Abstract
We investigate experimentally a dependency graph approach to the distributed parallel computation of the generic transitive closure problem. A parallel coarse-grained algorithm is derived from a finegrained algorithm. Its advantage is that approximately half of the work is organised as totally independent computation sequences of several processes. We consider conceptual description of dependencies between operations as partial order graphs of events. Such graphs can be split into disjoint subgraphs corresponding to different phases of the computation. This approach helps to design a parallel algorithm in a way which guarantees large independence of actions. We also show that a transformation of the fine-grained algorithm into the coarse-grained is rather nontrivial, and that the straightforward approach does not work.
Work partially supported by GR/N09855 EPSRC grant.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
A. Aho, J. Hopcroft, J. Ullman, The design and analysis of computer algorithms, Addison-Wesley (1974)
K. Chan. A. Gibbons, M. Marcello, W. Rytter, On The PVM Computations of Transitive Closure and Algebraic Path Problem, EuroPVM’98, Springer Verlag, Lecture Notes in Computing 1998
A. Gibbons, W. Rytter, Efficient parallel algorithms, Cambridge University Press (1988)
H.T. Kung and Jaspal Subhlok, A new approach for automatic parallelization of blocked linear algebra computations, 122–129, Supercomputing ‘91. Proceedings of the 1991 Conference on Supercomputing, 1992
Hans-Werner Lang, Transitive Closure on the Instrucion Systolic Array, 295–304, Proc. Int. Conf. on Systolic Arrays, San Diego, K. Bromley, S.Y. Kung, E. Swartzlander, 1991
F.T. Leighton, Introduction to Parallel Algorithms and Architectures: Arrays, Trees, Hypercubes, Morgan Kaufmann Publishers 1991
B. Maggs, S. Plotkin, Minimum cost spanning trees as a path finding problem, IPL 26 (1987) 191–293
Gunter Rote, A systolic array algorithm for the algebraic path problem, Computing, 34,3, 191–219, 1985
Alexandre Tiskin, All-pairs shortest paths computation in the BSP model, Proc. ICALP’01, 2001
Leslie G. Valiant, A bridging model for parallel computation, Communications of the ACM, 33,8, 103–111, 1990
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
Pagourtzis, A., Potapov, I., Rytter, W. (2001). PVM Computation of the Transitive Closure: The Dependency Graph Approach. In: Cotronis, Y., Dongarra, J. (eds) Recent Advances in Parallel Virtual Machine and Message Passing Interface. EuroPVM/MPI 2001. Lecture Notes in Computer Science, vol 2131. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45417-9_35
Download citation
DOI: https://doi.org/10.1007/3-540-45417-9_35
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-42609-7
Online ISBN: 978-3-540-45417-5
eBook Packages: Springer Book Archive