
City | Wailea |
Period | 1/3/95 → 1/6/95 |
Bibliographical note
Funding Information:This work was supported in part by the National Science Foundation under grant no. MIP-9221900.
Publisher Copyright:
© 1995 IEEE.