Design of Oppositional Glow worm Swarm Optimization based Resource Scheduler on Big Data Environment

Abstract

Big data is commonly used to support significant exploitation of processing resources, concentrating on on-demand services and resource scalability.  With numerousmethods available, controlling massive quantity of data in several data centers is still a tedious task. Particularly, resource scheduling (RS) is treated as a way of distributing resources by an efficient decision-making process with the aim of assisting desired tasks over time. The combination of heterogeneous computing resources using the Big Data users enables the chance of minimizing the energy utilization and maximizing resource efficiency. But the state of art RS techniques needs to boost the scheduling performance in the big data environment. In this aspect, this paper designs an Oppositional Glowworm Swarm Optimization based Resource Scheduling (OGSO-RS) scheme for big data environment. The proposed OGSO-RS technique aims to allocate the resources proficiently in the big data platform. The searching area and a large amount of data are provided as input to the geo-distributed datacenter, where the population initialization of glowworms takes place. In addition, the MapReduce function computes the optimal resource, and thereby the efficiency can be improvised. Moreover, the load can be allotted to the datacenters by minimizing the computational cost and storage area.  In order to showcase the improved performance of the proposed OGSO-RS technique, a series of experiments were carried out. The simulation results highlighted the betterment of the RS efficiency of the OGSO-RS technique compared to other existing approaches.