Analysis of load balancing in cloud data centers

Cloud computing is a distributed computing system, where the user will utilize the dynamically provisioned resources including storage, processing, network, etc. This has given rise to cloud data centers, which constitutes virtual resources, that will be shared among multiple users. The major issue in cloud data centers is to handle the millions of simultaneous requests/loads from users. To handle such requests efficiently load balancing algorithms are devised. The incoming load has to be distributed fairly and consistently among the machines which are available. Thus, load balancing techniques deals in achieving high resource utilization by sharing the load efficiently. In this work, Modified Central Load Balancer (MCLB) algorithm is proposed, where the load is balanced among all the available virtual machines thereby avoiding overloading and under loading of virtual machines. Allocation of jobs is done by considering the priority and the state of the virtual machine which helps in the fair allocation of the jobs and efficient user utilization. The MCLB algorithm is simulated using CloudSim and it is compared with existing Round Robin algorithm, Throttled algorithm and Equally Spread Current Execution Load algorithm. The comparison analysis shows that MCLB outperforms the remaining in performance evaluation metrics such as response time, data center processing time and total cost.