Abstract
Discrete event simulation is usually time consuming. Recently, there has been a great deal of interest in using parallel computers to speed up the simulation process. Before the parallel simulation approach is applied, it is important to understand the inherent parallelism of simulation applications. A simple technique called critical path analysis was proposed to study paralllelism of simulation applications. This paper describes three critical path analysis algorithms based on different event-scheduling (process scheduling) policies. These algorithms are much simpler than a previous approach, where the events must be recorded in a trace and an extra pass is required to process the event trace. In our approach, the critical path analysis algorithms are integrated with the sequential simulation. At the end of the sequential simulation, the optimal parallel execution time is also computed. Livny proposed an algorithm similar to our approach (His, however, was designed for a specific language). Our algorithms can be integrated with sequential simulation programs written by users or be integrated with simulation languages. Another advantage of our algorithms over previous approaches is that ours can be used to study load balancing under different event-scheduling policies. Since our algorithms can be easily inserted in sequential simulation programs, critical path analysis can be applied to existing sequential programs without difficulty. The results can then be used to predict the performance of parallel simulation on similar applications. An example is given to show how useful information can be obtained from our algorithms.
- 1 BAILEY, M., AND LIN, Y.-B. Synchronization strategies for logic-level simulation. To appear in Int. J. Comput. Simul., 1992.]]Google Scholar
- 2 BERRY, O. Performance evaluation of the Time Warp distributed simulation mechanism. Ph.D. dissertation, Univ. of Southern California, May 1986.]] Google Scholar
- 3 BERRY, O. Private communication. 1992.]]Google Scholar
- 4 BERRY, O., AND JEFFERSON, D. Critical path analysis of distributed simulation. In Proceed- ~ngs 1985 SCS Mult~conference on D~str~buted Szmulation (Jan. 1985), pp. 57-60.]]Google Scholar
- 5 CHANDY, K. M., AND MISRA, J. Distributed simulation: A case study in design and verification of distributed programs. 1EEE Trans. Softw. Eng. SE-5, 5 (Sept. 1979), 440-452.]]Google Scholar
- 6 FUJIMOTO, R.M. Lookahead in parallel discrete event simulation. In Proceedings International Conference on Parallel Processing, Vol. III, 1988, pp. 34-41.]]Google Scholar
- 7 FUS~MOTO, R. M. Parallel d~screte event simulation, Commun. ACM 33, 10 (Oct. 1990), 31-53.]] Google Scholar
- 8 HEIDELBERGER, r. Discrete event simulations and parallel processing: Statistical properties. SIAM J. Sci. Stat. Comput. 9, 6 (Nov. 1988), 1114-1132.]] Google Scholar
- 9 JEFFERSON, D. Virtual time. ACM Trans. Program. Lang. Syst. 7, (July 1985), 404-425.]] Google Scholar
- 10 LAMPOR% L. Time, clocks, and the ordering of events in a distributed system. Commun. ACM, 21, (July 1978), 558-565.]] Google Scholar
- 11 LIN, Y.-B., ANn LAZOWSKA, E.D. Optimality considerations for Time Warp parallel simulation. In Proceedings 1990 SCS Multzconference on D~strzbuted Simulation (Jan. 1990), pp. 29-34.]]Google Scholar
- 12 LIN, Y.-B., AND LAZOWSKA, E. D. A time-division algorithm for parallel simulation ACM Trans. Model. Comput. Simul. 1, i (1991), 73-83.]] Google Scholar
- 13 LIN, Y.-B., AND LAZOWSKA, E.D. Effects of process scheduling in parallel smlulation. Int. J. Comput. Szmul., 2, i (1992).]]Google Scholar
- 14 LIN, Y.-B., AND MAK, V.K. On simulating a large-scale mobile radio system. Submitted for publication, 1992.]]Google Scholar
- 15 LIN, Y.-B., LAZOWSKA, E. D., ANn BAER, J.-L. Conservative parallel simulation for systems with no lookahead. In Proceedings 1990 SCS Multzconference on D~strzbuted S~mulat~on (Jan. 1990), pp. 144 149.]]Google Scholar
- 16 LrvNY, M A study of parallelism in distributed simulation. In Proceedzngs 1985 SCS Multtconference on Distrtbuted Simulation (Jan. 1985), pp. 94-98.]]Google Scholar
- 17 LUBACHEVSKY, B. Efficient distributed event-driven simulations of multiple-loop networks. Commun. ACM 21, 2 (March 1989).]] Google Scholar
- 18 NANDY, B., AND LOUCKS, W. M. An algorithm for partitioning and mapping conservative parallel simulation onto multicomputers. In Proceedtngs 6th Workshop on Parallel and Distributed Stmulatzon (Jan. 1992), pp. 139-146]]Google Scholar
- 19 NICOL, D. M. The cost of conservative synchronization in parallel discrete event simulations. Tech. Rep., Dept. of Computer Science, College of Wilham and Mary, 1989.]]Google Scholar
- 20 REIHER, P, FUJIMOTO, R., BELLENOT, S., AND JEFFERSON, D. Cancellation strategies in optimistic execution systems. In Proc. 1990 SCS Multiconference on Distributed Simulation (Jan. 1990), pp. 112-121.]]Google Scholar
- 21 SOKOL, L. M., MUTCHLER, P. A., AND WEISSMAN, J. B. The role of event granularity m parallel simulation design. In Proceedings 6th Workshop on Parallel and Distrzbuted Sirnulatton (Jan. 1992)~ pp. 178-185.]]Google Scholar
- 22 WACNER, D. B., AND LAZOWSKA, E.D. Parallel simulation of queueing networks: Limitations and potentials. In Proceedings 1989 ACM SIGMETRICS and Performance '89 Conference (1989), pp. 146-155.]] Google Scholar
Index Terms
- Parallelism analyzers for parallel discrete event simulation
Recommendations
Parallel discrete event simulation
Special issue on simulationParallel discrete event simulation (PDES), sometimes called distributed simulation, refers to the execution of a single discrete event simulation program on a parallel computer. PDES has attracted a considerable amount of interest in recent years. From ...
Prediction of memory consumption in conservative parallel simulation
This paper proposes a critical path-like analyzer to predict the amount of memory consumed in a specific Chandy-Misra simulation: Segments of code are inserted into the existing sequential simulation program, and this modified simulation program is ...
Prediction of memory consumption in conservative parallel simulation
PADS '95: Proceedings of the ninth workshop on Parallel and distributed simulationThis paper proposes a critical path-like analyzer to predict the amount of memory consumed in a specific Chandy-Misra simulation: Segments of code are inserted into the existing sequential simulation program, and this modified simulation program is ...
Comments