GRANLOG: Um Modelo para Análise Automática de Granulosidade na Programação em Lógica
Resumo
Este texto descreve um estudo sobre análise automática de granulosidade na programação em lógica. A análise de granulosidade determina o tamanho dos grãos, ou seja, a complexidade dos módulos que serão executados seqüencialmente num único processador. A análise de granulosidade é vital para exploração eficiente do paralelismo. Inicialmente, são apresentados os princípios básicos que motivam e orientam o desenvolvimento deste trabalho. Logo após, é apresentado um modelo para geração de informações de granulosidade na programação em lógica, denominado GRANLOG. Finalmente, são descritas duas aplicações para o modelo proposto, ou seja, auxilio a decisões de escalonamento e simulação da execução de programas.
Referências
CHANG. Jung-Herng e DESPAIN, Alvin M. Semi-lntelligent Backtracking or Prolog Based on Static Data Dependency Analysis. Proceedings of the Symposium on Logic Programming. IEEE. p.10-21, 1985.
DEBRAY. Saumya K. Static Inference or Modes and Data Depencies in Logic Programs. ACM Transactions on Programming Languages and Systems, v.11, n.3, p.418-450, July 1989.
DEBRAY. Saumya K. e LIN. Nai-Wei. Cosi Analysls or Logic Programs. ACM Transactions on Programming Languages and Systems, v.15, n.5, p.826-875, November 1993.
DEBRAY, Saumya K.; GARCÍA, Pedro L ; HERMENEGILDO. Manuel e LIN, Nai-Wei. Estimating the Computacional Cost or Logic Programs. Proceedings of the International Static Analysis Symposium. Namur, Belgium, Springer-Verlag Lecture Notes in Computer Science, v.864, September 1994.
GARCÍA. Pedro L; HERMENEGILDO, Manuel e DEBRAY, Saumya K. Towards Granularity Based Control or Parallelism in Logic Programs. Proceedings of the lnternational Symposium on Parallel Computation, Linz. Austra, September 1994.
HERMENEGILDO, M. e ROSSI, F. On the Correctness and Efficiency or Independent And Parallelism in Logic Programs. Proceedings of the 1989 North American Conference on Logic Programs, MIT Press, pp. 369-390, October 1989.
KALÉ L. V. Completeness and Full Parallelism or Parallel Loglc Programming Schemes. Proceedings of the Fourth IEEE Symposium on Logic Programming, IEEE. p.125-133, 1987.
KING, A. e SOPER, P. Granularity Control for Concurrent Logic Programs. Proceedings of the International Symposium on Computer and Information Sciences, Nevsehir, Turkey, October 1990.
KRUATRACHUE. Boontee e LEWIS, Ted. Grain Size Determination for Parallel Processing. IEEE Software, p.23-32, January 1988.
LIN, Nai-Wei. Automatic: Complexity Analysis or Logic Programs. University of Arizona, 1993. (Ph.D. Thesis).
MCCREARY, Carolyn e GILL, Helen. Automatic Determination of Grain Size for Efficient Parallel Processing. Communications of the ACM, v.32, n.9, p.1073-1078, September 1989.
MELLISH, C.S. Some Global Otimizations for a Prolog Computer. Journal of Logic Programming, April 1985.
MISHRA, P. Toward a Theory or Types In Prolog. Proceedings of the First IEEE Symposium on Logic Programming, IEEE. p. 289-298, 1984.
SARKAR, V. Partitioning and Scheduling Parallel Programs for Multiprocessors. MIT Press, 1989.
TICK, Evan. Compile-Time Granularity Analysis or Parallel Logic: Programming Languages. New Generation Computing, v.7, n.2-3, p.325-337, 1990.
WERNER, Otilia; YAMIN. Adenauer C.; BARBOSA. Jorge L V. e GEYER, Cláudio F. R. OPERA Project: an Approach Towards Parallelism Exploitation on Logic Programming. Proceedings of the Tenth Logic Programming Workshop, Zurich, Institui für Informatik der Univesität Zürich, September 1994.
YARDENI, E. e SHAPIRO, E. A Type System for Logic: Programs. Concurrent Prolog: Collected Papers, MIT Press, v.2, p.211-244, 1987.
ZHONG, X.; TICK, E.; DUVVURU, S.; HANSEN. L.; SASTRY A.V.S. e SUNDARARAJAN, R. Towards an Efficient Compile-Time Granularity Analysis Algorithm. Proceedings of the Fifth Generation Computer Systems, Tokyo, Japan. p.809-816, 1992.