Scalable clustered time warp and logic simulation

Abstract

An algorithm called Clustered Time Warp (CTW) is introduced for the parallel simulation of discrete event models on a general purpose distributed memory architecture. The scalability of CTW is investigated using three different queueing models and different service-time distributions. The results show that the algorithm acts to limit the explosion of rollbacks exhibited by Time Warp. Further, the memory requirements for CTW are found to be three times smaller than that of Time Warp for one model and half as large for the two other models.