Abstract
This paper presents the results of multitasking a Navier-Stokes algorithm on the CRAY-2. The algorithm is a compact difference scheme for the solution of the incompressible, two-dimensional, time-dependent Navier-Stokes equations. Two implementations of multitasking on the CRAY-2 are considered: macrotasking (parallelism at the subroutine level) and microtasking (parallelism at the do-loop level). These two techniques are briefly described. The implementation of the algorithm is discussed in relation to these techniques, and the results for three problem sizes are presented. The timing results for both techniques are, in general, comparable with differences ranging between 2% and 14%, depending on the problem size. The best achieved speedup in a dedicated environment is 3.62 for macrotasking and 3.32 for microtasking. The task granularity for both techniques is computed, and the synchronization costs are estimated. For macrotasks of granularity of up to 0.5 msec, microtasking outperformed macrotasking, while the latter outperformed the former for granularity of over one msec.
Similar content being viewed by others
References
Axelrod, T. S. 1986. Effects of synchronization barriers on multiprocessor performance. J. Parallel Computing, 3: 129–140.
Bieterman, M. 1987. Microtasking general purpose partial differential equation software on the CRAY X-MP. Boeing Tech. Rept. ETA-TR-68, Boeing Computer Services (Nov.).
Chen, S. S., Dongarra, J. J., and Hsiung, C. C. 1984. Multiprocessing linear algebra algorithms on the CRAY X-MP-2: Experiences with small granularity. J. Parallel & Distributed Computing, 1: 22–31.
CRAY-2 Multitasking Programmer's Manual. 1988. Cray Pub. SN-2026 B, Cray Research, Inc. (June).
Fatoohi, R. A., and Grosch, C. E. 1987a. Implementation of an ADI method on parallel computers. J. Scientific Computing, 2 (June), 175–193.
Fatoohi, R. A., and Grosch, C. E. 1987b. Implementation of a four color cell relaxation scheme on the MPP, FLEX/32 and CRAY-2. In Conference Proceedings—The 1987 International Conference on Parallel Processing (St. Charles, Aug. 17–21), Penn State Press, pp. 424–426.
Fatoohi, R. A., and Grosch, C. E. 1988. Implementation and analysis of a Navier-Stokes algorithm on parallel computers. In Conference Proceedings—The 1988 International Conference on Parallel Processing, Vol III (St. Charles, Aug. 15–19), Penn State Press, pp. 235–242.
Gatski, T. B., Grosch, C. E., and Rose, M. E. 1982. A numerical study of the two-dimensional Navier-Stokes equations in vorticity-velocity variables. J. Comput. Phys., 48: 1–22.
Larson, J. L. 1984. Multitasking on the CRAY X-MP-2 multiprocessor. Computer, 17 (July), 62–69.
Mandell, D. A. 1987. Experiences and results multitasking a hydrodynamics code on global and local memory machines. In Conference Proceedings—The 1987 International Conference on Parallel Processing (St. Charles, Aug. 17–21), Penn State Press, pp. 415–420.
Misegades, K., Krause, L., and Booth, M. 1987. Microtasking of fluid mechanics codes on the CRAY X-MP. In Applications of Parallel Processing in Fluid Mechanics, FED—Vol. 47, American Society of Mechanical Engineers, New York, pp. 19–25.
Seager, M. K. 1986. Parallelizing conjugate gradient for the CRAY X-MP. J. Parallel Computing, 3: 35–47.
Swisshelm, J. M. 1988. Multitasking a three-dimensional Navier-Stokes algorithm on the CRAY-2. In Conference Proceedings-Supercomputing '88, Vol II, pp. 14–24, (Orlando, Nov. 14–18), IEEE Computer Society Press.
Author information
Authors and Affiliations
Additional information
This research was supported by NASA Contract No. NAS2-11555 while the author was an employee of Sterling Software under contract to the Numerical Aerodynamic Simulation Systems Divison at NASA Ames Research Center, Moffett Field, CA 94035.
Rights and permissions
About this article
Cite this article
Fatoohi, R.A. Multitasking a Navier-Stokes algorithm on the CRAY-2. J Supercomput 3, 109–124 (1989). https://doi.org/10.1007/BF00129845
Issue Date:
DOI: https://doi.org/10.1007/BF00129845