Locality optimizations for jacobi iteration on distributed parallel systems

Authors:
Yonggang Che;Zhenghua Wang;Xiaomei Li;Laurence T. Yang
Affiliations:
School of Computer, National University of Defense Technology, Changsha, P. R. China;School of Computer, National University of Defense Technology, Changsha, P. R. China;Institute of Equipment and Command Technology, Beijing, P.R. China;Department of Computer Science, St. Francis Xavier University, Antigonish, NS, Canada
Venue:
ISPA'04 Proceedings of the Second international conference on Parallel and Distributed Processing and Applications
Year:
2004

Citing 6
Cited 0

Loop skewing: the wavefront method revisited

International Journal of Parallel Programming
New tiling techniques to improve cache temporal locality

Proceedings of the ACM SIGPLAN 1999 conference on Programming language design and implementation
Architecture-independent parallelism for both shared- and distributed-memory machines using the Filaments package

Parallel Computing
Optimizing inter-nest data locality

CASES '02 Proceedings of the 2002 international conference on Compilers, architecture, and synthesis for embedded systems
Using Time Skewing to Eliminate Idle Time due to Memory Bandwidth and Network Limitations

IPDPS '00 Proceedings of the 14th International Symposium on Parallel and Distributed Processing
(R) Scheduling of Wavefront Parallelism on Scalable Shared-memory Multiprocessors

ICPP '96 Proceedings of the Proceedings of the 1996 International Conference on Parallel Processing - Volume 3

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we propose an inter-nest cache reuse optimization method for Jacobi codes. This method is easy to apply, but effective in that it enhances cache locality of the Jacobi codes while preserving their coarse grain parallelism. We compare our method to two previous locality enhancement techniques that can be used for Jacobi codes: time skewing and new tiling. We quantitatively calculate the main contributing factors to the runtime of different Jacobi codes. We also perform experiments on a PC cluster to verify our analysis. The results show that our method performs poorer than time skewing and new tiling for uniprocessor, but performs better for distributed parallel system.