EBTASIC: An Entropy-Based TOPSIS Algorithm for Task Scheduling in IaaS Clouds

Abstract

Objectives: To propose an algorithm to balance resource utilization and revenue generation in the cloud environment. Methods: This study proposes the Entropy-Based TOPSIS algorithm for task scheduling in IaaS Clouds (EBTASIC) to balance resource utilization and revenue generation using the objective-based Entropy Weighting Method (EWM) and Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). Findings/Novelty: Various performance evaluation factors are calculated using EBTASIC and compared with baseline algorithms First Come First Serve (FCFS) and Earliest Deadline First (EDF) algorithms over 12750 lease requests with hard deadlines. The actual response time of EBTASIC is 37.61 percent faster than FCFS and 47.95 percent faster than EDF. Total time spent on lease execution by EBTASIC is reduced by 3.43 percent when compared to FCFS and 3.99 percent when compared to EDF. Turnaround time of EBTASIC is lowered by 21.14 percent when compared to FCFS and 28.81 percent when compared to EDF. EBTASIC throughput is enhanced by 40.80 percent over FCFS and 54.3 percent over EDF. The average response time of EBTASIC is shortened by 9.38 percent compared to FCFS and 14.81 percent compared to EDF. Resource utilization of the proposed algorithm EBTASIC is enhanced by 25.54 percent over FCFS and 6.79 percent over the EDF algorithm. Keywords: Task and VM Scheduling; MCDM Techniques; TOPSIS; Entropy Weighting Method; EBTASIC Algorithm