Design and Development of Agent Based Architecture for Effective Resource Utilization in a Grid Environment

Software agents are the autonomous, problem-solving computational entities capable of effective operation in dynamic and open environments. Intelligent agent is a type of software agent, autonomous in nature which observes and acts upon environment and performs some task at each host. Agent can also coordinate, reason, solve a problem, clone and merge with other agents. A Grid is a set of resources (such as CPU, Memory, Disk, Applications, and Database) distributed over wide area networks and supports large scale distributed applications. Resources in grid are geographically distributed, linked through internet, to create virtual super computer for vast amount of computing capacity to solve complex problems. Genetic Algorithm works with key parameters such as fitness function, crossover probability and mutation probability and optimizes task scheduling. The paper proposes a software agent based architecture to utilize the resources effectively in Grid environment. The architecture is compared and analyzed the resource utilization with three algorithms namely Shortest job first, Arbitrarily Scheduling Algorithm and Genetic algorithm. The efficiency of resources utilization is analyzed and suitable algorithm is suggested.