A Novel QoS-Aware Load Balancing Mechanism in Cloud Environment

Abstract

Efficient low-level resource provisioning and QoS guaranteed are key challenges for cloud computing. When using virtual machines cluster to tackle various tasks scheduling, the target is to assign the tasks to each of the available nodes evenly in in premise of ensuring QoS of services, and it also means that the cloud providers should consider to reduce the load of overload nodes, and improve resource utilization of under-loading nodes. There are some limitations when applying these classic scheduling algorithms to the cloud computing environment. In order to solve this problem, we propose a novel QoS-aware load balancing mechanism in cloud environment. The key of this mechanism includes QoS model, resource model, and task model. We conduct a CloudSim based experiment to evaluate our method using a realistic dataset, and the results show that the algorithm proposed effectively shortens the waiting time in comparison to RR algorithm and Max-Min algorithm.