D. B. Wagner
E. D. Lazowska
ABSTRACT
This paper concerns the parallel simulation of queueing network models (QNMs) using the conservative (Chandy-Misra) paradigm. Most empirical studies of conservative parallel simulation have used QNMs as benchmarks. For the most part, these studies concluded that the conservative paradigm is unsuitable for speeding up the simulation of QNMs, or that it is only suitable for simulating a very limited subclass of these models (e.g., those containing only FCFS servers). In this paper we argue that these are unnecessarily pessimistic conclusions. On the one hand, we show that the structure of some QNMs inherently limits the attainable simulation speedup. On the other hand, we show that QNMs without such limitations can be efficiently simulated using some recently introduced implementation techniques.
We present an analytic method for determining an upper bound on speedup, and use this method to identify QNM structures that will exhibit poor simulation performance. We then survey a number of promising implementation techniques, some of which are quite general in nature and others of which apply specifically to QNMs. We show how to extend the latter to a larger class of service disciplines than had been considered previously.
- 1.A.O. Allen. Probability, Statistics, and Queueing Theory with Computer Science Applications. Academic Press, New York, NY, 1978. Google Scholar
Digital Library
- 2.R.E. Bryant. Simulation of Packet Communication Architecture Computer Systems. Technical Report MIT, LCS,TR-188, Massachussetts Insitute of Technology, Cambridge, Mass., 1977. Google ScholarDigital Library
- 3.K.M. Chandy and J. Misra. Asynchronous Distributed Simulation via a Sequence of Parallel Computations. Communications o} the A CM, 24(11):198-206, November 1981. Google ScholarDigital Library
- 4.R.M. Fujimoto. Lookahead in Parallel Discrete Event Simulation. In Proc. International Conference on Parallel Processing, St. Charles, IL, August 1988.Google Scholar
- 5.R.M. Fujimoto. Performance Measurements of Distributed Simulation Strategies. In Distributed Simulation 1988, pages 14-20. Society for Computer Simulation International, San Diego, CA, February 1988.Google Scholar
- 6.D. Jefferson, B. Beckman, F. Wieland, L. Blume, M. DiLoreto, P. Hontalas, P. Laroche, K. Sturdevant, J. Tupman, V. Warren, J. Wedel, H. Younger, and S. Bellenot. Time Warp Operating System. In Proc. Eleventh A CM Symposium on Operating Systems Principles, pages 77-93, Austin, TX, November 1987. ACM. Google ScholarDigital Library
- 7.D.R. Jefferson. Virtual Time. A CM Transactions on Programming Languages and Systems, 7(3):404-425, July 1985. Google ScholarDigital Library
- 8.E.D. Lazowska, J. Zahorjan, G.S. Graham, and K.C. Sevcik. Quantitative System Performance: Computer System Analysis Using Queueing Network Models. Prentice-Hall, Englewood Cliffs, N:I, 1984. Google ScholarDigital Library
- 9.Y-B. Lin, August 1988. Personal communication.Google Scholar
- 10.Y-B. Lin, :I-L. Baer, and E.D. Lazowska. Tailoring a Parallel Trace-Driven Simulation Technique to Specific Multiprocessor Cache Coherence Protocols. In Distributed Simulation 1989. Society for Computer Simulation International, San Diego, CA, March 1989.Google Scholar
- 11.J. Mists. Distributed Discrete-Event Simulation. A CM Computing Surveys, 18(1):39-60, March 1986. Google ScholarDigital Library
- 12.R.R. Muntz and J.W. Wong. Asymptotic Properties of Closed Queueing Network Models. In Proc. 8th Princeton Conference on Information Sciences and Systems, 1974.Google Scholar
- 13.D.M. Nicol. Parallel Discrete-Event Simulation of FCFS Stochastic Queueing Networks. In Parallel Programming: Experience with Applications, Languages, and Systems, pages 124-137. ACM SIGPLAN, July 1988. Google ScholarDigital Library
- 14.:I.K. Peacock, J.W. Wong, and E.G. Manning. Distributed Simulation Using a Network of Processors. Computer Networks, 3(1):44-56, March 1979.Google Scholar
- 15.D.A. Reed and A.D. Malony. Parallel Discrete Event Simulation: the Chandy-Misra Approach. In Distributed Simulation 1988, pages 8-13. Society for Computer Simulation International, San Diego, CA, February 1988.Google Scholar
- 16.D.A. Reed, A.D. Malony, and B.D. McCredie. Parallel Discrete Event Simulation Using Shared Memory. IEEE Transactions on Software Engineering, 14(4):541-553, April 1988. Google ScholarDigital Library
- 17.P.F. Reynolds. A Shared Resource Algorithm for Distributed Simulation. In Proc. 9th International Symposium on Computer Architecture, pages 259-266, Austin, TX, 1982. IEEE. Google ScholarDigital Library
- 18.C.H. Sauer, E.A. MacNair, and S. Salza. A Language for Extended Queueing Networks. IBM Journal of Research and Development, 24(6):747-755, November 1980.Google ScholarDigital Library
- 19.H.M. Taylor and S. Karlin. An Introduction to Stochastic Modeling. Academic Press, Orlando, FL, 1984.Google Scholar
- 20.D.B. Wagner. Conservative Parallel Simulation: Principles and Practice. PhD thesis, University of Washington, Seattle, WA, 1989. (In preparation.). Google ScholarDigital Library
- 21.D.B. Wagner and E.D. Lazowsl~. Parallel Simulation of Queueing Networks: Limitations and Potentials. Technical Report 88-09-05, Department of Computer Science, University of Washington, September 1988.Google Scholar
- 22.D.B. Wagner, E.D. Lazowska, and B.N. Bershad. Techniques for Efficient Shared-Memory Parallel Simulation. In Distributed Simulation 1989. Society for Computer Simulation International, San Diego, CA, March 1989.Google Scholar
Index Terms
- Parallel simulation of queueing networks: limitations and potentials
Recommendations
Parallel simulation of queueing networks: limitations and potentials
This paper concerns the parallel simulation of queueing network models (QNMs) using the conservative (Chandy-Misra) paradigm. Most empirical studies of conservative parallel simulation have used QNMs as benchmarks. For the most part, these studies ...
Conservative Parallel Simulation of Priority Class Queuing Networks
A conservative synchronization protocol for the parallel simulation of queuing networkshaving C job priority classes, where a job's class is fixed, is described. This problem has long vexed designers of conservative synchronization protocols because of ...
Monotone queuing networks and time parallel simulation
ASMTA'11: Proceedings of the 18th international conference on Analytical and stochastic modeling techniques and applicationsWe show how we can make more efficient the time parallel simulation of some queueing networks. We had previously introduced some improvements of two classical time parallel simulation approaches: the precomputation of states at specific time instants ...
Comments