作者: Jing Bi , Haitao Yuan , Wei Tan , MengChu Zhou , Yushun Fan
DOI: 10.1109/TASE.2015.2503325
关键词: Differentiated service 、 Computer science 、 Profit maximization 、 Metaheuristic 、 Simulated annealing 、 Service provider 、 Cloud computing 、 Distributed computing 、 Dynamic priority scheduling 、 Provisioning
摘要: A key factor of win–win cloud economy is how to trade off between the application performance from customers and profit providers. Current researches on resource allocation do not sufficiently address issues minimizing energy cost maximizing revenue for various applications running in virtualized data centers (VCDCs). This paper presents a new approach optimize VCDC based service-level agreements (SLAs) service providers customers. precise model external internal request arrival rates proposed virtual machines at different classes. An analytic probabilistic then developed non-steady states. In addition, smart controller fine-grained provisioning sharing among multiple applications. Furthermore, novel dynamic hybrid metaheuristic algorithm formulated maximization problem, simulated annealing particle swarm optimization. The can guarantee that differentiated qualities be provided with higher overall lower cost. advantage validated trace-driven simulations.