A probabilistic performance model for conservative simulation protocol

Abstract

Extensive research has been conducted on the performance of simulation systems. In this paper, a performance analysis of the conservative parallel simulation protocol is presented. We analyzed several performance factors for the conservative simulation of a model with an arbitrary number of logical processes. The analysis probabilistically identifies the critical path in a conservative model, and also estimates the number of safe messages processed by an logical process during its busy time. This can be used to estimate the busy time of a logical process and hence the duration for which it is blocked. The analytical results are verified using simulation of a real world wireless network simulation model.