Performance Evaluation of Task Scheduling in Cloud Based on Load Balancing and Ant Colony Optimization

Abstract

Cloud computing became a successful model till the date google proposed it as a business model. It earned lots of praise. Very high volume of data and tasks are submitted to the clouds day by day. Therefore Its management system is too busy with no of large data and no of tasks. It’s necessary to design an effective and efficient algorithm which can give better performance for task scheduling and allocation. Scheduling of tasks in Cloud is an important field of research in order to improve the performance of the cloud system environment. In this work, we proposed a task scheduling technique Load Balancing Ant Colony Scheduling(LBACS) which makes improvement in makespan, waiting time for tasks in cloud system and imbalance degree of virtual machines. It’s based on Ant Colony Optimization(ACO). It also supports load-balancing property. Algorithm is implemented on CloudSim version 4.0. Experimental results show that LBACS algorithmic technic outperforms best as compare to FCFS, RR, GA and ACO. It gives approximately 15-25% improvement over RR, 20-25% improvement over GA, 10-15% improvement over ACO and 30-40% improvement over FCFS algorithm for makespan, waiting time and Imbalance Degree of Vms(ID).