Title: An improved dynamic task scheduling algorithm based on INTOPSIS and PSO
Authors: Aditya Makwe; Priyesh Kanungo; Deepak Sukheja
Addresses: Institute of Engineering and Technology, Devi Ahilya Vishwavidyalaya, Indore, MP, India ' School of Computer Science and IT, Devi Ahilya Vishwavidyalaya, Indore, MP, India ' VNR Vignana Jyothi Institute of Engineering and Technology, Telangana, Hyderabad, India
Abstract: In cloud computing, efficient scheduling policy is needed to schedule user tasks on its resource. Due to the availability of many cloud service providers, allocating hosts and virtual machines of its data centre to user tasks requires an efficient scheduling technique. To address this problem, this study aims to discuss interval neutrosophic technique for order of preference by similarity to ideal solution (INTOPSIS) scheduling policy with particle swarm optimisation (PSO). First, INTOPSIS is used to determine rank of tasks; second, PSO is used to schedule the tasks on the virtual machine. Cloudsim is used to simulate the effectiveness of the proposed technique. The work's performance is compared to AHP-TOPSIS, TOPSIS-PSO, and PSO techniques in terms of average makespan, resource consumption, transmission delay and with ABC, IABC, and TOPSIS-PSO in terms of cost. Proposed method shows 2.5% to 5% decrement in transmission delay and 30% to 40% decrease in processing cost.
Keywords: particle swarm optimisation; PSO; INTOPSIS; cloud computing; task measurement index; TMI.
International Journal of Cloud Computing, 2024 Vol.13 No.4, pp.368 - 403
Received: 13 Sep 2023
Accepted: 06 Nov 2023
Published online: 20 Aug 2024 *