A Novel Amended Dynamic Round Robin Scheduling Algorithm for Timeshared Systems

A Novel Amended Dynamic Round Robin Scheduling Algorithm for Timeshared Systems

Uferah Shafi1, Munam Shah1, Abdul Wahid1, Kamran Abbasi2, Qaisar Javaid3, Muhammad Asghar4, and Muhammad Haider1

1Department of Computer Science, COMSATS University Islamabad, Pakistan

2Department of Distance Continuing and Computer Education, University of Sindh, Pakistan

3Department of Computer Science, International Islamic University, Pakistan

4Department of Computer Science, Bahuddin Zikriya University, Pakistan

 

Abstract: Central Processing Unit (CPU) is the most significant resource and its scheduling is one of the main functions of an operating system. In timeshared systems, Round Robin (RR) is most widely used scheduling algorithm. The efficiency of RR algorithm is influenced by the quantum time, if quantum is small, there will be overheads of more context switches and if quantum time is large, then given algorithm will perform as First Come First Served (FCFS) in which there is more risk of starvation. In this paper, a new CPU scheduling algorithm is proposed named as Amended Dynamic Round Robin (ADRR) based on CPU burst time. The primary goal of ADRR is to improve the conventional RR scheduling algorithm using the active quantum time notion. Quantum time is cyclically adjusted based on CPU burst time. We evaluate and compare the performance of our proposed ADRR algorithm based on certain parameters such as, waiting time, turnaround time etc. and compare the performance of our proposed algorithm. Our numerical analysis and simulation results in MATLAB reveals that ADRR outperforms other well-known algorithms such as conventional Round Robin, Improved Round Robin (IRR), Optimum Multilevel Dynamic Round Robin (OMDRR) and Priority Based Round Robin (PRR).

Keywords: CPU, scheduling algorithm, round robin scheduling FCFS, ADRR.

Received August 10, 2015; accepted February 15, 2017

https://doi.org/10.34028/iajit/17/1/11

Full text  

Read 1118 times Last modified on Thursday, 26 December 2019 07:05
Share
Top
We use cookies to improve our website. By continuing to use this website, you are giving consent to cookies being used. More details…