Analysis of mobile agents' fault-tolerant behavior

Abstract

As mobile agents are the medium for implementing various executions, their behaviors are paramount for network performance. Clearly, one of the pivotal tasks ahead, if mobile agents are to have significant impact, is to explore quantitative studies on the behaviors of mobile agents, which can reveal the inherence of the mobile agent approach and ultimately guide future research. This issue, unlike that of system design using mobile agents, has not yet been adequately addressed so far. We propose a fault-tolerant model for mobile agents executing in large distributed networks and analyze the life expectancy of mobile agents in our model. The key idea is the use of stochastic regularities of mobile agents' behavior - all the mobile agents in the network as a whole can be stochastically characterized though a single mobile agent may act randomly. In effect, our analytical results reveal new theoretical insights into the statistical behaviors of mobile agents and provide useful tools for effectively managing mobile agents in large networks.