基于分簇聚类的云架构数据中心网络流量调度方法  

A Clustering-based Network Traffic Scheduling Method for Cloud Architecture Data Center

作　　者：韩茂玲[1] HAN Mao-ling(Department of Information Engineering,Yantai Vocational College,Yantai 264670,China)

机构地区：[1]烟台职业学院信息工程系,山东烟台264670

出　　处：《辽东学院学报（自然科学版）》2021年第2期123-129,共7页Journal of Eastern Liaoning University:Natural Science Edition

摘　　要：云架构数据中心网络流量具有种类繁杂的特点,应用传统的流量调度方法存在丢包率高、CPU利用率低的问题,为此利用分簇聚类技术实现对网络流量调度方法的设计。分别从控制层、核心层、聚合层、边缘层四个层级,搭建云架构数据中心网络的拓扑结构。检测云架构数据中心网络流量,并结合检测结果利用分簇聚类技术实现对网络流量的聚类处理。通过分配网络带宽、设置基本调度约束和流量分类调度控制三个步骤,实现数据中心网络流量调度。实验结果表明,该方法的丢包率降低了6.46%,且对CPU的利用率更高,可以有效提升云架构数据中心的数据传输效率。To solve the problems of high packet loss rate and low CPU utilization of traditional traffic scheduling methods,a clustering-based network traffic scheduling method for cloud architecture data center was proposed.From the control layer,core layer,aggregation layer and edge layer,the topology of cloud architecture data center network was built.The network traffic of cloud architecture data center was detected,and the cluster clustering technology was used to cluster the network traffic.The network traffic scheduling of data center was realized by allocating network bandwidth,setting basic scheduling constraints and traffic classification scheduling control.The experimental results showed that the packet loss rate was reduced by 6.46%,and the CPU utilization was higher,showing that the method could effectively improve the data transmission efficiency of cloud architecture data center.

关 键 词：分簇聚类 云架构 数据中心 网络流量调度 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]