Cloud-based load balancing using double Q-learning for improved Quality of Service

Cloud computing improves the performance of software applications by providing on-demand usage, high availability, reliability, and agility. However, during peak traffic conditions the resources in cloud services can become over-utilized, impairing the ability to provide performance levels specified in service-level agreements. Therefore, a load balancing algorithm that provides an efficient and fair allocation of cloud resources while providing high availability to end users is a timely necessity. In this paper, we propose a load balancing scheme to distribute the workload among virtual servers using a modified version of the double Q-learning algorithm. The proposed algorithm is implemented on a load balancing controller and leverages user requests using software defined network technologies. The results reveal a considerable reduction in terms of unsatisfied cloud consumers compared to already existing popular algorithms. In short, this work will serve as a future guide for load balancing implementations in cloud environments that require higher Quality of Service.