Comparative Analysis of Load Balancing Algorithm in Cloud Computing

Abstract

Cloud computing is the delivery of computing services—including servers, storage, databases, networking, software, analytics, and intelligence over the Internet. To provide computing services to users in efficient manner, there is a need of effective utilization of resources across the VM. But in today’s world there are billions of users who are dependent on cloud computing services this leads to a huge no. of traffic across data centers. While serving users’ request, some data center will have a maximum number of users' task traffic and some data center will have less amount of traffic. These problems create unbalancing of loads across the data centers. To rectify this problem, we need a load balancer or load balancing technique. Load Balancing technique is a mechanism of redistributing users' task/request across the data centers so that no virtual machine should have maximum no. of users' task traffic, less no. of users’ task traffic or no users’ request. The effectiveness of a load balancing technique depends upon the algorithms used behind it. The cloud analyst simulator is such a tool which uses various types of algorithms, for example—throttled algorithm, round robin algorithm, etc., and different service broker policy like closest data center, and some other parameters (e.g., data center configuration), with the help of cloud analyst we can find out data center service request time and total cost for particular data center. The objective of this paper is to compare the service request time and total cost (VM cost and total data transfer cost) of a data center by using existing load balancing algorithms with different service broker policies.