Abstract
Address algebras are a powerful set of basic tools for the design and implementation of parallel programs. Address algebras give insightful shorthands for mapping abstract algorithms with their explicit problem dimensions and data communications to real machines with all the implications of communications topologies, fixed processor dimensions and non-uniform data access. These techniques are quite general and should be a fundamental tool for any parallel programmer today. The basic ideas of address algebra will be discussed along with many examples of their practical application.
Preview
Unable to display preview. Download preview PDF.
References
1 J.W. Cooley and J.W. Tukey, Math. Comp. 19, 1965, pp. 297–301.
2 G.C. Fox, et al., Solving Problems on Concurrent Processors, Vol. 1, Prentice Hall,1988.
3 Knuth uses these techniques extensively in his classic: D.E. Knuth, The Art of Computer Programming, Addison-Wesley, 1981.
4 An excellent text is P. Fletcher, Regular Mapping of Multi-Dimensional Data on Parallel Processors, Technical Report TR-HJ-93-05, CSIRO Division of Information Technology, Canberra, Australia, May, 1993.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Brown, J. (1996). Addressing algebra as a tool for parallel program development. In: Waśniewski, J., Dongarra, J., Madsen, K., Olesen, D. (eds) Applied Parallel Computing Industrial Computation and Optimization. PARA 1996. Lecture Notes in Computer Science, vol 1184. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-62095-8_12
Download citation
DOI: https://doi.org/10.1007/3-540-62095-8_12
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-62095-2
Online ISBN: 978-3-540-49643-4
eBook Packages: Springer Book Archive