Extended Heartbeat Mechanism for Fault Detection Service Methodology

Abstract

Fault detection methodology is a crucial part in providing a scalable, dependable and high availability of grid computing environment. The most popular technique that used in detecting fault is heartbeat mechanism where it monitors the grid resources in a very short interval. However, this technique has its weakness as it requires a period of times before the node is realized to be faulty and therefore delaying the recovery actions to be taken. This is due to unindexed status for each transaction and need to wait for a certain time interval before realizing the nodes has failed. In this paper, fault detection mechanism and service using extended heartbeat mechanism is proposed. This technique introduced the use of index server for indexing the transaction and utilizing pinging service for pushing mechanism. The model outperformed the existing techniques by reducing the time taken to detect fault in approximately 30%. Also, the mechanism provides a basis for customizable recovery actions to be deployed.