Abstract
This paper presents extensions of processor assignment statements, the language constructs to support them and the methodology for their implementation on distributed memory computers. The data exchange produced by these algorithms follows a generalization of the concept of hypercubic communication pattern. Due to the efficiency of the introduced weighted dynamic hypercubic communication patterns, the performance of programs built using the methodology are comparable to the obtained using classical message passing programs. The concept of balanced processor assignment statement eases the portability of pram algorithms to multicomputers.
Chapter PDF
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Fortune S., Wyllie, J.: Parallelism in Random Access Machines. STOC. (1978) 114–118
Hansen P. B.: Do hypercubes sort faster than tree machines?. Concurrency: Practice and Experience 6 (2) (1994) 143–151
Hoare, C. A. R.: Proof of a Program: Find. Communications of the ACM. 14 39–45
Murphy, T.A., Vaughan J.: On the Relative Performance of Diffusion and Dimension Exchange Load Balancing in Hypercubes. Proceedings of the PDP’97. IEEE Computer Society Press. (1997) 29–34
Rodríguez, C., Roda J.L., Morales, D. G., Almeida, F.: h-relation Models for Current Standard Parallel Platforms. Proceedings of the Euro-Par’98. Springer Verlag LNCS. (1998)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Rodríguez, C., Sande, F., León, C., Coloma, I., Delgado, A. (1998). Load balancing and processor assignment statements. In: Pritchard, D., Reeve, J. (eds) Euro-Par’98 Parallel Processing. Euro-Par 1998. Lecture Notes in Computer Science, vol 1470. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0057899
Download citation
DOI: https://doi.org/10.1007/BFb0057899
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-64952-6
Online ISBN: 978-3-540-49920-6
eBook Packages: Springer Book Archive