A Hybrid Approach Using ACO-GA for Task Scheduling in Cloud

Abstract

AbstractCloud computing is a fast-growing technology in today's world that offers a wide variety of services based on the need of its users. Cloud provides services like computing, storage, and networking which are accessible through the Internet. In the computing cloud, task scheduling plays a vital role in optimizing resource utilization and providing quality of service (QoS) to the customer. Efficient task scheduling is needed to fulfill the user requirement and system performance. Traditional task scheduling algorithms in a cloud platform like max–min, shortest job first, and round-robin are not so effective in reducing makespan and cost. The main motive behind this research is to apply a combination of the genetic algorithm for task scheduling in a cloud environment and analyzes its effect, as the genetic algorithm can efficiently solve NP-hard problems. This paper proposes a hybrid approach task that combines the advantages of the two most widely used evolutionary algorithms: genetic algorithm (GA) and ant colony optimization (ACO) for improving the scheduling in the cloud using hybrid approach to overcome the limitation of unnecessary diversity. The experimental result conclusively proved that there is an 18% to 20% reduction in cost and makespan by ACO-GA-based task scheduling in the cloud as compared to simple GA and ACO.KeywordsAnt colony optimization (ACO)Genetic algorithm (GA)Task scheduling