A Load Balancing Analysis of Cloud Base Application with different Service Broker Policies

Abstract

provisioning and resource optimization are the key issues in cloud computing. To balance the load in across virtual machine load balancing algorithms are classified into two categories i.e. static, dynamic. For homogeneous and stable environment we prefer static load balancing algorithms. For heterogeneous, dynamic environment we prefer dynamic load balancing algorithms. Load balancing may take place in the public, private or hybrid cloud. In this paper, we focus on a load balancing policy i.e. Closest data Center with different no of virtual machines. The evaluation metrics is the response time and data center processing time. Cloud Environment is simulated for the scenario of Internet banking of an international bank in simulation toolkit CloudAnalyst. Using these two evaluation metrics we identify that for real deployment of such customers application what should be a threshold value of key parameters which are supported by the Cluster of users across the Globe. KeywordsCloudlet, Clouds, DVFS, VM, CPU;