基于包簇概念的云资源分配成本优化模型  

Cost Optimization Model for Cloud Resource Allocation Based on Packet Cluster

作　　者：吕腾飞 陈世平[1,2] 王磊 LU Tengfei;CHEN Shiping;WANG Lei(School of Optical-Electrical and Computer Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China;Network and Information Center Office,University of Shanghai for Science and Technology,Shanghai 200093,China)

机构地区：[1]上海理工大学光电信息与计算机工程学院,上海200093 [2]上海理工大学信息化办公室,上海200093

出　　处：《电子科技》2019年第3期31-35,46,共6页Electronic Science and Technology

基　　金：国家自然科学基金(61472256);上海市教委科研创新重点项目(12zz137);上海市一流学科建设项目(S1201YLXK)~~

摘　　要：针对云数据中心因传统资源管理方式造成云服务成本过高问题,文中提出一种基于包、簇概念的资源集中管理分配优化模型。将用户的具体需求抽象为一个个独立的需求包,将数据中心的各类资源整合成一个个资源簇,并将CPU、RAM、带宽作为指标,建立成本评估模型,利用基于包簇概念下的改进粒子群算法,实现需求包到资源簇的部署映射。仿真实验结果表明,该优化模型可以有效降低资源分配过程中的营运成本,稳定提高资源平均利用率。To resolve the high cost in cloud computing data center caused by traditional resource management,a cost optimization model for cloud resource allocation based on packet-cluster was proposed.The model abstracted the specific needs of the users into individual requirements packages and integrated various resources of data center into resource clusters.Through the improved particle swarm algorithm based on the concept of package clustering,the deployment mapping of demand packages to resource clusters were finally implemented.The experimental results showed that the model was not only effective to reduce operating costs in resource allocation process but also improved the resource utilization of physical servers.

关 键 词：云计算 数据中心 资源分配 虚拟机 包簇概念 粒子群算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]