Scalability-aware Scheduling Optimization Algorithm for Multi-Objective Cloud Task Scheduling Problem

This paper put forward a novel scalability-aware scheduling optimization algorithm called Cloud Scalable Multi-Objective Cat Swarm Optimization Based Simulated Annealing (CSM-CSOSA) for solving task scheduling optimization problem in cloud computing environment. The novelty of the algorithm is based on the improvement of its local search procedure using improved simulated annealing optimization approach. The goal is to provide a solution that can adapt the dynamic changing cloud tasks and resources while minimizing the amount of time and cost of processing task in order to meet cloud consumers' QoS expectations. In order to determine the performance of our proposed task scheduling algorithm. A task scheduling model based on execution time and execution cost objectives is presented. Implementation of the algorithm is carried out using CloudSim tool and evaluated based on metrics of execution time, execution cost, and scalability. Comparison with similar task scheduling algorithms like Multi-Objective Genetic Algorithm (MOGA), Multi-Objective Scheduling Based on Ant Colony Optimization (MOSACO) and Multi-Objective Particle Swarm Optimization (MOPSO) is carried out. The results obtained show our proposed algorithm has achieved a remarkable performance in term of minimizing task execution time, execution cost and returned better scalability performance. The proposed algorithm is therefore recommended for large task scheduling for cloud computing environment.