Dynamic Fine-Grained Resource Provisioning for Heterogeneous Applications in Virtualized Cloud Data Center

Abstract

The balance between customer-perceived application performance and cloud provider's profit is a key to achieve win-win in cloud economy. Current researches on cloud resource allocation do not sufficiently address the issue of minimizing energy cost and maximizing revenue for various applications in virtualized cloud data center (VCDC). This paper presents a new approach to realize the optimization of VCDC's profit based on the service-level agreements between cloud providers and customers. A precise model of the external and internal request arrival rates is proposed for virtual machines of different service classes. An analytic probabilistic model is then developed for non-equilibrium VCDC states. Next, a smart controller is proposed for fine-grained resource provisioning and sharing among multiple applications. A novel hybrid meta-heuristic algorithm based on simulated annealing and particle swarm optimization is developed to solve the formulated profit maximization problem. The proposed algorithm can guarantee that differentiated service qualities can be provided with higher overall performance and lower energy cost. Finally, the effectiveness of the proposed approach is validated with trace-driven simulation.