Optimization and scaling of shared-memory and message-passing implementations of the ZEUS hydrodynamics algorithm

ABSTRACT

We compare the performance of shared-memory and message-passing versions of the ZEUS algorithm for astrophysical fluid dynamics on a 64-processor HP/Convex Exemplar SPP-2000. Single-processor optimization is guided by timing several versions of simple loops whose structure typifies the main performance bottlenecks. Overhead is minimized in the message-passing implementation through the use of non-blocking communication operations. Our benchmark results agree reasonably well with the predictions of a simple performance model. The message-passing version of ZEUS scales better than the shared-memory one primarily because, under shared-memory, (unless data-layout directives are utilized) the domain decomposition is effectively one-dimensional.

REFERENCES

• 1: Clarke, David A., M. L. Norman, and R. A. Fiedler, "ZEUS-3D User Manual", LCA Technical Report no. 15, 1994.
• 2: Kevin Dowd, High performance computing, O'Reilly & Associates, Inc., Sebastopol, CA, 1993
• 3: Ian Foster, Designing and Building Parallel Programs: Concepts and Tools for Parallel Software Engineering, Addison-Wesley Longman Publishing Co., Inc., Boston, MA, 1995
• 4: Gropp, William, E. Lusk, and A. Skjellum, "Using MPI", MIT Press, 1994.
• 5: Marc Snir , Steve W. Otto , David W. Walker , Jack Dongarra , Steven Huss-Lederman, MPI: The Complete Reference, MIT Press, Cambridge, MA, 1995
• 6: Stone, James M., and M. Norman, Astrophysical Journal Supplement Series, vol. 80, pp. 753-818, 1992.