Improving task scheduling by using a fuzzy reasoner

Issue title: Special collection of papers on ‘Current Fuzzy Logic-based Software Applications and Systems’

Guest editors: Konstantina Chrysafiadi

Article type: Research Article

Authors: Chrysafiadi, Konstantina

Affiliations: Department of Informatics, University of Piraeus, Piraeus, Greece | E-mail: [email protected]

Correspondence: [*] Corresponding author: Departmenmt of Informatics, University of Piraeus, Piraeus, Greece. E-mail: [email protected].

Abstract: In this paper a rule-based fuzzy scheduling method is presented. Aim of this method is to handle the uncertainty and vagueness that characterize the complicated process of task scheduling. It uses fuzzy sets to describe both a task’s deadline and a task’s remaining execution time. The operation of the presented scheduling method is based on a rule-based reasoner that decides dynamically about the priority of the tasks that wait to be executed. This reasoner is triggered each time a change occurs (e.g. the execution of a task ends, a new task arrives etc.). It has been compared with Earlied Deadline First (EDF) scheduling algorithm. The results showed that the presented fuzzy scheduling method improves the scheduling process, since it minimizes both, the mean waiting time and the mean turnaround time.

Keywords: Task scheduling, fuzzy logic, rule-based fuzzy reasoner, SRTF, EDF

DOI: 10.3233/IDT-190110

Journal: Intelligent Decision Technologies, vol. 14, no. 2, pp. 165-170, 2020