Parallel Simulation of Statistical Multiplexers Parallel Simulation of Statistical Multiplexers

Abstract

The simulation of high{speed telecommunication systems such as ATM (Asyn-chronous Transfer Mode) networks has generally required excessively long run times. This paper reviews alternative approaches using parallelism to speed up simulations of discrete event systems, and telecommunication networks in particular. Subsequently, a new simulation method is introduced for the fast parallel simulation of a common network element, namely, a work{conserving nite capacity statistical multiplexer of bursty ON/OFF sources arriving on input links of equal peak rate. The primary performance measure of interest is the cell loss ratio, due to buuer overrows. The proposed method is based on two principal techniques: (1) the derivation of low{level (cell level) statistics from a higher level (burst level) simulation and (2) parallel execution of the burst level simulation program. For the latter, a time{division parallel simulation method is used where simulations operating at diierent intervals of simulated time are executed concurrently on diierent processors. Both techniques contribute to the overall speedup. Furthermore, these techniques support simulations that are driven by traces of actual network traac (trace{driven simulation), in addition to standard models for source traac. An analysis of this technique is described, indicating that it ooers excellent potential for delivering good performance. Measurements of an implementation running on a 32 processor KSR{2 multiprocessor demonstrate that, for certain model parameter settings, the simulator is able to simulate up to 10 billion cell arrivals per second of wallclock time.

1 Figure or Table

Cite this paper

@inproceedings{Nikolaidis1994ParallelSO, title={Parallel Simulation of Statistical Multiplexers Parallel Simulation of Statistical Multiplexers}, author={Ioanis Nikolaidis and Richard M. Fujimoto and C. Anthony Cooper}, year={1994} }