Abstract
In the present work, fine-grain parallel applications of a wide class to be executed on distributed-memory multiprocessors are investigated. The problem of determining the maximum degree of parallelism in the control flow of such applications is under consideration. An efficient procedure (called R-procedure) for solving the indicated problem is introduced and stated. The procedure operates on a control flow graph of an application, and it provides fast calculation of the parallelism degree through the technique of reduction of the control flow graph. The R-procedure is shown to have the time complexity of at most O[n 3/z 3], with n and z standing for the number of processes in the application and parallelism degree, respectively.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Paragon XP/S Product Overview, Beaverton, Ore.: Supercomputer Systems Division, Intel Corp. (1991)
Duato, J., Yalamanchili, S., Ni, L.M.: Interconnection Networks: An Engineering Approach. Los Alamitos, Calif. IEEE CS Press (1997)
Chu, W.W., Lan, L.M-T.: Task Allocation and Precedence Relations for Distributed Real-Time Systems. IEEE Trans. Comput., Vol. C-36. 6 (1987) 667–679
Chu, W.W., et al.: Task Allocation in Distributed Data Processing. IEEE Computer. 11 (1980) 57–69
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Zotov, I.V., Titov, V.S. (2001). An O[n 3/z 3] Reduction Procedure for Determining the Maximum Degree of Parallelism in Parallel Applications. In: Malyshkin, V. (eds) Parallel Computing Technologies. PaCT 2001. Lecture Notes in Computer Science, vol 2127. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-44743-1_14
Download citation
DOI: https://doi.org/10.1007/3-540-44743-1_14
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-42522-9
Online ISBN: 978-3-540-44743-6
eBook Packages: Springer Book Archive