基于MILP的云计算数据中心扩张策略优化模型  

Optimizing Expansion Strategies Based on MILP for Cloud Computing Data Centers

作　　者：宋俊锋[1] 

机构地区：[1]丽水学院工程与设计学院,浙江丽水323000

出　　处：《湘潭大学自然科学学报》2015年第4期105-110,共6页Natural Science Journal of Xiangtan University

基　　金：浙江省教育科学规划项目(2014SCG386);浙江省教育技术研究规划课题(JB055)

摘　　要：随着云计算系统的日益发展,云服务提供商需要对大规模数据中心进行不断扩张,为此,提出一种基于混合整数线性规划(MILP)的数据中心扩张策略优化模型.综合考虑了流量负载、能耗、电力成本、土地成本、税率、服务器利用率等多种约束,构建MILP模型,在满足用户的服务级别协议(SLA)下,求解最优扩张策略,使服务商利润最大化.其中,扩张策略包括扩建现有数据中心和在何时何地新建数据中心等措施.仿真实验结果表明,该模型能够根据实际情况获得最优扩张策略,提高了利润.With the increasing development of cloud computing systems, the large-scale data center is need to expand by cloud service providers, therefore, an optimizing expansion strategies based on mixed in- teger-linear programming（MILP） for cloud computing data centers is proposed. The energy consumption, electricity costs, land costs, tax rate, server utilization and many kinds of constraint are considered to build MILP model, under meet the user service level agreement （SLA）, solving the optimal expansion strategy to maximize the profits of service providers. The expansion strategy includes the expansion of the existing data center or / and when and where to build new data centers. The simulation results show that the model can obtain the optimal expansion strategy according to the actual situation, and improve the profit.

关 键 词：云计算 数据中心扩张 混合整数线性规划 优化模型 服务级别协议 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]