Job Scheduling In Cloud Computing Based On Adaptive Job Size Based Queuing Process

Abstract

Cloud computing is one of the most emerging distributed technology. It enables remotes accessing, storing, retrieving and processing of data from anywhere anytime. The main feature of this cloud technology is resource sharing at highly reliable and securable mode. The vast usage of this technology with several features had a greater impact, especially in the need for efficient methodology in resource management. Generally, there are two main components in cloud computing such as task scheduling and resource allocation. In the cloud, the major process is the computing of remote data centers. Resource scheduling and task execution are one of the major problems that still exist. It needs effective resource utilization that is assigning of available resources according to the request equally. The major issue still existing in cloud computing is time and computation cost. In this paper, we propose an Adaptive Work Size Based Queuing Process (AWSQP), this enables quick data access to the virtual machine (VM’s). The main principle is to achieve quick processing, task priority and minimum sized task are taken for the queue. At the same time, the task is executed within the deadline but without increasing the cost. Our proposed mechanism AWSQP works effectively based on the data size and chooses the best cost-effective path. The data access completion time is evaluated by the request/response time along with the mean and variance of the network service time. Then AWSQP applies the best path for the task with high priority and continues the same for the entire queue. Based on the proposed task scheduling mechanism, an experiment is conducted between CETS, ACTS, and AWSQP on the aspects of computation cost, communication cost, execution time, CPU utilization and bandwidth. The obtained results prove that achieved performance by AWSQP is far better than the existing approaches.