Cloud task scheduling using nature inspired meta-heuristic algorithm

In this paper we investigate the application of Meta-Heuristic for cloud task scheduling on Hadoop. Hadoop is an open source implementation of MapReduce framework which extensively used for processing computational intensive jobs on huge amount of data over multi-node cluster. In order to achieve an efficient execution schedule, the scheduling algorithm requires to determining the order and the node on which tasks will be executed. A scheduling algorithm uses execution time, order of task arrival and location of data (i.e., assign task to the node which contains the required data) to determine the best execution schedule. We use Particle Swarm Optimization (PSO) to determine the tasks execution schedule and compare with tasks schedules obtained from other techniques like Genetic Algorithm (GA), Brute Force (BF) algorithm, First In First Out (FIFO) algorithm and Delay Scheduling Policy (DSP) algorithm. The results of this study prove the significance of PSO algorithm for cloud task scheduling over other algorithms.