A hierarchical control framework of load balancing and resource allocation of cloud computing services

Service providers must guarantee Quality of Service (QoS) requirements of the co-hosted applications in a data center and simultaneously achieve optimal utilization of their infrastructure under varying workload. This paper presents a hierarchical control framework that aims at compromising antagonistic objectives inside a data center. The local control level tackles simultaneously the problems of resource allocation and admission control of virtual machines while the upper level addresses together the load balancing of the incoming requests and placement of virtual machines into a cluster of physical servers. Numerical results show that the cooperation of the two control layers guarantees the satisfaction of the system's constraints and the user's requirements towards the fluctuations of incoming requests.