Extending self-healing in grid environment by pulse monitoring

Abstract

The increasing complexity of Grid services and systems demands larger human effort for system configuration and performance management. The physical management of these systems is not only tedious but also error-prone. Inherently, effective monitoring is essential for Grid management, which can further facilitate self-management of the Grid as advocated by Autonomic computing i.e. shifting the burden of managing systems from people to technologies. This paper proposes a pulse monitoring based Grid Monitoring System that passes the monitoring data to Fault Discovery and Healing System, which has been designed for facilitating self-healing in Grid. The Grid monitoring system demonstrates that through efficient monitoring, self-healing can be achieved in a Grid environment and Grid can work autonomously, thus minimizing the physical management and service-downtime.