Abstract | ||
---|---|---|
The execution, postponement of tasks due to intertask synchronization for data transfer decreases the execution efficiency of parallel tasks. Especially, when there is intertask-dependent array, such execution. postponement occurs clearly since the size of array data is relatively large. In order to decrease this execution postponement, we proposed incremental processing which controls task execution according to the pattern of data production/consumption of array data between two tasks with intertask-dependent arrays. With incremental processing, two tasks can, be executed incrementally if they have similar access order of array data. In this paper, we propose the access pattern analysis method for intertask-dependent arrays managed individually. |
Year | Venue | Keywords |
---|---|---|
1999 | INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED PROCESSING TECHNIQUES AND APPLICATIONS, VOL VI, PROCEEDINGS | parallelizing compiler, incremental processing, access pattern analysis for array data |
Field | DocType | Citations |
Computer science,Parallel computing,Distributed computing | Conference | 0 |
PageRank | References | Authors |
0.34 | 1 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Katsushi Sanda | 1 | 0 | 0.34 |
Koichi Asakura | 2 | 15 | 9.76 |
Toyohide Watanabe | 3 | 475 | 119.79 |