Abstract
Today, the service provider has to serve many users and the increase of the requests or tasks from the users to the cloud providers has become one of the scalable techniques to proposed the services. Many scheduling algorithms have been proposed to schedule the tasks in cloud computing environment such as (SJF) and (FCFS) algorithms. This paper aims to improve the shortest job first scheduling algorithm in the cloud computing. In tasks scheduling (TS), the most important parameters are makespan and response time. Therefore, we have proposed a Modified Shortest Job First algorithm (MSJF) to minimize the completion time of the last task (Makespan) and minimize the average response time with maximizing the resources utilization. MSJF has two functions, one is calculating the average of tasks length and the other one is Load-Balancing between the virtual machines. One of the important advantages of MSJF is sending the longest tasks to the fastest Machine. We compared the result of our proposed algorithm MSJF with SJF and FCFS. The performance of MSJF is better compared to SJF and FCFS.