An optimum checkpointing-based fault tolerant algorithm using mobile agent in distributed systems

Abstract

Traditional message passing based checkpointing and rollback recovery algorithms perform well for tightly coupled systems. In wide area distributed systems these algorithms may suffer from large overhead due to message passing delay and network traffic. Mobile agent (MA) provides a new technology for implementing fault tolerant mechanism in distributed systems. MA has the merits of flexibility, low network traffic, efficiency, less communication latency, as well as highly asynchronous. So to design checkpointing and rollback recovery algorithms for wide area distributed systems, mobile agents are introduced. One of effective factors on performance of mobile agent based algorithms is number of migrations. In this paper we propose an optimum mobile agent based checkpointing protocol which attempts to eliminate useless migrations in different phases of checkpointing algorithm. The proposed algorithm will improve performance and network traffic greatly. The effectiveness of proposed algorithms is evaluated by extensive simulation under a dataset including 7000 different connected graphs. Our results show a significant improvement in number of migrations and average execution time when these solutions are simulated. © 2011 Springer-Verlag.