Parallel Approaches to a Numerically-Intensive Application Using PVM
This paper proposes some schemes for the parallelization of a quantum reactive scattering code that integrates a two dimensional Schrödinger equation. The parallelization is based on both task farm and pipeline models whose processes communicate via the message-passing paradigm. The various versions of the code are implemented using the Parallel Virtual Machine (PVM) communication library. Benchmarks of the two proposed parallel models are performed on a CRAY T3E system and the measured performances are then discussed.
KeywordsSynchronization Time Communication Time Coupling Matrix Slave Node Load Imbalance
Unable to display preview. Download preview PDF.
- 1.Non Equilibrium Processes in Partially Ionized Gases. M. Capitelli and J.N. Barsdley Eds. Plenum, New York, 1990.Google Scholar
- 5.G. Fox, R. D. Williams, and P. C. Messina. Parallel Computing Works!. Morgan Kaufmann Publishers, Inc, 1994.Google Scholar
- 6.R. Baraglia, D. Laforenza, and A. Laganà. Parallelization Strategies for a Reduced Dimensionality Calculation of Quantum Reactive Scattering Cross section on a Hypercube Machine. Lecture Notes in Computer Science, 919:554–561, 1995.Google Scholar
- 7.A.J.G. Hey. Experiment in MIMD Parallelism. Proc. of International Conference PARLE’89, 28–42, 1989.Google Scholar
- 8.R. Baraglia, R. Ferrini, D. Laforenza, and A. Laganà. On the Optimization of a Task Farm Model for the Parallel Integration of a Two-Dimensional Schrödinger Equations on Distributed Memory Architectures. ICA3PP, IEEE Third International Conference on Algorithms and Architectures for Parallel Processing, 543–556, December 1997.Google Scholar
- 9.R. Baraglia, R. Ferrini, D. Laforenza, and A. Laganà. On the Optimization of a Pipeline Model to Integrate Reduced Dimensionality Schrödinger Equations for Distributed Memory Architectures. The International Journal of Supercomputer Applications and High Performance Computing, Vol. 13, no. 1, Spring 1999, pp. 49–62.CrossRefGoogle Scholar
- 10.A. L. Beguelin, J. J. Dongarra, G. A. Geist, W. Jiang, R. Mancheck, and V. S. Sunderam. PVM: Parallel Virtual Machine: A users’ Guide and Tutorial for Networked Parallel Computing. The MIT Press, 1994.Google Scholar