A resource allocation algorithm for heterogeneous jobs in grid environment

Grid is a large-scale, heterogeneous, dynamic collection of independent systems which are geographically distributed and interconnected with high speed networks to meet ever increasing demand of high performance computing. Allocating distributed computational resources for end-user applications is one of the most complex and challenging tasks in Grid computing. Brokers are required for submitting the jobs of the end users to accomplish their demands of resources. To allocate resources in Grid, the Grid requires the description of a framework that can permit brokers to communicate to accomplish an efficient utilization of the resources themselves. The user applications must simultaneously assure both constraints of job execution and policies that govern system usage. This paper discusses an algorithm for resource allocation for heterogeneous jobs, subject to a set of constraints. The strategy used tries to reduce the cost of the Grid users' while allocating resources to Grid users. The resource allocation strategy discussed can control the task assignment to Grid resources with the objective to minimize the execution cost of the Grid users. The technique is analytically examined to obtain results under various environment conditions.