Abstract
This paper surveys efficient techniques for estimating, via simulation, the probabilities of certain rare events in queueing and reliability models. The rare events of interest are long waiting times or buffer overflows in queueing systems, and system failure events in reliability models of highly dependable computing systems. The general approach to speeding up such simulations is to accelerate the occurrence of the rare events by using importance sampling. In importance sampling, the system is simulated using a new set of input probability distributions, and unbiased estimates are recovered by multiplying the simulation output by a likelihood ratio. Our focus is on describing asymptotically optimal importance sampling techniques. Using asymptotically optimal importance sampling, only a fixed number of samples are required to get accurate estimates, no matter how rare the event of interest is. In practice, this means that the required run lengths can be reduced by many orders of magnitude, compared to standard simulation. The queueing systems studied include simple queues (e.g., GI/GI/1) and discrete time queues with multiple autocorrelated arrival processes that arise in the analysis of Asynchronous Transfer Mode communications switches. References for results on Jackson networks and and tree structured networks of ATM switches are given. Both Markovian and non-Markovian reliability models are treated.
Chapter PDF
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Al-Qaq, W.A., M. Devetsikiotis, and K.R. Townsend. 1993. Importance sampling methodologies for simulation of communication systems with adaptive equalizers and time-varying channels. IEEE Journal on Selected Areas in Communications 11: 317–327.
Anantharam, V. 1988. How large delays build up in a GI/G/1 queue. Queueing Systems 5: 345–368.
Anantharam, V., P. Heidelberger, and P. Tsoucas. 1990. Analysis of rare events in continuous time Markov chains via time reversal and fluid approximation. IBM Research Report RC 16280. Yorktown Heights, New York.
Arsham, H., A. Fuerverger, D.L. McLeish, J. Kreimer, and R.Y. Rubinstein. 1989. Sensitivity analysis and the “what if” problem in simulation analysis. Math. Comput. Modelling 12: 193–219.
Asmussen, S. 1982. Conditioned limit theorems relating a random walk to its associate. Advances in Applied Probability 14: 143–170.
Asmussen, S. 1985. Conjugate processes and the simulation of ruin problems. Stochastic Processes and their Applications 20: 213–229.
Asmussen, S. 1987. Applied Probability and Queues. New York, NY: J. Wiley & Sons, Inc.
Asmussen, S. 1989. Risk theory in a Markovian environment. Scand. Actuarial J., 69–100.
Asmussen, S. 1990. Exponential families and regression in the Monte Carlo study of queues and random walks. The Annals of Statistics 18: 1851–1867.
Asmussen, S. 1993. Busy period analysis, rare events and transient behaviour in fluid models. To appear in Journal of Applied Mathematics and Stochastic Analysis.
Asmussen, S., and R.Y. Rubinstein. 1992. The efficiency and heavy traffic properties of the score function method in sensitivity analysis of queueing models. Advances in Applied Probability 24: 172–201.
Asmussen, S., R.Y. Rubinstein, and C.L. Wang. 1992. Efficient regenerative rare events simulation via the likelihood ratio method. Preprint.
Barlow, R.E., and F. Proschan. 1981. Statistical Theory of Reliability and Life Testing. New York, NY: Holt, Reinhart and Winston, Inc.
Breiman, L. 1968. Probability. Reading, MA: Addison-Wesley.
Brown, M. 1990. Error bounds for exponential approximations of geometric convolutions. The Annals of Probability 18: 1388–1402.
Bucklew, J. 1990. Large Deviation Techniques in Decision, Simulation and Estimation. New York, NY: J. Wiley & Sons, Inc.
Bucklew, J.A., P. Ney, and J.S. Sadowsky. 1990. Monte Carlo simulation and large deviations theory for uniformly recurrent Markov chains. J. Appl. Prob. 27: 44–99.
Carrasco, J.A. 1991. Failure distance-based simulation of repairable faulttolerant systems. In Proceedings of the Fifth International Conference on Modeling Techniques and Tools for Computer Performance Evaluation, 337–351.
Carrasco, J.A. 1991. Efficient transient simulation of failure/repair Markovian models. In Proceedings of the Tenth Symposium on Reliable and Distributed Computing, 152–161, IEEE Computer Society Press, Pisa, Italy.
Chang, C.S. 1992. Stability, queue length and delay, Part II: Stochastic queueing networks. IBM Research Report RC 17709, Yorktown Heights, New York. Part of the report is published in Proceedings of the IEEE CDC'92 Conference, 1005–1010, IEEE Computer Society Press, Tucson, Arizona, 1992.
Chang, C.S., P. Heidelberger, S. Juneja, and P. Shahabuddin. 1992. Effective bandwidth and fast simulation of ATM intree networks. IBM Research Report RC 18586, Yorktown Heights, New York. To appear in Proceedings of the Performance '93 Conference.
Chang, C.S., P. Heidelberger, S. Juneja, and P. Shahabuddin. 1993. The application of effective bandwidth to fast simulation of communication networks. IBM Research Report RC 18877, Yorktown Heights, New York.
Chen, H., and A. Mandelbaum. 1991. Discrete flow networks: bottleneck analysis and fluid approximations. Mathematics of Operations Research 16: 408–446.
Çinlar, E. 1975. Introduction to Stochastic Processes. Englewood Cliffs, NJ: Prentice Hall, Inc..
Cogburn, R. 1975. A uniform theory for sums of Markov chain transition probabilities. The Annals of Probability 3: 191–214.
Cottrell, M., J.C, Fort, and G. Malgouyres. 1983. Large deviations and rare events in the study of stochastic algorithms. IEEE Transactions on Automatic Control AC-28: 907–920.
Crane, M.A., and D.L. Iglehart. 1975. Simulating stable stochastic systems III: regenerative processes and discrete event simulation. Operations Research 23: 33–45.
Csiszár, I., T.M. Cover and B.-S. Choi. 1987. Conditional limit theorems under Markov conditioning. IEEE Transactions on Information Theory 33: 788–801.
Devetsikiotis, M., and K.R. Townsend. 1992. On the efficient simulation of large communication networks using importance sampling. In Proceedings of IEEE Globecom '92, IEEE Computer Society Press.
Devetsikiotis, M., and K.R. Townsend. 1992. A dynamic importance sampling methodology for the efficient estimation of rare event probabilities in regenerative simulations of queueing systems. In Proceedings of the IEEE ICC '92 Conference, 1290–1297, IEEE Computer Society Press.
Devetsikiotis, M., and K.R. Townsend. 1993. Statistical optimization of dynamic importance sampling parameters for efficient simulation of communication networks. To appear in IEEE/ACM Transactions on Networking.
Ellis, R. 1984. Large deviations for a general class of random vectors. Annals of Probability 12: 1–12.
Feller, W. 1971. An Introduction to Probability Theory and its Applications. Vol. 2 (Second Edition). New York, NY: J. Wiley & Sons, Inc.
Frater, M.R., and B.D.O. Anderson. 1989. Fast estimation of the statistics of excessive backlogs in tandem networks of queues. Australian Telecommunications Research 23: 49–55.
Frater, M.R., R.R. Bitmead, R.A. Kennedy, and B.D.O. Anderson. 1989. Rare events and reverse time models. In Proceedings of the 28th Conference on Decision and Control, 1180–1183, IEEE Press.
Frater, M.R., T.M. Lenon, and B.D.O. Anderson. 1991. Optimally efficient estimation of the statistics of rare events in queueing networks. IEEE Transactions on Automatic Control 36: 1395–1405.
Frater, M.R., J. Walrand, and B.D.O. Anderson. 1990. Optimally efficient simulation of buffer overflows in queues with deterministic service times via importance sampling. Australian Telecommunications Research 24: 1–8.
Gärtner, J. 1977. On large deviations from invariant measure. Theory Probab. Appl. 22: 24–39.
Geist, R.M. and M.K. Smotherman. 1989. Ultrahigh reliability estimates through simulation. In Proceedings of the Annual Reliability and Maintainability Symposium, 350–355, IEEE Press.
Gertsbakh, I.B. 1984. Asymptotic methods in reliability theory: A review. Advances in Applied Probability 16: 147–175.
Gibbens, R.J., and P.J. Hunt. 1991. Effective bandwidths for the multi-type UAS channel. Queueing Systems 9: 17–28.
Glasserman, P. 1993. Stochastic monotonicity and conditional Monte Carlo for likelihood ratios. Advances in Applied Probability 25: 103–115.
Glynn, P.W. 1989. A GSMP formalism for discrete event systems. Proceedings of the IEEE 77: 14–23.
Glynn, P.W. 1992. Importance sampling for Markov chains: asymptotics for the variance. Technical Report, Department of Operations Research, Stanford University, To appear in Stochastic Models.
Glynn, P.W., and P. Heidelberger. 1992. Experiments with initial transient deletion for parallel, replicated steady-state simulations. Management Science 38: 400–418.
Glynn, P.W., and D.L. Iglehart. 1989. Importance sampling for stochastic simulations. Management Science 35: 1367–1392.
Glynn, P.W., and W. Whitt. 1992. The asymptotic efficiency of simulation estimators. Operations Research 40: 505–520.
Goyal, A., W.C. Carter, E. de Souza e Silva, S.S. Lavenberg, and K.S. Trivedi. 1986. The System Availability Estimator. In Proceedings of the Sixteenth International Symposium on Fault-Tolerant Computing, 84–89, IEEE Computer Society Press.
Goyal, A., and S.S. Lavenberg. 1987. Modeling and analysis of computer system availability. IBM Journal of Research and Development 31, 6: 651–664.
Goyal, A., P. Heidelberger, and P. Shahabuddin. 1987. Measure specific dynamic importance sampling for availability simulations. In 1987 Winter Simulation Conference Proceedings, 351–357, IEEE Press.
Goyal, A., P. Shahabuddin, P. Heidelberger, V.F. Nicola, and P.W. Glynn. 1992. A unified framework for simulating Markovian models of highly reliable systems. IEEE Transactions on Computers C-41: 36–51.
Guérin, R., H. Ahmadi and M. Naghshineh. 1991. Equivalent capacity and its application to bandwidth allocation in high-speed networks. IEEE J. Select. Areas Commun. 9: 968–981.
Halton, J.H. 1970. A retrospective and prospective survey of the Monte Carlo method. SIAM Review 12: 1–60.
Hammersley, J.M., and D.C. Handscomb, 1964. Monte Carlo Methods. London: Methuen and Co., Ltd.
Heidelberger, P. 1988. Discrete event simulations and parallel processing: statistical properties. SIAM Journal on Scientific and Statistical Computing 9: 1114–1132.
Heidelberger, P., Nicola, V.F., and Shahabuddin, P. 1992. Simultaneous and efficient simulation of highly dependable systems with different underlying distributions. In Proceedings of the 1992 Winter Simulation Conference, 458–465, IEEE Press.
Heidelberger, P., P. Shahabuddin, and V.F. Nicola. 1993. Bounded relative error in estimating transient measures of highly dependable non-Markovian systems. IBM Research Report RC 18794, Yorktown Heights, New York.
Heidelberger, P., and P. Tsoucas. 1991. Reverse time simulation of rare events. IBM Technical Disclosures Bulletin 34, No. 3: 163–165.
Iglehart, D.L. 1972. Extreme values in the GI/G/1 queue. Annals of Mathematical Statistics 43: 627–635.
Jensen, A. 1953. Markov chains as an aid in the study of Markov processes. Skand. Aktuarietidskr. 36: 87–91.
Juneja, S. 1993. Efficient Rare Event Simulation of Stochastic Systems. Ph.D. Thesis, Department of Operations Research, Stanford University, California.
Juneja, S., and P. Shahabuddin. 1992. Fast simulation of Markovian reliability/availability models with general repair policies. In Proceedings of the Twenty-Second International Symposium on Fault-Tolerant Computing, 150–159, IEEE Computer Society Press.
Kalos, M.H., and P.A. Whitlock. 1986. Monte Carlo Methods, Volume I: Basics. New York, NY: John Wiley & Sons, Inc.
Keilson, J. 1979. Markov Chain Models — Rarity and Exponentiality, New York, NY: Springer Verlag.
Kelly, F.P. 1979. Reversibility and Stochastic Networks. New York, NY: John Wiley & Sons, Inc.
Kelly, F.P. 1991. Effective bandwidths at multi-class queues. Queueing Systems 9: 5–16.
Kesidis, G., and J. Walrand. 1993. Quick simulation of ATM buffers with on-off multiclass Markov fluid sources. To appear in ACM Transactions on Modeling and Computer Simulation.
Lehtonen, T., and H. Nyrhinen. 1992. Simulating level-crossing probabilities by importance sampling. Advances in Applied Probability 24: 858–874.
Lehtonen, T., and H. Nyrhinen. 1992. On asymptotically efficient simulation of ruin probabilities in a Markovian environment. Scand. Actuarial. J. 60–75.
Lewis, E.E., and F. Bohm. 1984. Monte Carlo simulation of Markov unreliability models. Nuclear Engineering and Design 77: 49–62.
Lewis, P.A.W., and G.S. Shedler. 1979. Simulation of nonhomogeneous Poisson processes by thinning. Naval Research Logistics Quarterly 26, 403–413.
Nakayama, M.K. 1991. Simulation of Highly Reliable Markovian and NonMarkovian Systems. Ph.D. Thesis, Department of Operations Research, Stanford University, California.
Nakayama, M.K. 1991. Asymptotics for likelihood ratio derivative estimators in simulations of highly reliable Markovian systems. IBM Research Report RC 17357, Yorktown Heights, NY.
Nakayama, M.K. 1993. A characterization of the simple failure biasing method for simulations of highly reliable Markovian systems. IBM Research Report RC 18721, Yorktown Heights, New York.
Nakayama, M.K. 1993. General conditions for bounded relative error in simulations of highly reliable Markovian systems. IBM Research Report RC 18993. Yorktown Heights, New York.
Nakayama, M.K., A. Goyal and P.W. Glynn. 1990. Likelihood ratio sensitivity analysis for Markovian models of highly dependable systems. IBM Research Report RC 15400, Yorktown Heights, New York. To appear in Operations Research.
Nicol, D.M., and D.L. Palumbo. 1993. Reliability analysis of complex models using SURE bounds. ICASE NASA Langley Research Center Technical Report 93-14.
Nicola, V.F., P. Heidelberger and P. Shahabuddin. 1992. Uniformization and exponential transformation: techniques for fast simulation of highly dependable non-Markovian systems. In Proceedings of the Twenty-Second International Symposium on Fault-Tolerant Computing, 130–139, IEEE Computer Society Press.
Nicola, V.F., M.K. Nakayama, P. Heidelberger and A. Goyal. 1990. Fast simulation of dependability models with general failure, repair and maintenance processes. In Proceedings of the Twentieth International Symposium on Fault-Tolerant Computing, 491–498, IEEE Computer Society Press.
Nicola, V.F., M.K. Nakayama, P. Heidelberger, and A. Goyal. 1991. Fast simulation of highly dependable systems with general failure and repair processes. IBM Research Report RC 16993. Yorktown Heights, New York. To appear in IEEE Transactions on Computers.
Nicola, V.F., P. Shahabuddin, P. Heidelberger and P.W. Glynn. 1993. Fast simulation of steady-state availability in non-Markovian highly dependable systems. In Proceedings of the Twenty-Third International Symposium on Fault-Tolerant Computing, 38–47, IEEE Computer Society Press.
Nicola, V.F., P. Shahabuddin, and P. Heidelberger. 1993. Techniques for fast simulation of highly dependable systems. IBM Research Report RC 18956, Yorktown Heights, New York.
Parekh, S., and J. Walrand. 1989. A quick simulation method for excessive backlogs in networks of queues. IEEE Transactions on Automatic Control 34: 54–56.
Rubinstein, R.Y. 1991. How to optimize discrete-event systems from a single sample path by the score function method. Annals of Operations Research 27: 175–212.
Sadowsky, J.S. 1991. Large deviations and efficient simulation of excessive backlogs in a GI/G/m queue. IEEE Transactions on Automatic Control 36: 1383–1394.
Sadowsky, J.S. 1993. On the optimality and stability of exponential twisting in Monte Carlo estimation. IEEE Transactions on Information Theory 39: 119–128.
Sadowsky, J.S., and J.A. Bucklew. 1990. On large deviations theory and asymptotically efficient Monte Carlo estimation. IEEE Transactions on Information Theory 36: 579–588.
Shahabuddin, P. 1990. Simulation and Analysis of Highly Reliable Systems. Ph.D. Thesis, Department of Operations Research, Stanford University, California.
Shahabuddin, P. 1991. Importance sampling for the simulation of highly reliable Markovian systems. IBM Research Report RC 16729, Yorktown Heights, New York. To appear in Management Science.
Shahabuddin, P. 1993. Fast transient simulation of Markovian models of highly dependable systems. IBM Research Report RC 18587, Yorktown Heights, New York. To appear in Proceedings of the Performance '93 Conference.
Shahabuddin, P., and M.K. Nakayama. 1993. Estimation of reliability and its derivatives for large time horizons in Markovian systems. IBM Research Report RC 18864, Yorktown Heights, New York. To appear in Proceedings of 1993 Winter Simulation Conference.
Shahabuddin, P., V.F. Nicola, P. Heidelberger, A. Goyal, and P.W. Glynn. 1988. Variance Reduction in Mean Time to Failure Simulations. In 1988 Winter Simulation Conference Proceedings, 491–499, IEEE Press.
Shanthikumar, J.G. 1986. Uniformization and hybrid simulation/analytic models of renewal processes. Operations Research 34: 573–580.
Shwartz, A., and A. Weiss. 1993. Induced rare events: analysis via time reversal and large deviations. To appear in Advances in Applied Probability.
Siegmund, D. 1976. Importance sampling in the Monte Carlo study of sequential tests. The Annals of Statistics 4: 673–684.
Smith, W.L. 1955. Regenerative stochastic processes. Proc. Roy. Soc. Ser. A. 232: 6–31.
Tsoucas, P. 1989. Rare events in series of queues. IBM Research Report RC 15530, Yorktown Heights, New York.
Weiss, A. 1986. A new technique for analyzing large traffic systems. Advances in Applied Probability 18: 506–532.
Whitt, W. 1980. Continuity of generalized semi-Markov processes. Mathematics of Operations Research 5: 494–501.
Whitt, W. 1993. Tail probability with statistical multiplexing and effective bandwidths in multi-class queues. To appear in Telecommunication Systems.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1993 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Heidelberger, P. (1993). Fast simulation of rare events in queueing and reliability models. In: Donatiello, L., Nelson, R. (eds) Performance Evaluation of Computer and Communication Systems. Performance SIGMETRICS 1993 1993. Lecture Notes in Computer Science, vol 729. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0013853
Download citation
DOI: https://doi.org/10.1007/BFb0013853
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-57297-8
Online ISBN: 978-3-540-48044-0
eBook Packages: Springer Book Archive