Performance evaluation of hierarchical distributed simulators

ABSTRACT

A model can be decomposed into components, then mapped onto a network of simulators for execution. With different levels of decompositions, we are interested in finding the optimal model decomposition that can be implemented on a particular structure of distributed simulators. This is done by comparing performance measures such as the minimum response time or the maximum throughout per unit of hardware complexity. We claim that the need for a model capable of performance evaluation is an important concern in designing distributed simulators. Thus a methodology for performance evaluation and simulation modelling of hierarchically specified distributed simulators is required. From the empirical results, we found that simulating a given distributed simulator model is suitable to get optimal performances, that is, each model has an optimal decomposition level in terms of performance. The influential performance measures are the average of simulation run times and the product of the average run time and the hardware complexity of the given simulator model.

REFERENCES

• 1: Balk, D.K. and Zeigler, B,P., "Performance ~odelling and Simulation of Hierarchical Distributed Simulators", Technical R.~ort CSC- ~, Department of Computer Science, Wayne State University, Detroit, Michigan, May 1985.
• 2: David A. Bennett , Christopher A. Landauer, Simulation of a distributed system for performance modelling, ACM SIGMETRICS Performance Evaluation Review, v.8 n.3, p.49-56, Fall 1979
• 3: Bryant, R.E., "Simulation on A Distributed System", IEEE Proceedings of Distributed Computing Systems, 1979.
• 4: Arturo I. Concepcion, Distributed simulation on multiprocessors: specification, design, and architecture (discrete event, computer), 1985
• 5: Concepcion, A.I., Baik, D.K. and Zeigler, B.P., "Distributed Simulation of Hierarchical ~odels", Proc. of the Worksho~ on Parallel Processinq us~ the HEP, Norman, Oklahoma, March 1985.
• 6: David L. Davidson , Paul F. Reynolds, Jr., Performance analysis of a distributed simulation algorithm based on active logical processes, Proceedings of the 15th conference on Winter simulation, p.267-270, December 12-14, 1983, Arlington, Virginia, United States
• 7: Donald F. DuBois, A Hierarchical Modeling System for computer networks, ACM SIGMETRICS Performance Evaluation Review, v.11 n.1, p.147-155, Spring 1981
• 8: Franklin, M.A., Warm, D.F. and Wong, K.F., "Parallel Machines and Algorithms for Discrete-Event Simulation", Proc. of Parallel Processing Conf., IEEE, 1984.
• 9: Livny, M., "A Study of Parallelism in Distributed Simulation", Proc. of the Conf. on Distributed Simulation, San Diego, Jan. 1985.
• 10: David M. Nicol , Paul F. Reynolds, Jr., Problem oriented protocol design, Proceedings of the 16th conference on Winter simulation, p.470-474, December 28-30, 1984, Dallas, TX
• 11: Nicoi, D,M. and Reynolds, P.F., "A Statistical Approach to Dynamic Partioning", Proc._..._~. of the Conf. on Distributed SimuIatiop, San Diego, Jan. 1985.
• 12: Rozenblit,J.W.,'~Experimental Frames for Distributed Simulation Architectures", Proc. of Distributed Simulation, San Diego, Jan. 1985.
• 13: Russell, E.C., BuildinQ Simulation Models with SIMSCRIPT llLk_~,, CACJ, L.A., 1983.
• 14: Bernard P. Zeigler, Theory of Modelling and Simulation, Krieger Publishing Co., Inc., Melbourne, FL, 1984
• 15: Bernard P. Zeigler, Multifacetted modelling and discrete event simulation, Academic Press Professional, Inc., San Diego, CA, 1984