Adaptive and distributed fault monitoring of grid by mobile agents

Abstract

Efficient fault monitoring in the grid environment is a must that enables the stable provision of distributed computing resources without the interruption even in the case of grid node faults and dynamic reconfiguration. This paper proposes a mobile agent based approach for the fault detection of grid systems. In this approach, mobile agents embedding an extended Adaptive and Distributed System level Diagnosis (ADSD) algorithm are deployed to the nodes on a grid, and detect and diagnose the node failures in an adaptive and distributed way. We also present the experiment results. © Springer-Verlag 2004.