Name
Abstract | ||
|---|---|---|
In this work, we present new proposals based on the owner-compute rule for the parallelization of irregular loops with dependences. The parallel code increases the available parallelism through the distribution of the statements inside each iteration instead of the whole iterations of the loop. Additionally, our proposal presents as main features the reordering of the layout of the indirection entries, optimizing data locality, and the efficient load balancing. Inspector and executor phases are fully parallel, without synchronizations and uncoupled, allowing the reuse of the information of the inspector. Experimental results on a SGI O2000 system prove that our approach exhibits a high performance, even when compared to well-known parallelization strategies. |
Year | DOI | Venue |
|---|---|---|
2003 | 10.1007/978-3-540-45209-6_43 | LECTURE NOTES IN COMPUTER SCIENCE |
Keywords | Field | DocType |
load balance | Locality,Indirection,Synchronization,Executor,Load balancing (computing),Task parallelism,Parallel algorithm,Computer science,Reuse,Parallel computing,Distributed computing | Conference |
Volume | ISSN | Citations |
2790 | 0302-9743 | 0 |
PageRank | References | Authors |
0.34 | 9 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| David E. Singh | 1 | 191 | 23.13 |
| María J. Martín | 2 | 174 | 27.68 |
| Francisco F. Rivera | 3 | 177 | 26.17 |