A testbed for parallel simulation performance prediction

ABSTRACT

The overwhelming complexity of influencing factors determining the performance of parallel simulation executions demands a performance oriented development of logical process simulators. This paper presents an incremental code development process that supports early performance predictions of Time Warp protocols and several of its optimizations. A set of tools, N-MAP, for performance prediction and visualization has been developed, representing a testbed for a detailed sensitivity analysis of the various Time Warp execution parameters. As an example, the effects of various performance factors like the event structure underlying the simulation task, the average LVT progression per simulation step, the commitment rate, state saving overhead, etc. are demonstrated. We show how the scenario management features provided by the N-MAP tool can be efficiently utilized to predict performance sensitivities. For the particular example, the Time Warp protocol, though highly involved, N-MAP was able to predict the per formance sensitivity that was measured from the full implementation executing on the Meiko CS-2.

REFERENCES

• 1: Ian F. Akyildiz , Liang Chen , Samir Ranjan Das , Richard M. Fujimoto , Richard F. Serfozo, The effect of memory capacity on Time Warp performance, Journal of Parallel and Distributed Computing, v.18 n.4, p.411-422, Aug. 1993 [doi>10.1006/jpdc.1993.1074]
• 2: Giovanni Chiola , Alois Ferscha, Distributed Simulation of Timed Petri Nets: Exploiting the Net Structure to Obtain Efficiency, Proceedings of the 14th International Conference on Application and Theory of Petri Nets, p.146-165, June 21-25, 1993
• 3: Samir Das , Richard Fujimoto , Kiran Panesar , Don Allison , Maria Hybinette, GTW: a time warp system for shared memory multiprocessors, Proceedings of the 26th conference on Winter simulation, p.1332-1339, December 11-14, 1994, Orlando, Florida, United States
• 4: Samir R. Das , Richard M. Fujimoto, An adaptive memory management protocol for Time Warp parallel simulation, Proceedings of the 1994 ACM SIGMETRICS conference on Measurement and modeling of computer systems, p.201-210, May 16-20, 1994, Nashville, Tennessee, United States
• 5: Ferscha, A. 1996. Parallel and distributed simulation of discrete event systems. In Parallel and Distributed Computing Handbook, ed. A. Y. Zomaya, 1003-1041. McGraw-Hill.
• 6: Ferscha, A. and G. Chiola. 1995. Adaptive distributed simulation of Petri net models. In Proceedings of 1995 Summer Computer Simulation Conference (SCSC '95).
• 7: Fujimoto, R. M. 1990. Performance of Time Warp under sythetic workloads. In Proceedings of the SCS Multiconference on Distributed Simulation, ed. D. Nicol, 23-28.
• 8: Fujimoto, R. M. 1993. Parallel discrete event simulation: Will the field survive? ORSA Journal of Computing, 5(3):218-230.
• 9: A. Gupta , I. F. Akyildiz , R. M. Fujimoto, Performance Analysis of Time Warp with Multiple Homogeneous Processors, IEEE Transactions on Software Engineering, v.17 n.10, p.1013-1027, October 1991 [doi>10.1109/32.99190]
• 10: Lin, Y-B. 1993. Will parallel simulation research survive? ORSA Journal of Computing, 5(3):236-238.
• 11: Lin, Y-B. and Lazowska, E. 1990. Determining the global virtual time in a distributed simulation, in 1990 International Conference on Parallel Processing, (III)201-209.
• 12: William Pugh, Skip lists: a probabilistic alternative to balanced trees, Communications of the ACM, v.33 n.6, p.668-676, June 1990 [doi>10.1145/78973.78977]
• 13: Robert Rönngren , Rassul Ayani , Richard M. Fujimoto , Samir R. Das, Efficient implementation of event sets in Time Warp, Proceedings of the seventh workshop on Parallel and distributed simulation, p.101-108, May 16-19, 1993, San Diego, California, United States