基于多元回归和Tiebout模型的云服务定价优化方法  

Cloud Service Pricing Optimization Method Based on Tiebout Model and Multiple Regression

作　　者：张德成 王杨 蒋秀林 余超平 ZHANG De-cheng;WANG Yang;JIANG Xiu-lin;YU Chao-ping(School of Basic Courses,Bengbu Medical College,Bengbu Anhui 233000,China;School of Computer and Information,Anhui Normal University,Wuhu Anhui 241000, China)

机构地区：[1]蚌埠医学院公共基础学院,安徽蚌埠233000 [2]安徽师范大学计算机与信息学院,安徽芜湖241000

出　　处：《长春师范大学学报》2018年第12期18-24,共7页Journal of Changchun Normal University

基　　金：国家自然科学基金项目"面向服务质量优化的充电传感网资源分配研究"(61871412);安徽省优秀青年人才基金项目"教学型PACS系统关键技术研究"(2009SQRZ127);教育部产学合作协同育人项目"基于云平台的健康管理APP"(201701045038);安徽省重大人文社科基金项目"基于Multi-agent理论的Web社团与演化路径研究"(SK2014ZD033);安徽省大学生创新创业训练项目"基于物联网的智能社区卫生模拟系统设计研究"(201710367028)

摘　　要：云服务市场快速增长,为了节约运营成本,提升用户满意度,云服务定价优化问题受到学术界和云服务提供商的密切关注。本文在Tiebout模型基础上提出了一种基于多元回归的云服务定价优化方法(Optimized Pricing Model for Cloud Services,OPMCS)。首先分析Tiebout模型并提炼出影响云服务定价的主要因素;再结合云服务平台特征,利用多元回归建立定价优化模型。模型采用密度峰值聚类对资源进行分类以高效合理地将资源分配给用户。此外,考虑到影响定价的因素会随时间而变化,模型引入动态时间因子进行价格修正。实验结果表明,本文提出的OPMCS具有一定的可用性和合理性。With the rapid increase of cloud services market,the pricing issue has been closely paid attention to by cloud service distributors in order to save operating costs and improve users satisfaction.On the basis of Tiebout theory and multiple regression,an optimized pricing model for cloud services(OPMCS)is proposed.Firstly,several major key factors affecting the pricing of public cloud services are refined through the analysis of Tiebout model.Furthermore,the model introduces the method of density peak clustering to classify owned resources in order to share them with users efficiently and reasonably.In addition,the importance of the factors affecting pricing will change along with time,therefore,we add dynamic time factor to modify the pricing model.Experiment results show that the OPMCS is of reasonability and availability to some extent.

关 键 词：多元回归 TIEBOUT模型 云服务 密度峰值聚类 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]