Abstract: Cloud computing is that the next generation of computation. Probably people will have everything they have on the cloud. Cloud computing provides resources to shopper on demand. The resources are also software package resources or hardware resources. Cloud computing architectures square measure distributed, parallel and serves the requirements of multiple purchasers in numerous situations. This distributed design deploys resources distributive to deliver services expeditiously to users in numerous geographical channels. Purchasers in a very distributed setting generate request haphazardly in any processor. Therefore the major disadvantage of this randomness is related to task assignment. The unequal task assignment to the processor creates imbalance i.e., a number of the processors square measure over laden and a few of them square measure beneath loaded. The target of load balancing is to transfer the load from over laden method to beneath loaded method transparently. Load balancing is one in all the central problems in cloud computing. To realize high performance, minimum interval and high resource utilization magnitude relation we want to transfer the tasks between nodes in cloud network. Load balancing technique is employed to distribute tasks from over loaded nodes to beneath loaded or idle nodes. In following sections we tend to square measure discuss concerning cloud computing, load balancing techniques and also the planned work of our load balancing system.Proposed load balancing algorithm is simulated on Cloud Analyst toolkit. Performance is analyzed on the parameters of overall response time, data transfer, average data center servicing time and total cost of usage. Results are compared with three existing load balancing algorithms namely Round Robin, Equally Spread Current Execution Load, and Throttled. Results on the basis of case studies performed shows more data transfer with minimum response time.

Keywords: Cloud Computing; Load balancing; Load balancing Algorithms; IaaS.