Porting Industrial Codes on High-Performance Computers

and distributed computing using PVM
  • Patrick R. Amestoy
  • Michel J. Daydé
Conference paper
Part of the ERCOFTAC Series book series (ERCO, volume 3)

Abstract

We address the main issues when porting existing codes from serial to parallel computers and when developing portable parallel software on MIMD multiprocessors (shared memory, virtual shared memory, distributed memory multiprocessors, and networks of computers). We especially address distributed/heterogeneous computing on multiprocessors and networks of computers using the PVM programming environment.

We illustrate this by using examples from our experience in porting industrial codes and in designing parallel numerical libraries. We report in some detail on the optimization and the parallelization of scientific applications coming from Centre National d’Etudes Spatiales, from Aerospatiale, and from the ESPRIT III EUROPORT-1 ‘PARALLEL AERO’ project.

Keywords

Virtual Machine Shared Memory Cache Line Matrix Vector Product RISC Processor 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. P.R. Amestoy and I.S. Duff, (1989), Vectorization of a multiprocessor multifrontal code, lilt. J. of Supercomputer Applies., 3, 41 59.Google Scholar
  2. P.R. Amestoy and I.S. Duff, (1993), Memory allocation issues in sparse multiprocessor multifrontal methods, Int. J. of Supercomputer Applies., 7, 64 82.Google Scholar
  3. P.R. Amestoy, M.J. Dayde, I.S. Duff, and P. Morere, (1995), Linear algebra calculations on a virtual shared memory computer, lilt Journal of High Speed Computing, 7, 21 43.Google Scholar
  4. P.R. Amestoy, (1991), Factorization of large sparse matrices based on a multifrontal approach in a multiprocessor environment, phd thesis, Institut, National Polytechnique de Toulouse. Available as CERFACS report TH / PA / 9 1 / 2.Google Scholar
  5. E. Anderson, Z. Bai, C. Bischof, J. Demmel, J. Dongarra, J. Du Croz, A. Greenbaum, S. Hammarling, A. Mckenney, S. Ostrouchov, and D. Sorensen, (1992), LAPACK : A portable linear algebra library for highperformance computers, SIAM, Philadelphia.Google Scholar
  6. A. Beguelin, J. Dongarra, A. Geist, and V. Sunderam, (1993), Visualization and debugging in an heterogeneous environment, IEEE Comp., 26(6),88 95Google Scholar
  7. A. Beguelin, J. Dongarra, A. Geist, R. Manchek, and V. Sunderam, (1991), A users’ guide to PVM Parallel Virtual Machine, Tech. Rep. ORNL/TM-11826, Oak Ridge National Laboratory, Oak Ridge, Tennessee.Google Scholar
  8. A. Beguelin, J. Dongarra, A. Geist, R. Manchek, and V. Sunderam, (1995), Recent enhancements to PVM, lilt Journal of Supercomputer Applications, 9, 108 127.Google Scholar
  9. A. Beguelin, (1993), Xab: a tool for monitoring PVM programs, in Workshop on Heterogeneous Processing, IEEE Computer Society Press, 92 97.Google Scholar
  10. R. Butler and E. Lusk, (1992), Users’Guide to the P4 Parallel Programming System, tech. rep., University of North Florida, Argonne National Laboratory.Google Scholar
  11. J.L. Charles, M.J. Dayde, A. Petitet, L. Prevost and E. Simmonet, (1993), Evaluation de calculateurs rnultiprocesseurs pour les logiciels et bibliotheques scientifiques du CNES: Rapport Final, tech. rep., CERFACS, Toulouse, France.Google Scholar
  12. J. Choi, J. Demmel, I. Dhillon, J. Dongarra, S. Ostrouchov, A. Petitet, K. Stanley, D. Walker, and R.C. Whaley, (1995), ScaLAPACK: A Portable Linear Algebra Library for Distributed Memory Computers - Design Issues and Performance, Tech. Rep. LAPACK Working Note 9 5, CS - 9 5 - 2 8 3, University of Tennessee.Google Scholar
  13. J. Choi, J. Dongarra, S. Ostrouchov, A. Petitet, D. Walker, AND R.C. Whaley, (1995), A Proposal for a Set of Parallel Basic Linear Algebra Subprograms, Tech. Rep. LAPACK Working Note 100, CS - 9 5 - 2 8 3, University of Tennessee.Google Scholar
  14. M.J.Daydé and I.S. Duff, (1991), Use of Level 3 BLAS in LU factorization in a multiprocessing environment on three vector multiprocessors, the ALLIANT FX/80, the CRAY-2, and the IBM 3090/VF, Int. J. of Supercomputer Applies., 5, 92 110.Google Scholar
  15. M.J.Daydé and I.S. Duff, (1995), Porting industrial codes and developing sparse linear solvers on parallel computers, Computing Systems in Engineering, 6, 295 305.CrossRefGoogle Scholar
  16. M.J. DaydéND I.S. Duff, (1996), A Block Implementation of Level 3 BLAS for RISC Processors, Tech. Rep. t o appear, ENSEEIHT-IRIT.Google Scholar
  17. M.J. Daydé, I.S. Duff, and A. Petitet,(1992), A Parallel Block Implementation of Level 3 BLAS Kernels for MIMD Vector Processors, Tech. Rep. TR/PA/92/74, CERFACS, Toulouse, France.Google Scholar
  18. M.J. Daydé, I.S. Duff, J.Y. L’Cellent, and L. Giraud, (1993), Evaluation d’ordinateurs vectoriels et paralleles sur un jeu de programmes representatifs des calculs a la division avions de I’Aerospatiale : Rapport Final, Tech. Rep. FR /PA/93/19, CERFACS, Toulouse, France.Google Scholar
  19. M.J. Daydé V. Van Kemenade, and J.B. Vos, (1995), Description, Validation and Evaluation of the New Parallel Version of NSMB, Tech. Rep. Esprit Project 8421 : PARALLEL AERO, WP6.1, Deliverable RD21.Google Scholar
  20. J.J.Dongarra and E. Grosse, (1987), Distribution of Mathematical Software Via Electronic Mail, Comm. ACM, 30, 403 407.CrossRefGoogle Scholar
  21. J. Dongarra and R.C. Whaley, (1995), A Us ers’Guide to the BLACS, Tech. Rep. CS-95-281, University of Tennessee, Knoxville, Tennessee, USA.Google Scholar
  22. J.J. Dongarra, J. Du Croz, I.S. Duff, and S. Hammarling, (1990A), Algorithm 679. A set of Level 3 Basic Linear Algebra Subprograms., ACM Transactions on Mathematical Software, 16, 1 17.Google Scholar
  23. J.J. Dongarra, J. Du Croz, I.S. Duff, and S. Hammarling, (1990b), Algorithm 679. A Set of Level 3 Basic Linear Algebra Subprograms: model implementation and test programs, ACM Transactions on Mathematical Software, 16, 18 28.MATHCrossRefGoogle Scholar
  24. J.J. Dongarra, J.J. Du Croz, S. Hammarling, and R.J. Hanson, (1988), An extented set of Fortran Basic Linear Algebra Subprograms, ACM Trans. Math. Softw., 14, 17 and 18 32.Google Scholar
  25. J. Dongarra, R. Hempel, A.J.G. Hey, and D.W. Walker, (1995), MPI ; A Message Passing Interface Standard, Int Journal of Supercomputer Applications, 8, (3/4).Google Scholar
  26. J.J. Dongarra, (1992), Performance of Various Computers Using Standard Linear-Algebra Software, Tech. Rep. CS-89-85, University of Tennessee, Knoxville, Tennessee, USA.Google Scholar
  27. I.S. Duff and J.K. Reid, (1983), The multifrontal solution of indefinite sparse symmetric linear systems, ACM Transactions oil Mathematical Software, 9, 302 325.MathSciNetGoogle Scholar
  28. I.S. Duff and J.K. Reid, (1984), The multifrontal solution of unsymrrietric sets of linear systems, SIAM Journal on Scientific and Statistical Computing, 5, 633 641.MathSciNetGoogle Scholar
  29. I.S Duff, R.G. Grimes, and J.G. Lewis, (1992), Users’Guide for the Harwell-Boeing Sparse Matrix Collection (Release I), Technical Report RAL 9 2 - 0 8 6, Rutherford Appleton Laboratory.Google Scholar
  30. Fortran 90, (1991), ISO/IEC 1539:1991 (E) and now ANSI X3.198-1992, tech. rep.Google Scholar
  31. A. Geist, A. Beguelin, J. Dongarra, W. Jiang, R. Manchek, and V. Sunderam, (1993), PVM 3 User’s Guide and Reference Manual, Tech. Rep. ORNL/TM-12187, Engineering Physics and Mathematics Division, Oak Ridge National Laboratory, Tennessee.Google Scholar
  32. S. Hariri and A. Varma, (1993), High-Performance Distributed Computing : Promises and Challenges, Concurrency : Practice mid Experience, 5, 233 238.Google Scholar
  33. R. Hempel, H. - C. Hope, U. Keller, and W. Krotz, (1995), PARM ACS V6.1 Specification, Tech. Report GmbH Technical Report, PALLAS.Google Scholar
  34. High Performance Fortran Forum, (1993), High Performance Fortran Language Specification, tech. rep., Rice University, Houston, Texas.Google Scholar
  35. C.L. Lawson, R.J. Hanson, D.R. Kincaid, and F.T. Krogh, (1979a), Basic Linear Algebra Subprograms for Fortran Usage, ACM Transactions oil Mathematical Software, 5, 308 323.Google Scholar
  36. C.L. Lawson, R.J. Hanson, D.R. Kincaid, and F. T. Krogh, (1979b), Algorithm 539. Basic linear algebra subprograms for Fortran usage, ACM Trans. Math. Softw., 5, 324 325.Google Scholar
  37. R. Schreiber. and H.D. Simon, (1992), Towards the Teraflop in CFD, tech. rep., NASA Ames Research Center, Moffett, Field, CA.Google Scholar
  38. V. Van Kemenade, M.J. Dayde, and J.B. Vos, (1995), Parallel Navier-Stokes Multi-Block Code to Solve Industrial Aerodynamic Design Problems on High Performance Computers, in Proceedings of HPCN 95 Europe, Milano, Italy.Google Scholar

Copyright information

© Kluwer Academic Publishers 1996

Authors and Affiliations

  • Patrick R. Amestoy
    • 1
  • Michel J. Daydé
    • 1
  1. 1.ENSEEIHT-IRITToulouseFrance

Personalised recommendations