Abstract
In this paper, we developed an efficient algorithm for obtaining a compact time-optimal schedule for loops without conditional branches. The running time of our algorithm is O(mn(p 2max +log(nLD))) where n (m) is the number of vertices (edges) in the associated data dependence graph, p max is the denominator of the maximum slope in its irreducible form, L is the maximum number of machine cycles it takes to execute a statement, and D is the maximum number of iterations spanned by a statement. When L and D are polynomially large in n, the running time becomes O(mn 3), which is the best result currently known.
This is a preview of subscription content, log in via an institution.
Preview
Unable to display preview. Download preview PDF.
References
A. Aiken and A. Nicolau, Optimal Loop Parallelization, in Proc. of the SIGPLAN '88 Conference on Programming Language Design and Implementation, 1988, 308–317.
R. Cytron, Doacross: Beyond vectorization for multiprocessors, in Proc. of the 1986 Int'l Conf. on Parallel Processing, Aug. 1986, 836–844.
K. Ebcioglu, A Compilation Technique for Software Pipelining of Loops with Conditional Jumps, in Proc. 20th Annual Workshop in Microprogramming, Dec. 1987, 69–79.
K. Iwano and K. Steiglitz, Testing for cycles in infinite graphs with periodic structure, in Proc. of 19th ACM Symposium on Theory of Computing, 46–55, 1987.
K. Iwano and S. Yeh, An Efficient Algorithm for Optimal Loop Parallelization, Research Report RT 0043, IBM Tokyo Research Laboratory, March, 1990.
R. M. Karp, A characterization of the minimum cycle mean in a digraph, Discrete Mathematics, 23(1978), 309–311.
M. Lam, Software Pipelining: An Effective Scheduling Technique for VLIW Machines, in Proc. of the SIGPLAN '88 Conference on Programming Language Design and Implementation, 1988, 318–328.
A. A. Munshi and B. Simons, Scheduling Sequential Loops on Parallel Processors, in Proc. 1987 Int'l. Conf. on Supercomputing, Lecture Notes in Computer Science, vol. 297, 392–415.
J. B. Orlin, Some problems on dynamic/periodic graphs, in Progress in Combinatorial Optimization, ed. W. R. Pulleyblank, 1984, Academic Press, Orlando, Florida.
J. B. Orlin and R. K. Ahuja, New scaling algorithms for assignment and minimum cycle mean problems, Sloan Working Paper 2019-88, Sloan School of Management, M.I.T., 1988.
B. R. Rau and C. D. Glaeser, Some Scheduling Techniques and an Easily Schedulable Horizontal Architecture for High Performance Scientific Computing, in Proc. 14th Annual Workshop on Micropramming, Dec. 1981, 183–198.
R. Reiter, Scheduling Parallel Computations, Journal ACM, 15 (1968), 59–599.
B. Su, S. Ding, and J. Xie, URPR — An Extension of URCR for Software Pipelining, in Proc. 19th Annual Workshop on Microprogramming, Oct. 1986, 94–103.
A. Zaky and P. Sadayappan, Optimal static scheduling of sequential loops on multiprocessors, in Proc. 18th International Conference on Parallel Processing, Aug. 1989, III-130–137.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1990 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Iwano, K., Yeh, S. (1990). An efficient algorithm for optimal loop parallelization (extended abstract). In: Asano, T., Ibaraki, T., Imai, H., Nishizeki, T. (eds) Algorithms. SIGAL 1990. Lecture Notes in Computer Science, vol 450. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-52921-7_69
Download citation
DOI: https://doi.org/10.1007/3-540-52921-7_69
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-52921-7
Online ISBN: 978-3-540-47177-6
eBook Packages: Springer Book Archive