Superfast parallel discrete event simulations

Nonconventional parallel simulations methods are presented, wherein speed-ups are not limited by the number of simulated components. The methods capitalize on Chandy and Sherman's space-time relaxation paradigm, and incorporate fast algorithms for solving recurrences. Special attention is paid to implementing these algorithms on currently available massively parallel SIMD computers. As examples, “superfast” simulations for open and closed queuing networks and for the slotted ALOHA communication protocol are discussed. Several of the simulations are implemented and show impressive computational speeds. This paper summarizes previous results of the authors and presents some new experiments and simulations.

[1]  Albert G. Greenberg,et al.  ICASE MASSIVELY PARALLEL ALGORITHMS FOR TRACE-DRIVEN CACHE SIMULATIONS , 1991 .

[2]  M. R. Leadbetter,et al.  Extremes and Related Properties of Random Sequences and Processes: Springer Series in Statistics , 1983 .

[3]  Albert G. Greenberg,et al.  Simple, Efficient Asynchronous Parallel Prefix Algorithms , 1987, ICPP.

[4]  P. Glynn,et al.  Departures from Many Queues in Series , 1991 .

[5]  Albert G. Greenberg,et al.  Efficient Parallel Algorithms for Linear Recurrence Computation , 1982, Inf. Process. Lett..

[6]  Kenneth E. Batcher,et al.  Sorting networks and their applications , 1968, AFIPS Spring Joint Computing Conference.

[7]  Albert G. Greenberg,et al.  Synchronous relaxation for parallel simulations with applications to circuit-switched networks , 1993, TOMC.

[8]  Albert G. Greenberg,et al.  Unboundedly parallel simulations via recurrence relations for network and reliability problems , 1990, 1990 Winter Simulation Conference Proceedings.

[9]  Lawrence G. Roberts,et al.  Dynamic allocation of satellite capacity through packet reservation , 1973, AFIPS National Computer Conference.

[10]  Albert G. Greenberg,et al.  Algorithms for unboundedly parallel simulations , 1991, TOCS.

[11]  Philip Heidelberger,et al.  Parallel trace-driven cache simulation by time partitioning , 1990, 1990 Winter Simulation Conference Proceedings.

[12]  Albert G. Greenberg,et al.  A Sweep Algorithm for Massively Parallel Simulation of Circuit-Switched Networks , 1993, J. Parallel Distributed Comput..

[13]  H. T. Kung,et al.  The Complexity of Parallel Evaluation of Linear Recurrences , 1977, JACM.

[14]  Albert G. Greenberg,et al.  Unboundedly parallel simulations via recurrence relations , 1990, SIGMETRICS '90.

[15]  K. G. Ramakrishnan,et al.  Panacea: An Integrated Set of Tools for Performance Analysis , 1989 .

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

[17]  Debasis Mitra,et al.  Control and coordination policies for systems with buffers , 1989, SIGMETRICS '89.

[18]  Jain J. Wang,et al.  Massively time-parallel, approximate simulation of loss queueing systems , 1994, Ann. Oper. Res..

[19]  Sanjeev Saxena,et al.  On Parallel Prefix Computation , 1994, Parallel Process. Lett..

[20]  Guy E. Blelloch,et al.  Scans as Primitive Parallel Operations , 1989, ICPP.

[21]  Albert G. Greenberg,et al.  Experience in massively parallel discrete event simulation , 1993, SPAA '93.

[22]  D. Iglehart,et al.  Multiple channel queues in heavy traffic. I , 1970, Advances in Applied Probability.

[23]  Clyde P. Kruskal,et al.  Searching, Merging, and Sorting in Parallel Computation , 1983, IEEE Transactions on Computers.