Dynamic load balancing of atomic structure programs on a PVM cluster
The MCHF package is a suite of programs that enable the calculation of atomic data required by many science and engineering disciplines. As a means of meeting its high computational demands, the package has previously been implemented in PVM. The codes have been used on a dedicated cluster of workstations with a static load balancing scheme. However, the cluster needs to be shared with other users, and different architecture workstations need to be embedded. In this paper, modifications of two well-known dynamic load balancing schemes are implemented and tested. The resulting codes exhibit perfect load balancing for a variety of system loads, facilitating the solution of large problems and the efficient utilization of current resources.
Unable to display preview. Download preview PDF.
- 3.C. C. Douglas, T. G. Mattson, and M. H. Schultz. Parallel programming systems for workstation clusters. Technical Report YALEU/DCS/TR-975, Yale University, 1993.Google Scholar
- 4.C. F. Fischer. The Hartree-Fock Method for Atoms: A numerical approach. J. Wiley & Sons, New York, 1977.Google Scholar
- 5.C. F. Fischer. The MCHF atomic structure package. Comput. Phys. Commun., 64:369, 1991.Google Scholar
- 6.C. F. Fischer, N. S. Scott, and J. Yoo. Multitasking the calculation of angular integrals on the CRAY-2 and CRAY X-MP. Parallel Computing, 8:385, 1988.Google Scholar
- 7.A. Geist, A. Beguelin, J. Dongarra, W. Jiang, R. Manchek, and V. Sunderam. PVM 3 users's guide and reference manual. Technical Report TM-12187, ORNL, 1993.Google Scholar
- 8.S. F. Hummel, E. Schonberg, and L. E. Flynn. Factoring: A practical and robust method for scheduling parallel loops. In Proceedings of Supercomputing '91 Conference, page 610. IEEE Press, Los Alamitos, California, 1991.Google Scholar
- 9.C. Polychronopoulos and D. Kuck. Guided self-scheduling: A practical scheduling scheme for parallel supercomputers. IEEE Transactions on Computers, 36:1425, 1987.Google Scholar
- 10.A. Stathopoulos and C. F. Fischer. A davidson program for finding a few selected extreme eigenpairs of a large, sparse, real, symmetric matrix. Comput. Phys. Commun., 79:268, 1994.Google Scholar
- 11.A. Stathopoulos, A. Ynnerman, and C. F. Fischer. A pvm implementation of the mchf atomic structure package. The International Journal of Supercomputer Applications and High Performance Computing, submitted.Google Scholar
- 12.V. S. Sunderam. A framework for parallel distributed computing. Concurrency: Practice and Experience, 2:315, 1990.Google Scholar
- 13.T. H. Tzen and L. M. Ni. A practical scheduling scheme for parallel computers. IEEE Transactions on parallel and distributed systems, 4:87, 1993.Google Scholar