Scheduling convex bipartite communications toward efficient GEN_BLOCK transformations

Authors:
Ching-Hsien Hsu;Shih-Chang Chen;Chao-Yang Lan;Chao-Tung Yang;Kuan-Ching Li
Affiliations:
Department of Computer Science and Information Engineering, Chung Hua University, Hsinchu, Taiwan , R.O.C.;Department of Computer Science and Information Engineering, Chung Hua University, Hsinchu, Taiwan , R.O.C.;Department of Computer Science and Information Engineering, Chung Hua University, Hsinchu, Taiwan , R.O.C.;Department of Computer Science and Information Engineering, Tunghai University, Taichung, Taiwan, R.O.C.;Department of Computer Science and Information Management, Providence University, Taichung, Taiwan,R.O.C.
Venue:
ISPA'05 Proceedings of the Third international conference on Parallel and Distributed Processing and Applications
Year:
2005

Citing 4
Cited 0

Processor reordering algorithms toward efficient GEN_BLOCK redistribution

Proceedings of the 2001 ACM symposium on Applied computing
Symbolic Communication Set Generation for Irregular Parallel Applications

The Journal of Supercomputing
A Divide-and-Conquer Algorithm for Irregular Redistribution in Parallelizing Compilers

The Journal of Supercomputing
An efficient algorithm for irregular redistributions in parallelizing compilers

ISPA'03 Proceedings of the 2003 international conference on Parallel and distributed processing and applications

Quantified Score

Hi-index	0.01

Visualization

Abstract

Irregular data redistribution is used to enhance data locality and algorithm performance on heterogeneous processor systems. In this paper, we present an efficient scheduling algorithm based on convex bipartite communications for irregular GEN_BLOCK transformations. The proposed technique consists of two phases: degree reduction phase, schedules communications involved in processors with degree greater than two; and coloring phase, schedules remaining communications of all processors with degree-2 and degree-1. To evaluate the performance of our algorithm, we have implemented the proposed technique along with three scheduling methods. The simulation results show improvement of total communication costs by the proposed algorithm.