Improved version of round robin scheduling algorithm based on analytic model

Abstract

Scheduling is one of the most important issues in the operating system, such as the tasks must be affected to the appropriate virtual machines, considering different factors at the same time to ensure better use of resources. A lot of research has been carried out to propose more efficient task scheduling algorithms. Round robin is one of the most powerful algorithms in this field but its main challenge is the choice of time quantum. The effectiveness of Round Robin depends on the choice of this parameter. In this paper, we aimed to overcome these challenge by proposing an improved Round Robin scheduling algorithm using a variable time quantum based on an analytic model. Our analytical model takes into consideration different parameters to determine the order of tasks execution. The use of burst time as parameters in our model ensures a more suitable time quantum. This algorithm can be applied in any operating system and therefore in the cloud computing environment. In order to assess the performance and effectiveness of the proposed approach, five different scenarios have been implemented; the comparisons made with existing works have shown that the proposed approach improves the average waiting time and the average turnaround time, which ensures better scheduling of tasks and better use of resources.