作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2019, Vol. 45 ›› Issue (6): 119-126. doi: 10.19678/j.issn.1000-3428.0050981

• 体系结构与软件技术 • 上一篇    下一篇

基于聚类优化的SDN多域自适应管理方法

姚蓝,兰巨龙,胡涛   

  1. 国家数字交换系统工程技术研究中心,郑州 450002
  • 收稿日期:2018-03-28 出版日期:2019-06-15 发布日期:2019-06-15
  • 作者简介:姚蓝(1982—),女,博士研究生,主研方向为软件定义网络;兰巨龙,教授、博士生导师;胡涛(通信作者),硕士研究生。
  • 基金资助:
    国家自然科学基金创新研究群体项目(61521003);国家网络空间安全专项课题(2017YFB0803204)。

Adaptive management method for SDN multi-domain based on clustering optimization

YAO Lan,LAN Julong,HU Tao   

  1. National Digital Switching System Engineering and Technology Research Center,Zhengzhou 450002,China
  • Received:2018-03-28 Online:2019-06-15 Published:2019-06-15

摘要: 针对软件定义网络(SDN)多域网络中存在子域规划及控制器部署不合理的问题,提出一种新的SDN多域自适应管理方法。基于改进的节点聚类思想,设计SDN子域划分优化算法,依据网络中的吸引度和归属度对节点进行聚类,完成子域规划。面向子域中不同网络代价,给出控制器自适应部署算法,并通过代价加权求和得到子域网络总开销,完成控制器部署。仿真结果表明,与控制器自适应部署算法、负载均衡算法等相比,该方法网络子域划分和控制器部署更加合理,可节约23%的备份控制器空间,改善网络的综合性能。

关键词: 软件定义网络, 子域规划, 节点聚类, 控制器部署, 自适应管理

Abstract: Aiming at the problem that sub-domain planning and controller deployment are unreasonable in Software Defined Networking(SDN) multi-domain networks,a new SDN multi-domain adaptive management method is proposed.Based on the improved node clustering idea,the SDN Sub-domain Dividing Optimization Algorithm(DDOA) is designed.The nodes in the network are clustered according to the degree of attraction and attribution,and the sub-domain planning is completed.For the different network costs in the sub-domain,the Controller Adaptive Deployment Algorithm(CADA) is given,and the cost of the sub-domain network is obtained by cost-weighted summation to complete the controller deployment.Simulation results show that compared with the controller adaptive deployment algorithm and load balancing algorithm,the network sub-domain division and controller deployment are more reasonable,which can save 23% backup controller space and improve the overall performance of the network.

Key words: Software-Defined Networking(SDN), sub-domain planning, node clustering, controller deployment, adaptive management

中图分类号: