An Optimum Time Quantum Using Linguistic Synthesis for Round Robin Scheduling Algorithm

TL;DR

This paper proposes a dynamic, fuzzy logic-based method to optimize the time quantum in Round Robin CPU scheduling, aiming to reduce context switches and improve turnaround time.

Contribution

It introduces a Mamdani Fuzzy Inference System to generate an optimal time quantum based on imprecise parameters, enhancing scheduling performance.

Findings

Reduced context switches

Improved turnaround time

Enhanced scheduling efficiency

Abstract

In Round Robin CPU scheduling algorithm the main concern is with the size of time quantum and the increased waiting and turnaround time. Decision for these is usually based on parameters which are assumed to be precise. However, in many cases the values of these parameters are vague and imprecise. The performance of fuzzy logic depends upon the ability to deal with Linguistic variables. With this intent, this paper attempts to generate an Optimal Time Quantum dynamically based on the parameters which are treated as Linguistic variables. This paper also includes Mamdani Fuzzy Inference System using Trapezoidal membership function, results in LRRTQ Fuzzy Inference System. In this paper, we present an algorithm to improve the performance of round robin scheduling algorithm. Numerical analysis based on LRRTQ results on proposed algorithm show the improvement in the performance of the system by reducing unnecessary context switches and also by providing reasonable turnaround time.