Dual-Level Parallel Analysis of Harbor Wave Response Using MPI and OpenMP

Abstract

The authors describe their experiences converting an existing serial production code to a parallel code combining both MPI and OpenMP. Such dual-level parallel codes will be able to take full advantage of the emerging class of high performance computer architectures using small clusters of shared-memory processors connected via a message-passing network. While the focus is restricted to a harbor response simulation code, the techniques presented herein are appropriate for a broad class of applications that explore a parameter space. The code modifications reduced the execution time of one test case from 3100 minutes on a single CPU to just over 12 minutes on 256 CPUs. Results demonstrate that dual-level parallelism allows substantial increases in model resolution combined with improvements in simulation turnaround time but, contrary to conventional wisdom, requires very little source code alteration.

Get full access to this article

View all access options for this article.

References

Bova, S. , Breshears, C. , Eigenmann, R. , Gabb, H. , Gaertner, G. , Kuhn, B. , Magro, B. , Salvini, S. , and Vatsa, V. 1999. Combining message-passing and directives in parallel applications . SIAM News 32(9).

Breshears, C. P. , and Fagg, G. E. 1998. A computation allocation model for distributed computing under MPI_Connect . In Proceedings of First Southern Conference on Computing.

Demirbilek, Z. , and Panchang, V. 1998. CGWAVE: A coastal surface water wave model of the mild-slope wave equation. Technical Report CHL-98-26, U.S. Army Waterways Experiment Station .

Dowd, K. , and Severance, C. 1998. High Performance Computing. Sebastopol, CA: O’Reilly and Associates .

El-Rewini, H. , and Lewis, T. G. 1998. Distributed and Parallel Computing. Greenwich, CT: Manning .

Fagg, G. E. , and Dongarra, J. J. 1996. PVMPI: An integration of the PVM and MPI systems. Technical Report UT-CS-96-328, University of Tennessee, Knoxville .

Foster, I. 1997. Designing and Building Parallel Programs: Concepts and Tools for Parallel Software Engineering. Reading, MA: Addison-Wesley .

Geist, A. , Beguelin, A. , Dongarra, J. , Jiang, W. , Manchek, R. , and Sunderam, V. 1994. PVM: A Users’ Guide and Tutorial for Network Parallel Computing. Cambridge: MIT Press .

Harkins, G. , Puckette, P. , and Dorrell, C. 1997. Physical model studies of Ponce de Leon Inlet, Florida. Technical Report CHL-97-23, Waterways Experiment Station .

10.

Leasure, B. , ed. 1994. Parallel Processing Model for High Level Programming Languages. X3H5/94-SD2 Revision L. ANSI.

11.

Nichols, B. , Buttlar, D. , and Farrell, J. P. 1996. Pthreads Programming. Sebastopol, CA: O’Reilly and Associates .

12.

OpenMP Architecture Review Board . 1997. OpenMP Fortran Application Program Interface, Version 1.0 [Online]. Available: http://www.openmp.org.

13.

Panchang, V. , Pearce, B. , Wei, G. , and Cushman-Roisin, B. 1991. Solution of the mild-slope wave problem by iteration . Applied Ocean Research 13: 187-199 .

14.

Smith, S. , and Smith, J. 1999. Numerical modeling of waves at Ponce de Leon Inlet, Florida . Journal of Waterway, Port, Coastal, and Ocean Engineering. Submitted for publication.

15.

Snir, M. , Otto, S. , Huss-Lederman, S. , Walker, D. , and Dongarra, J. 1998. MPI—The Complete Reference: Vol. 1. The MPI Core. 2nd ed. Cambridge: MIT Press .

16.

Xu, B. , Panchang, V. , and Demirbilek, Z. 1996. Exterior reflections in elliptic harbor wave models . Journal of Waterway, Port, Coastal, and Ocean Engineering 122: 118-126 .