Abstract
A time-parallel simulation obtains parallelism by partitioning the time domain of the simulation. An approximate time-parallel simulation algorithm named GG1K is developed for acyclic networks of loss FCFSG/G/1/K queues. The GG1K algorithm requires two phases. In the first phase, a similar system (i.e. aG/G/1/∞ queue) is simulated using the GLM algorithm. Then the resultant trajectory is transformed into an approximateG/G/1/K trajectory in the second phase. The closeness of the approximation is investigated theoretically and experimentally. Our results show that the approximation is highly accurate except whenK is very small (e.g. 5) in certain models. The algorithm exploits unbounded parallelism and can achieve near-linear speedup when the number of arrivals simulated is sufficiently large.
Similar content being viewed by others
References
S.G. Akl,The Design and Analysis of Parallel Algorithms (Prentice-Hall, 1989).
S. Andradóttir and T.J. Ott, Parallel simulation of communication networks through time segmentation, Working Paper.
V.C. Bhavsar and J.R. Isaac, Design and analysis of parallel Monte Carlo algorithms, SIAM J. Sci. Stat. Comput. 8(1987)73–95.
K.M. Chandy and R. Sherman, Space-time and simulation,Proc. 1989 SCS Multiconference on Distributed Simulation (1989), pp. 53–57.
C. Chang and P. Heidelberger, Fast simulation of packet loss rates in a shared buffer communications switch, ICASE Report 93-79, ICASE, Hampton, VA (1993).
R.M. Fujimoto, Parallel discrete event simulation, Commun. ACM 33(1990)31–53.
P. Glynn and P. Heidelberger, Analysis of parallel replicated simulations under a completion time constraint, ACM TOMACS 1(1991)3–23.
A.G. Greenberg, B.D. Lubachevsky and I. Mitrani, Unboundedly parallel simulations via recurrence relations,Proc. Conf. on Measurement and Modeling of Computer Systems (1990) pp. 1–12.
P. Heidelberger, Statistical analysis of parallel simulations,Proc. 1986 Winter Simulation Conf. (1986) pp. 290–295.
P. Heidelberger, Discrete event simulations and parallel processing: Statistical properties, SIAM J. Stat. Comput. 9(1988)1114–1132.
P. Heidelberger and H.S. Stone, Parallel trace-driven cache simulation by time partitioning,Proc. 1990 Winter Simulation Conf. (1990) pp. 734–737.
R.E. Lander and M.J. Fischer, Parallel prefix computation, J. ACM 27(1980)831–838.
J. Lane, ATM knits voice, data on any net, IEEE Spectrum 2(1994)42–45.
Y.B. Lin and E.A. Lazowska, Time-division algorithm for parallel simulation, ACM TOMACS 1(1991)73–83.
Y.B. Lin, Parallel trace-driven simulation for packet loss in finite-buffered voice multiplexers, Parallel Comput. 19(1993)219–228.
D. Mitra and I. Mitrani, Control and coordination policies for systems with buffers, ACM SIGMETRICS Perf. Eval. Rev. 17(1989)156–164.
I. Nikolaidis and R. Fujimoto, Parallel simulation of high-speed network multiplexers,IEEE Conf. on Decision and Control, DC-3(1993) pp. 2224–2229.
H.S. Stone,High Performance Computer Architecture (Addison-Wesley, 1987).
W. Verbiest, L. Pinnoo and B. Voeten, The impact of the ATM concept on video coders, IEEE J. Sel. Areas Commun. SAC-6(1988)1623–1631.
J.J. Wang and M. Abrams, Determining initial states for time-parallel simulations,Proc. 1993 Workshop on Parallel and Distributed Simulation (1993) pp. 19–26.
J.J. Wang and M. Abrams, Massively time-parallel, approximate simulation of loss queueing systems, Technical Report TR 94-09, Computer Science Department, VA Tech., VA (1994).
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Wang, J.J., Abrams, M. Massively time-parallel, approximate simulation of loss queueing systems. Ann Oper Res 53, 553–575 (1994). https://doi.org/10.1007/BF02136843
Issue Date:
DOI: https://doi.org/10.1007/BF02136843