Cache Optimization For Structured and Unstructured Grid Multigrid

Douglas CC, Hu J, Kowarschik M, Rüde U, Weiß C (2000)


Publication Type: Journal article

Publication year: 2000

Journal

Publisher: Kent State University Library

Book Volume: 10

Pages Range: 21-40

URI: http://www.emis.ams.org/journals/ETNA/vol.10.2000/pp21-40.dir/pp21-40.pdf

Abstract

Many current computer designs employ caches and a hierarchical memory architecture. The speed of a code depends on how well the cache structure is exploited. The number of cache misses provides a better measure for comparing algorithms than the number of multiplies. In this paper, suitable blocking strategies for both structured and unstructured grids will be introduced. They improve the cache usage without changing the underlying algorithm. In particular, bitwise compatibility is guaranteed between the standard and the high performance implementations of the algorithms. This is illustrated by comparisons for various multigrid algorithms on a selection of different computers for problems in two and three dimensions. The code restructuring can yield performance improvements of factors of 2-5. This allows the modified codes to achieve a much higher percentage of the peak performance of the CPU than is usually observed with standard implementations. Copyright © 2000, Kent State University.

Authors with CRIS profile

Related research project(s)

How to cite

APA:

Douglas, C.C., Hu, J., Kowarschik, M., Rüde, U., & Weiß, C. (2000). Cache Optimization For Structured and Unstructured Grid Multigrid. Electronic Transactions on Numerical Analysis, 10, 21-40.

MLA:

Douglas, C. C., et al. "Cache Optimization For Structured and Unstructured Grid Multigrid." Electronic Transactions on Numerical Analysis 10 (2000): 21-40.

BibTeX: Download