Two objectives big data task scheduling using swarm intelligence in cloud computing

Cloud computing is the latest and the most used type of distributed computing systems and also it covers most of their features. It has been widely used for its enormous benefits and its ability to cope with large scale data such as workflows and big data applications. On the other hand, scheduling algorithms; starting from traditional to Hyper-heuristic; are widely used in computing systems such as cloud computing to monitor the use of resources. However, these scheduling algorithms vary in term of their performance and most of these traditional and simple scheduling algorithms may not be efficient for large scale data. Although many scheduling algorithms have been implemented for cloud computing, it has been realized that most of the applications nowadays require different objectives that simple scheduling algorithms fail to achieve. Either one of the objective is violated or the results are far from the optimal solution. In this direction, this paper first gives review of some previous scheduling algorithms used in cloud. Then, it proposes a type of swarm intelligence called Particle Swarm Optimization (PSO) algorithm to diminish cost though meeting deadlines. The proposed method is evaluated using CloudSim and big data applications are used as sample of applications. From the results, it can be seen that PSO works better for big data applications and the cost is reduced to more than half when compared with ordinary scheduling algorithms such as First-Come-First-Serve (FCFS)