Enable efficient stage construction for replication based fault-tolerant execution of mobile agent

Abstract

Reliability as well as fault-tolerance is a fundamental issue for the development of robust mobile agent system. A number of research works is done in these areas. Some researchers adopt the use of spatial replication based approach since it can reduce blocking possibility of mobile agent execution. Unfortunately, this fault-tolerance scheme is not cost effective because it incurs great costs of time and communication, which leads to increased system load. In this paper, we present a dynamic stage construction protocol for mobile agent execution to maintain high performance without being forced to make tradeoffs in reliability. The protocol schedules task execution dynamically according to the execution ability of visited nodes, hence avoid the unnecessary revisit to them at latter stage and cut down total execution time. On the other hand, it has specific node selection strategy to guarantee reliability, decrease communication overhead between two consecutive stages and improve efficiency