Reduction of Makespan Using Ant Colony Optimization in Clouds

Abstract

Cloud computing has become a modern age technology. Every other organization is moving its infrastructure to the Cloud. It is now becoming a necessity rather than being a luxury. Cloud computing services are now becoming a household name among the end users. In order to deploy these services without any hindrance it is important to handle task scheduling. Task scheduling on the Cloud not only helps achieve judicious use of resources but also help reduce number of machines involved in execution. The algorithm which has been proposed as a part of this work is based on Ant Colony Optimization algorithm. This heuristic based optimization algorithm has been inspired from the foraging behavior of ants. One of the main features of such algorithms is that with each iteration we get an improved result until we achieve an optimal solution to our problem. The main objective of the proposed algorithm is to reduce the makespan i.e. the total execution time of the tasks at hand. This will be achieved by making an optimal schedule which gives minimum execution time. This algorithm has been compared with other conventional scheduling algorithms like First Come First Serve (FCFS) algorithm and Round Robin (RR) algorithm. The obtained results show that the algorithm based on Ant Colony Optimization gives better results than FCFS and RR algorithms. The algorithm has been implemented using an open source platform called CloudSim. It helps simulate and model Cloud computing scenarios to help in getting better results.