Analysis of Particle Swarm Optimization and Genetic Algorithm based on Task Scheduling in Cloud Computing Environment

Since the beginning of cloud computing technology, task scheduling problem has never been an easy work. Because of its NP-complete problem nature, a large number of task scheduling techniques have been suggested by different researchers to solve this complicated optimization problem. It is found worth to employ heuristics methods to get optimal or to arrive at near-optimal solutions. In this work, a combination of two heuristics algorithms was proposed: particle swarm optimization (PSO) and genetic algorithm (GA). Firstly, we list pros and cons of each algorithm and express its best interest to maximize the resource utilization. Secondly, we conduct a performance comparison approach based on two most critical objective functions of task scheduling problems which are execution time and computation cost of tasks in cloud computing. Thirdly, we compare our results with other existing heuristics algorithms from the literatures. The experimental results was examined with benchmark functions and results showed that the particle swarm optimization (PSO) performs better than genetic algorithm (GA) but they both present a similarity because of their population based search methods. The results also showed that the proposed hybrid models outperform the standard PSO and reduces dramatically the execution time and lower the processing cost on the computing resources. Keywords—Execution Time; Task Scheduling Algorithms; Particle Swarms (PSO); Genetic Algorithm (GA); Virtual Machines (VMs)