Using UML State Diagrams for Modelling the Performance of Parallel Programs

Uso de Diagramas de Estado UML para la Modelación del Desempeño de Programas Paralelos

Jorge Ortega Arjona

Departamento de Matemáticas, Facultad de Ciencias, UNAM e–mail: jloa@fciencias.unam.mx

Article received on June 28, 2007
Accepted on October 24,2007

Abstract

There are many possibilities to design a parallel program in order to obtain the best performance possible. The selection of a program structure, as an organisation of processes, impacts on the performance to be achieved, and depends on the problem to be solved. Now, in order to select a program structure as the best in terms of performance, the software designer requires performance modelling techniques to evaluate different alternatives. If the structure of the parallel program can be modelled as a set of interacting processes, described in terms of UML State Diagrams, this paper presents a performance modelling to estimate the average execution time of a parallel program. Performance modelling is achieved by calculating the average execution time of a parallel program, described as a set of processes which run with deterministically and exponentially distributed execution times.

Keywords: .Performance modelling, parallel program, UML State Diagram.

Resumen

Hay muchas posibilidades para diseñar un programa paralelo a fin de obtener el mejor desempeño posible. La selección de una estructura del programa, así como una organización de procesos, impacta sobre el desempeño a lograrse, y depende del problema a resolver. Ahora bien, para seleccionar una estructura del programa como la mejor en términos de desempeño, el diseñador de software requiere de técnicas de modelación para evaluar diferentes opciones. Si la estructura de un programa paralelo puede modelarse como un conjunto de procesos interactivos, descritos en términos de Diagramas de Estado de UML, este artículo presenta una modelación para estimar el tiempo de ejecución promedio de un programa paralelo, descrito como un conjunto de procesos que corren en tiempos de ejecución con distribuiciones determinística y exponencial.

Palabras clave: Modelación de desempeño, programa paralelo, Diagrama de Estado de UML.

DESCARGAR ARTÍCULO EN FORMATO PDF

References

1. Billinton, R., and Allan, R.N. (1992). Reliability Evaluation of Engineering Systems: Concepts and Techniques. Springer.        [ Links ]

2. Booch, G., Rumbaugh, J., and Jacobson I. (1998). The Unified Modeling Language User Guide. Addison–Wesley.        [ Links ]

3. Fowler, M., and Scott, K. (1997). UML Distilled. Addison–Wesley Longman Inc., Reading MA.        [ Links ]

4. Geist, A., Beguelin, A., Dongarra, J., Jiang, W., Manchek, R., and Sunderam, V. (1994). PVM: Parallel Virtual Machine. A User's Guide and Tutorial for Networked Parallel Computing. The MIT Press.        [ Links ]

5. Kleinrock, L. (1975). Queueing Systems. Volume 1: Theory. John Wiley and Sons.        [ Links ]

6. Lui, J.C.S., Muntz, R.R., and Towsley, D. (1998). Computer performance bounds of fork–join parallel programs under a multiprocessor environment. IEEE Transactions on Parallel and Distributed Systems, Vol. 9 No. 3.        [ Links ]

7. Montgomery, D.C. (1991). Design and Analysis of Experiments. John Wiley & Sons, Inc.        [ Links ]

8. Ortega–Arjona, J., and Roberts, G. (1998). Architectural Patterns for Parallel Programming. Proceedings of the 3^rd European Conference on Pattern Languages of Programming and Computing (EuroPLoP 98).        [ Links ]

9. Ortega–Arjona, J. (2000). The Communicating Sequential Elements pattern. An Architectural Pattern for Domain Parallelism. Proceedings of the 7^th Conference on Pattern Languages of Programming (PLoP 2000).        [ Links ]

10. Ortega–Arjona, J. (2004). The Manager–Workers pattern. An Activity Parallelism Architectural Pattern for Parallel Programming. Proceedings of the 9^th European Conference on Pattern Languages of Programming and Computing (EuroPLoP 2004).        [ Links ]

11. Pooley, R., and King, P. (1999) The Unified Modeling Language and Performance Engineering. IEE Proceedings – Software 146(2).        [ Links ]

12. Thomasian, A. and Bay, P. (1986). Analytic Queueing Network Models for Parallel Processing of Task Systems. IEEE Transactions on Computers, December 1986.        [ Links ]

13. Weiss, B. (1999). Introductory Statistics. Addison–Wesley.        [ Links ]