Study of load balancing algorithms for Cloud Computing

The significance of distributed applications is constantly rising due to technological advancements such as increasing internetworking of various computing devices and widespread usage of smart phones. Cloud computing is an information technology paradigm that enables ubiquitous access to shared pools of configurable system resources. Load balancing is one among various challenging issues in cloud computing. It is a mechanism to distribute the user requests among the virtual machines so that the requests are assigned proportional to the capacity of each virtual machine. Balancing of requests prevents the virtual machines from being either overloaded or under loaded. This paper presents a comparison of performance of load balancing algorithms in cloud computing. The algorithms considered for study are Throttled and Equally Spread Current Execution. Particle swarm optimization is also simulated to solve load balancing dynamically. The proposed work compares response time of all the three techniques. The simulation results show that particle swarm optimization performs better for dynamic changes in the system.