Description

Tasks of a resource management system are complicated in a heterogeneous, dynamic environment like Grid. A resource management system may have to deal with the resources on which it has no control and which are under the control of multiple administrative domains. Moreover, the system loads and status of the resources change frequently. So the information about the resources is often limited or dated. This work proposes an integrated framework for performance-based resource management in computational Grid environment. Objective of this work is to design and implement this framework for pursuing two definite goals - maintaining the quality of services required for a batch of jobs, while keeping the resource utilization cost as minimum as possible. The work demonstrates the effectiveness of the multi-agent system for managing execution of multiple concurrent jobs by developing a tool, called PRAGMA (Performance based Resource Brokering and Adaptive execution in Grid for Multiple concurrent Applications). This work also evaluates the efficiency of this tool by performing different sets of experiments using different batches of jobs on a local Grid test bed.