Using state estimating to enhance real time and fault tolerance in distributed simulation

Ma Min,ShiYao Jin,Shengning Wu

2006-01-01

Abstract:In simulation field, researchers pay more attention about characteristic of real time and fault tolerance. For WAN and distributed environment, the network transport delay and network partition problems are difficult to be handled, more over HLA, as a current popular framework for distributed simulation does not support real time and fault tolerance. According to analysis of simulation models, we divide system models into continuous state model and discrete state model two types. Hermite interpolating and Markov process theory are proposed to handle network delay and partition problems by substituting real computing data with estimated data, so simulation system has the ability of passing through the transient failure or data transport delay. Consequently real time and fault tolerance characteristics of whole simulation system are improved. In order to guarantee the accuracy of simulation result, we use optimistic recovery to restore system and substitute estimated data with delayed real compute data when the data estimating error exceeds the error limit. At last experimental results show that proposed method is effective with condition of continuous system model and small step (look-head).