An Assessment of MPI Environments for Windows NT
In this paper we evaluate the MPI environments currently available for Windows NT on the Intel IA32 and Compaq/DEC Alpha architectures. We present benchmark results for low-level communication and for the NAS Parallel Benchmarks to allow comparison with other systems, but our primary interest is determining real application performance and robustness in production cluster environments. For this we use PAFEC-FE, a large FORTRAN code for finite-element analysis. We present results from three MPI implementations, two architectures, and three networking technologies (10 and 100 Mbit/s Ethernet and 1 Gbit/s Myrinet).
Unable to display preview. Download preview PDF.
- 1.D. Ridge, D. Becker, P. Merkey, and T. Sterling. Beowulf: Harnessing the power of parallelism in a pile-of-PC 's. In Proc.1997 IEEE Aerospace Conference.Google Scholar
- 2.D. A. Nicole, K. Takeda, and I. C. Wolton. Running HPC codes on DEC alphas and NT. In Proc. HPCI Conference '98, Manchester, January 1998.Google Scholar
- 3.D. Emerson, K. Maguire, K. Takeda, and D. A. Nicole. An evaluation of commodity supercomputers for CFD applications. In Proc. Parallel CFD '98, Taiwan, May 1998.Google Scholar
- 4.K. Takeda, O. R. Tutty, and D. A. Nicole. Parallel vortex methods on commodity supercomputers; an investigation into bluff body far wake behaviour. In Proc.3rd International Workshop on Vortex Flows and Related Numerical Methods, Toulouse, August 1998.Google Scholar
- 5.S. J. Cox, G. J. Daniell, and D. A. Nicole. Maximum entropy, parallel computation and lotteries. In Proc.1998 International Conference on Parallel and Distributed Processing Techniques and Applications, Las Vegas, pp. 1252–1258, 1998.Google Scholar
- 6.M. A. Baker. MPI on NT: the current status and performance of the available environments. In Proc. EuroPVM/MPI 98, September 1998.Google Scholar
- 7.W. Gropp, E. Lusk, N. Doss, and A. Skjellum. A high-performance portable implementation of the message passing interface. Parallel Computing, 22:457–468, 1996.Google Scholar
- 8.M. Marinho and J. G. Silva. WMPI message passing interface for Win32 clusters. Instituto de Engenharia de Coimbra, Portugal and Departamento de Engenharia Informatica, Universidade de Coimbra, Portugal. http://dsg.dei.uc.pt/wmpi/intro.html.Google Scholar
- 9.L. S. Hebert, W. Seefeld, A. Skjellum, C. D. Taylor, and R. Dimitrov. MPI for NT: two generations of implementations and experience with the message passing interface for clusters and SMP environments. In Proc.1998 International Conference on Parallel and Distributed Processing Techniques and Applications, Las Vegas, July 1998.Google Scholar
- 10.M. Lauria and A. Chien. MPI-FM: high-performance MPI on workstation clusters. Journal of Parallel and Distributed Computing, 40 (1): January 1997.Google Scholar
- 11.D. Bailey, T. Harris, W. Saphir, R. Wijngaart, A. Woo, and M. Yarrow. The NAS parallel benchmarks 2.0. NAS report number NAS–95-020, December 1995.Google Scholar
- 12.P. C. Macey. Advanced design tools for loudspeakers using vibroacoustic finite and boundary element models. Proc. IO, 17 (7):3356–3367, 1995.Google Scholar