Abstract
In irregular scientific computational problems one is periodically forced to choosea delay point where some overhead cost is suffered to ensure correctness, or to improve subsequent performance. Examples of delay points are problem remappings, and global synchronizations. One sometimes has considerable latitude in choosing the placement and frequency of delay points; we consider the problem of scheduling delay points so as to minimize the overal execution time. We illustrate the problem with two examples, a regridding method which changes the problem discretization during the course of the computation, and a method for solving sparse triangular systems of linear equations. We show that one can optimally choose delay points in polynomial time using dynamic programming. However, the cost models underlying this approach are often unknown. We consequently examine a scheduling heuristic based on maximizing performance locally, and empirically show it to be nearly optimal on both problems. We explain this phenomenon analytically by identifying underlying assumptions which imply that overall performance is maximized asymptotically if local performance is maximized.
Similar content being viewed by others
References
S. Baden,Dynamic Load Balancing of a Vortex Calculation Running on Multiprocessors, Technical Report, University of California at Berkley, Department of Computer Science (1986).
G. Cybenko and T. G. Allen, Parallel Algorithms for Classification and Clustering. InProc. of the 31st Annual Int'l Tech. Symp. on Optical and Optoelectronic Applied Science and Engineering, San Diego, California (August 1987).
M. J. Berger and S. H. Bokhari, A Partitioning Strategy for Nonuniform Problems on Multiprocessors,IEEE Trans. on Computers C-36(5):570–580 (May 1987).
E. V. Denardo,Dynamic Programming: Models and Applications, Prentice-Hall, Englewood Cliffs, New Jersey (1982).
W. H. Press, B. P. Flannery, S. A. Teukolsky, and W. T. Vettering,Numerical Recipes. Cambridge University Press, New York (1986).
M. J. Berger and J. Oliger, Adaptive Mesh Refinement for Hyperbolic Partial Differential Equations,J. of Computational Physics 53:484–512 (1984).
D. M. Nicol and J. C. Townsend, Accurate Modeling of Parallel Scientific Computations. InProc. of the 1989 SIGMETRICS Conference, pp. 165–170, Berkeley, California (May 1989).
D. M. Nicol and J. H. Saltz, Dynamic Remapping of Parallel Computations with Varying Resource Demands,IEEE Trans. on Computers 37(9):1073–1087 (September 1988).
L. Lamport, The Parallel Execution of Do Loops,Communications of the ACM 17(2):83–93 (1974).
J. Saltz, Aggregation Methods for Solving Sparse Triangular Systems on Multiprocessors,SIAM J. Sci. and Stat. Computation (to appear).
A. Greenbaum,Solving Sparse Triangular Linear Systems Using Fortran with Parallel Extensions on the NYU Ultracomputer Prototype. Report 99, NYU Ultracomputer Note (April 1986).
Y. Saad and M. Schultz,Parallel Implementations of Preconditioned Conjugate Gradient Methods. Department of Computer Science YALEU/DCS/TR-425, Yale University (October 1985).
M. R. Garey and D. S. Johnson,Computers and Intractability, W. H. Freeman and Co., New York (1979).
H. Elman,Iterative Methods for Large Sparse Nonsymmetric Systems of Linear Equations, Department of Computer Science YALEU/DCS/TR-229, Yale University (April 1982).
N. Matelan, The Flex/32 Multicomputer. InProc. of the 12th Int'l Symp. on Computer Architecture, pp. 209–213, Computer Society Press (June 1985).
Additional information
This research was supported in part by the National Aeronautics and Space Administration under NASA contract NAS1-18107 while the author consulted at ICASE, Mail Stop 132C, NASA Langley Research Center, Hampton, Virginia 23665.
Supported in part by NASA contract NAS1-18107, the Office of Naval Research under Contract No. N00014-86-K-0654, and NSF Grant DCR 8106181.
Rights and permissions
About this article
Cite this article
Nicol, D.M., Saltz, J.H. & Townsend, J.C. Delay point schedules for irregular parallel computations. Int J Parallel Prog 18, 69–90 (1989). https://doi.org/10.1007/BF01409746
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF01409746