Abstract
The High Performance Fortran (HPF) language and the Message Passing Interface (MPI) are two widely used methods to achieve parallelism on today's clusters and multiprocessor supercomputers. HPF is a distinct language providing extensions to Fortran 90/95 to express parallel execution paths and regions. MPI is a library of communication calls that can be inserted into modern high-level languages (C and Fortran). This paper discusses the use of the two approaches in a parallel finite element application for liquid composite manufacturing process modeling. The unstructured nature of the code provides an excellent opportunity to test both the computation and communication effectiveness of the two approaches. We discuss performance results based on implementations conducted on a modern massively parallel computing platform with a highly tuned processor interconnection network.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Koelbel, C. H., Loveman, D. B., Schreiber, R. S., Steele, G. L., Jr. and Zosel, M. E.: The High Performance Fortran Handbook, The MIT Press, 1994.
Gropp, W., Lusk, E. and Skjellum, A.: Using MPI-Portable Parallel Programming with the Message-Passing Interface, The MIT Press, 1994.
Mohan, R. V., Ngo, N. D. and Tamma, K. K.: On a pure finite-element-based methodology for resin transfer mold filling simulations, Polymer Engrg. Sci. 39(1) (1999), 000–000.
Kumar, V., Grama, A., Gupta, A. and Karypis, G.: Introduction to Parallel Computing: Design and Analysis of Algorithms, Benjamin/Cummings, 1994.
Foster, I.: Designing and Building Parallel Programs, Addison-Wesley, 1995.
Mohan, R. V., Shires, D. R., Mark, A. and Tamma, K. K.: Advanced manufacturing of large scale composite structures: process modeling, manufacturing simulations and massively parallel computing platforms, J. Adv. Engrg. Software 29(3-6) (1998), 249–264.
Carey, G. F., Barragy, E., McIay, R. and Sharma, M.: Element-by-element vector and parallel computations, Comm. Appl. Numer. Methods 4 (1988), 299–308.
Shires, D., Mohan, R. and Mark, A.: Strategies and experiences using high performance Fortran, Technical Report ARL-TR-2437, U.S. Army Research Laboratory, Aberdeen Proving Ground, MD, 2001.
Japan Association for High-Performance Fortran: HPF/JA Language Specification, January 1999.
Karypis, G. and Kumar, V.: METIS: A Software Package for Partitioning Unstructured Graphs, Partitioning Meshes, and Computing Fill-Reducing Orderings of Sparse Matrices, University of Minnesota and the Army HPC Research Center, 1997.
Chilimbi, T., Hill, M. and Larus, J.: Making pointer-based data structures cache conscious, Computer 33(12) (2000), 67–74.
Cuthill, E. and McKee, J.: Reducing the bandwidth of sparse symmetric matrices, In: 24th National Conference, Assoc. Comput. Machinery, 1969, pp. 157–172. HPF AND MPI APPROACHES AND PERFORMANCE 167
Shires, D., Mohan, R. and Mark, A.: Optimization and performance of a Fortran 90 MPIbased unstructured code on large scale parallel systems, Internat. Conf. Parallel and Distributed Processing Techniques and Applications, 2001.
Miles, D.: Portland Group, June 2000. Presentation at the U.S. Army Research Laboratory, Aberdeen Proving Ground, MD.
Portland Group. PGHPF User's Guide - Version 3.0, 2000.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Shires, D., Mohan, R. An Evaluation of HPF and MPI Approaches and Performance in Unstructured Finite Element Simulations. Journal of Mathematical Modelling and Algorithms 1, 153–167 (2002). https://doi.org/10.1023/A:1020534421393
Issue Date:
DOI: https://doi.org/10.1023/A:1020534421393