An Efficient Algorithm for Load Balancing in Cloud Computing
Muhammad Rizwan Arshad, Sheheryar Malik · 2014
Cloud computing has emerged as a new model of computing. It enables its users to have a large pool of on-demand shared resources available over the internet. One of the major challenges in cloud computing is the load balancing of resources to optimize the resource usage and reliability. Load balancing is the allocation of workload among several nodes to make sure that there is not a single node which is overwhelmed. Despite being highly reliable, a cloud infrastructure can commence a setback for their availability/reliability due to inappropriate distribution of workloads. Load balancing is one of the technique to evenly distribute the workloads on the computing resources. It helps in achieving the best possible resources' utilization, which leads to increase the overall performance of cloud environment. In this paper, a new algorithm for load balancing in the cloud environment is proposed. To achieve the load balancing, the proposed algorithm maintains a task status table to estimate the real-time load of virtual machines and the expected completion time of the tasks, which can allocate the workload among nodes and realize the load balance. The extensive experiments revealed that the proposed load balancing algorithm performs better than the other existing load balancing techniques/algorithms in cloud environment. It optimizes the resources utilization by reducing the standard deviation of resource usage, reduces overall makespan time for tasks, and fast to reach resources equilibrium state getting proper load distribution.