Stability of event synchronisation in distributed discrete event simulation

This paper is concerned with the behaviour of message queues in distributed discrete event simulators. We view a logical process in a distributed simulation as comprising a message sequencer with associated message queues, followed by an event processor. We show that, with standard stochastic assumptions for message arrival and time-stamp processes, the message queues are unstable for conservative sequencing, and for conservative sequencing with maximum lookahead and hence for optimistic resequencing, and for any resequencing algorithm that does not employ interprocessor “flow control”. These results point towards certain fundamental limits on the performance of distributed simulation of open queueing networks.

[1]  D.M. Nicol,et al.  High performance parallelized discrete event simulation of stochastic queueing networks , 1988, 1988 Winter Simulation Conference Proceedings.

[2]  K. Mani Chandy,et al.  Asynchronous distributed simulation via a sequence of parallel computations , 1981, CACM.

[3]  Marc A. Kaplan,et al.  A sufficient condition for nonergodicity of a Markov chain , 1979 .

[4]  Jayadev Misra,et al.  Distributed discrete-event simulation , 1986, CSUR.

[5]  Michael Kaplan,et al.  A sufficient condition of nonergodicity of a Markov chain (Corresp.) , 1979, IEEE Trans. Inf. Theory.

[6]  Edward D. Lazowska,et al.  Parallel simulation of queueing networks: limitations and potentials , 1989, SIGMETRICS '89.

[7]  Jon B. Weissman,et al.  MTW: an empirical performance study , 1991, 1991 Winter Simulation Conference Proceedings..

[8]  Isi Mitrani,et al.  Analysis and Optimum Performance of Two Message-Passing Parallel Processors Synchronized by Rollback , 1984, Perform. Evaluation.

[9]  Wojciech Szpankowski,et al.  Towards computable stability criteria for some multidimensional stochastic processes , 1989 .

[10]  J. Harrison.  Assembly-like queues , 1973, Journal of Applied Probability.

[11]  Leonard Kleinrock,et al.  Queueing Systems: Volume I-Theory , 1975 .

[12]  David R. Jefferson,et al.  Virtual time , 1985, ICPP.

[13]  B. Eddy Patuwo,et al.  The effect of synchronization requirements on the performance of distributed simulations , 1993, PADS '93.

[14]  Leonard Kleinrock,et al.  On Distributed Systems Performance , 1990, Comput. Networks ISDN Syst..

[15]  R. M. Fujimoto,et al.  Parallel discrete event simulation , 1989, WSC '89.

[16]  David M. Nicol,et al.  The cost of conservative synchronization in parallel discrete event simulations , 1993, JACM.

[17]  Boris D. Lubachevsky,et al.  Efficient distributed event-driven simulations of multiple-loop networks , 1988, CACM.

[18]  Leonard Kleinrock,et al.  Theory, Volume 1, Queueing Systems , 1975 .

[19]  Leonard Kleinrock,et al.  Bounds and approximations for self-initiating distributed simulation without lookahead , 1991, TOMC.

[20]  Erhan Çinlar,et al.  Introduction to stochastic processes , 1974 .

[21]  Philippe Robert,et al.  Analysis of Update Response Times in a Distributed Data Base Maintained by the Conservative Time Stamps Ordering Algorithm , 1983, Performance.

[22]  W. Szpankowski.  Some sufficient conditions for non-ergodicity of markov chains , 1985, Journal of Applied Probability.

[23]  G. Latouche.  QUEUES WITH PAIRED CUSTOMERS , 1981 .