Efficient resource clustering and scheduling algorithms for solving the job scheduling problems in the cloud environment

Abstract

Cloud computing storms the today s world by making a revolution in technology and satisfies the requirements of the current business. It is very important to use good scheduling policies for the effective usage of cloud resources that leads to a successful business environment. But, till now it s a challenge for the cloud environments to chedule their jobs among the resources in an efficient way, because these problems come under the theory of Non Deterministic polynomial hard problem. Resource clustering plays a vital role in job scheduling problem in the cloud environments. It is essential to develop good resource clustering policies to produce efficient job scheduling algorithms in the cloud environment whereas, many existing systems failed to produce a successful policy for the formation of resource clusters, job scheduling and resulting in optimal solution. Metaheuristic, evolutionary and hyperheuristic approaches are the most used methodologies in the distributed computing such as the grid computing and cloud computing for developing the job scheduling algorithms. Some of the major parameters such as arrival time, response time and device utilization should be considered while developing the scheduling and resource clustering algorithms. In order to identify the efficiency of the developed algorithms, the above parameters should be evaluated using a performance evaluation model. A good performance evaluation model identifies the accurate arrival rate, response time of jobs and exact percentage of device utilization. But most of the existing performance evaluation models failed to produce the optimal solution. A good scheduling approach should provide optimal solution for all the four types of tasks and resources which had not happened before. The above points specify that, the proposed approach should address and overcome all these disadvantages of the existing systems newline