Benchmarking FALCON's MATLAB-to-Fortran 90 Compiler on an SGI Power Challenge
Resumo
Esse artigo apresenta uma visão geral do compilador de MATLAB para Fortran 90 desenvolvido para o ambiente de programação para processamento de alto desempenho FALCON. Esse compilador utiliza técnicas estáticas e dinâmicas para inferência. Essas técnicas são otimizadas com análise simbólica e com métodos desenvolvidos para determinação do domínio das variáveis. Os testes de desempenho apresentados nesse artigo mostram que os programas gerados pelo compilador podem ser executados até 1000 vezes mais rápidos que a versão interpretada pelo MATLAB. Esses programas também tem um desempenho muito superior aos programas gerados pelo compilador MATLAB disponível comercialmente. Finalmente, para a mairia dos programas testados, o compilador gerou programas que rodam tão rápido quanto versões escritas a mão para os mesmos algoritmos.
Referências
AHO, A., SETHI, R., AND ULLMAN, J. Compilers: Principies, Techniques and Tools. Addison-Wesley Publishing Company, 1985.
BARRBTT, R., BERRY, M., CHAN, T., DEMMEL, J., DONATO, J., DONGARRA, J., EIJKHOUT, V., POZO, R., ROMINE, C., AND VAN DER VORST, H. Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods. SIAM, 1993.
BLUME, W., DOALLO, R., EIGENMANN, R., GROUT, J., HOEFLINGER, J., LAWRENCE, T., LEE, J., PADUA, D., PAEK, Y., POTTENGER, B., RAUCHWERGER, L., AND TU, P. Parallel Programming with Polaris. IEEE Computer 29, 12 (December 1996), 78-82.
BLUME, W., EIGENMANN, R., FAIGIN, K., GROUT, J., HOEFLINGER, J., PADUA, D., PETERSEN, P., POTTENGER, B., RAUCHWERGBR, L., TU, P., AND WEATHERFORD, S. Polaris: Improving the Effectiveness of Parallelizing Compilers. In Languages and Compilers for Parallel Computing (August 1994), K. Pingali, U. Banerjee, D. Gelernter, A. Nicolau, and D. Padua, Eds., Lecture Notes in Computer Science, vol. 892, Springer-Verlag, pp. 141-154. 7th International Workshop, Ithaca, NY, USA.
BUDD, T. An APL Compiler. Springer-Verlag, 1988.
CYTRON, R., FERRANTE, J., ROSEN, B. K., WEGMAN, M. N., AND ZADECK, F. K. Efficiently Computing Static Single Assignment Form and the Control Dependence Graph. ACM Transactions on Programming Language and Systems 19, 4 (October 1991), 451-490.
DEROSE, L. A. Compiler Techniques for MATLAB Programs. PhD thesis, University of lllinois at Urbana-Champaign, Department of Computer Science, 1996.
DEROSE, L., GALLIVAN, K., GALLOPOULOS, E., MARSOLF, B., AND PADUA, D. FALCON: A MATLAB Interactive Restructuring Compiler. In Languages and Compilers for Parallel Computing (August 1995), C.-H. Huang, P. Sadayappan, U. Banerjee, D. Gelernter, A. Nicolau, and D. Padua, Eds., Lecture Notes in Computer Science, vol. 1033, Springer-Verlag, pp. 269-288. 8th International Workshop, Columbus, Ohio.
DEROSE, L., GALLIVAN, K., GALLOPOULOS, E., MARSOLF, B., AND PADUA, D. FALCON: An Environment for the Development of Scientific Libraries and Applications. In Proc. of the KBUP95: First international workshop on Knowledge-Based systems for the (re)Use of Program libraries (Sophia Antipolis, France, November 1995).
GALLOPOULOS, E., HOUSTIS, E., AND RICE, J. R. Computer as Thinker/Doer: Problem-Solving Environments for Computational Science. IEEE Computational Science & Engineering 1, 2 (Summer 1994), 11-23.
GARCIA, A. L. Numerical Methods for Physics. Prentice Hall, 1994.
GILMAN, L., AND ROSE, A. APL : An lnteractive Approach. Wiley, 1984.
THE MATH WORKS, INC. MATLAB, High-Performance Numeric Computation and Visualization Software. User's Guide, 1992.
THE MATH WORKS, INC. MATLAB Compiler, 1995.
MATHEWS, J. H. Numerical Methods for Mathematics, Science and Engineering, 2nd ed. Prentice Hall, 1992.
SCHWARTZ, J. T. Automatic Data Structure Choice in a Language of a Very High Level. Communications of the ACM 18 (1975), 722-728.
Tu, P., AND PADUA, D. Gated SSA-Based Demand-Driven Symbolic Analysis for Parallelizing Compilers. In Proceedings of the 9th ACM Intemational Conference on Supercomputing (Barcelona, Spain, July 1995), pp. 414-423.
BARRBTT, R., BERRY, M., CHAN, T., DEMMEL, J., DONATO, J., DONGARRA, J., EIJKHOUT, V., POZO, R., ROMINE, C., AND VAN DER VORST, H. Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods. SIAM, 1993.
BLUME, W., DOALLO, R., EIGENMANN, R., GROUT, J., HOEFLINGER, J., LAWRENCE, T., LEE, J., PADUA, D., PAEK, Y., POTTENGER, B., RAUCHWERGER, L., AND TU, P. Parallel Programming with Polaris. IEEE Computer 29, 12 (December 1996), 78-82.
BLUME, W., EIGENMANN, R., FAIGIN, K., GROUT, J., HOEFLINGER, J., PADUA, D., PETERSEN, P., POTTENGER, B., RAUCHWERGBR, L., TU, P., AND WEATHERFORD, S. Polaris: Improving the Effectiveness of Parallelizing Compilers. In Languages and Compilers for Parallel Computing (August 1994), K. Pingali, U. Banerjee, D. Gelernter, A. Nicolau, and D. Padua, Eds., Lecture Notes in Computer Science, vol. 892, Springer-Verlag, pp. 141-154. 7th International Workshop, Ithaca, NY, USA.
BUDD, T. An APL Compiler. Springer-Verlag, 1988.
CYTRON, R., FERRANTE, J., ROSEN, B. K., WEGMAN, M. N., AND ZADECK, F. K. Efficiently Computing Static Single Assignment Form and the Control Dependence Graph. ACM Transactions on Programming Language and Systems 19, 4 (October 1991), 451-490.
DEROSE, L. A. Compiler Techniques for MATLAB Programs. PhD thesis, University of lllinois at Urbana-Champaign, Department of Computer Science, 1996.
DEROSE, L., GALLIVAN, K., GALLOPOULOS, E., MARSOLF, B., AND PADUA, D. FALCON: A MATLAB Interactive Restructuring Compiler. In Languages and Compilers for Parallel Computing (August 1995), C.-H. Huang, P. Sadayappan, U. Banerjee, D. Gelernter, A. Nicolau, and D. Padua, Eds., Lecture Notes in Computer Science, vol. 1033, Springer-Verlag, pp. 269-288. 8th International Workshop, Columbus, Ohio.
DEROSE, L., GALLIVAN, K., GALLOPOULOS, E., MARSOLF, B., AND PADUA, D. FALCON: An Environment for the Development of Scientific Libraries and Applications. In Proc. of the KBUP95: First international workshop on Knowledge-Based systems for the (re)Use of Program libraries (Sophia Antipolis, France, November 1995).
GALLOPOULOS, E., HOUSTIS, E., AND RICE, J. R. Computer as Thinker/Doer: Problem-Solving Environments for Computational Science. IEEE Computational Science & Engineering 1, 2 (Summer 1994), 11-23.
GARCIA, A. L. Numerical Methods for Physics. Prentice Hall, 1994.
GILMAN, L., AND ROSE, A. APL : An lnteractive Approach. Wiley, 1984.
THE MATH WORKS, INC. MATLAB, High-Performance Numeric Computation and Visualization Software. User's Guide, 1992.
THE MATH WORKS, INC. MATLAB Compiler, 1995.
MATHEWS, J. H. Numerical Methods for Mathematics, Science and Engineering, 2nd ed. Prentice Hall, 1992.
SCHWARTZ, J. T. Automatic Data Structure Choice in a Language of a Very High Level. Communications of the ACM 18 (1975), 722-728.
Tu, P., AND PADUA, D. Gated SSA-Based Demand-Driven Symbolic Analysis for Parallelizing Compilers. In Proceedings of the 9th ACM Intemational Conference on Supercomputing (Barcelona, Spain, July 1995), pp. 414-423.
Publicado
07/10/1997
Como Citar
DE ROSE, Luiz; PADUA, David.
Benchmarking FALCON's MATLAB-to-Fortran 90 Compiler on an SGI Power Challenge. In: INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE AND HIGH PERFORMANCE COMPUTING (SBAC-PAD), 9. , 1997, Campos do Jordão/SP.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
1997
.
p. 285-299.
DOI: https://doi.org/10.5753/sbac-pad.1997.22631.
