A Novel Approach Towards an Efficient Load Balancing Algorithm in Cloud Computing

Abstract

The need of cloud computing (CC) technology is growing vastly, as it is easier and cheaper to deploy the large scale applications on cloud. Hence the system stability and user amusement is very important. To enact it, it is paramount to use load balancing algorithms (LBA). It acts in the medial of virtual machines (VM) and the data centers (DC) and makes the system dynamic. In the interest to perform load balancing many algorithms has been contemplated to minimize overall response time (ORT), data center processing time (DPT), processing cost. This article analyses the contraposition performances of three subsisting algorithms namely round robin (RR), equally spread current execution (ESCE) and the throttled algorithm (THR). On the basis of study of subsisting algorithms we have contemplated a new algorithm named as “ProposedDualIndex” (PDI) which performs better and overcome the issues of subsisting algorithms. On the thoroughfare to determine a new optimal LBA we maintain two tables for virtual machines (VMs). One is for available VM and other for busy VM by doing so we eliminate process of multiple scanning. Hence reduce the available VM searching time which eventually reduces processing time and cost. The simulation of contemplated scheme is done on cloud analyst tool and assessed on three metrics that are ORT, Datacenter processing time and processing cost. Specifically, the results show that PDI algorithm takes less time than the other three algorithms to access cloud data. The contemplated algorithm can be used for faster accessing of data on CC.