MOTS‐ACO: An improved ant colony optimiser for multi‐objective task scheduling optimisation problem in cloud data centres

Abstract

Task scheduling in cloud data centres is an optimisation problem that aims to minimise power consumption and task makespan as well as ensures the quality of service delivered to cloud consumers. Although there are several existing task scheduling approaches, these methods mainly focus on optimising makespans of tasks while ignoring critical issues. This paper presents a comprehensive multi-objective task scheduling model based on an improved Ant Colony Optimisation (ACO) algorithm, referred to as MOTS-ACO. In order to promote the diversity of the Pareto set and accelerate the convergence speed, adaptive distribution probability is incorporated into the proposed algorithm, specifically in the process of updating the global rule. The performance of MOTS-ACO is compared with several existing multi-objectives task scheduling algorithms based on the makespan time, turnaround time, power efficiency and load balancing parameters. The results show the superiority of MOTS-ACO in terms of the makespan time, turnaround time, power efficiency and load balancing. Moreover, the proposed MOTS-ACO algorithm introduces more diversity in the search and accelerates the convergence speed towards the Pareto optimal solution. K E Y W O R D S ACO, cloud computing, load balancing, makespan, power efficiency, task scheduling, turnaround time This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

SLA Aware Task-Scheduling Algorithm in Cloud Computing Using Whale Optimization Algorithm

“…Objectives of resource scheduling algorithms modeled by diferent nature-inspired algorithms [7] Adaptive PSO Makespan, throughput, resource utilization [9] MCFCM-PSO Load balancing, makespan [10] PSO-RDAL Makespan, response time, penalty cost, total execution cost [11] CR-PSO Makespan, execution time, execution cost, energy consumption [5] GA-ACO Response time, task completion time, throughput [12] PSO-ACO Makespan, resource utilization, total computation cost [13] CCSA Makespan, overall cost [14] MOTS-ACO Makespan, turnaround time, power consumption [15] EDA-GA Task completion time, load balance of tasks [16] EPETS Energy consumption [17] MVO-GA Task transfer time [18] GACCRATS Makespan, customer satisfaction [19] MGGS Makespan, response time, QoS [20] OCSA Makespan, cost [24] CGA Task completion time, total execution cost [25] IWC Task scheduling time, scheduling cost [26] SLNO Energy, power consumption, resource utilization [27] GCWOAS2 Task completion time, load balance of virtual resources [28] GAGELS Makespan, resource utilization [29] DILS Makespan, learning rate…”

“…It was compared over MO-ACO, Min-Min, and ACO algorithms, and the results showed that CCSA outperforms existing algorithms. In [14], scheduling approach designed by updating pheromone leads to increase in acceleration of ant exploration in solution space. MOTS-ACO was used as methodology for designing scheduling problem.…”

An Enhanced Trust Scheduling Algorithm for Medical Applications in a Heterogeneous Cloud Computing Environment

“…It specifies that 2 instances of type1 virtual machine can be executed, but if it supposes there are continuous type 1 request, then the type 1 and type 2 instances will be pushed to a starvation state. This is the problem considered and solved [20][21][22][23][24][25].…”

An Efficient Trust-Aware Task Scheduling Algorithm in Cloud Computing Using Firefly Optimization

“…The results revealed that FPSO-GA had a fuzzy nature, which proved that it outperforms the existing approaches by efficiently balancing the load among VMs. In [ 22 ], the authors formulated an efficient power-aware task scheduling algorithm developed using a modified ACO approach. This algorithm was modeled based on updating pareto by accelerating the convergence using adaptive probability distribution.…”