电子学报 ›› 2021, Vol. 49 ›› Issue (3): 484-491.DOI: 10.12263/DZXB.20200120

• 学术论文 • 上一篇    下一篇

一种基于图神经网络的SDN路由性能预测模型

车向北1, 康文倩1, 邓彬1, 杨柯涵2, 李剑2   

  1. 1. 深圳供电局有限公司, 广东深圳 518000;
    2. 北京邮电大学计算机学院, 北京 100876
  • 收稿日期:2020-01-20 修回日期:2020-04-17 出版日期:2021-03-25 发布日期:2021-03-25
  • 作者简介:车向北 男,1984年08月出生,陕西宝鸡人,深圳供电局有限公司高级工程师.主要从事电力监控系统网络安全工作.E-mail:chexiangbei@163.com;康文倩 女,1988年04月出生,江苏徐州人,深圳供电局有限公司工程师.主要从事电力监控系统网络安全工作.E-mail:wenqiankang@163.com;邓彬 男,1989年4月出生,湖北黄石人,深圳供电局有限公司系统运行部工程师.主要从事调度自动化系统、智能调度技术等方向研究.E-mail:15219495096@163.com;杨柯涵 男,1996年4月出生,陕西汉中人,北京邮电大学计算机学院硕士研究生,主要研究领域为深度学习,计算机网络.E-mail:kehanyang@bupt.edu.cn;李剑 男,1976年12月出生,陕西西安人,北京邮电大学计算机学院教授、博士生导师.主要从事密码学、网络空间安全、量子信息、计算机应用、人工智能、软件定义网络等方向研究.E-mail:lijian@bupt.edu.cn
  • 基金资助:
    国家自然科学基金(No.U1636106);北京市自然科学基金(No.4182006)

A Prediction Model of SDN Routing Performance Based on Graph Neural Network

CHE Xiang-bei1, KANG Wen-qian1, DENG Bing1, YANG Ke-han2, LI Jian2   

  1. 1. Shenzhen Power Supply Bureau Co., Ltd., Shenzhen, Guangdong 518000, China;
    2. School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China
  • Received:2020-01-20 Revised:2020-04-17 Online:2021-03-25 Published:2021-03-25
  • Supported by:
     

摘要: 软件定义网络作为未来网络架构的发展方向,通过分离数据平面与控制平面高效设定路由方案.而在路由方案的优化过程中,准确预估给定路由方案下的网络性能是其关键.本文基于图神经网络建模网络中物理链路与路由方案路径的关系,在给定的路由方案与网络流量下对网络中的各项端到端性能指标(如延迟、抖动)进行准确预估,以辅助优化路由方案.本文基于OMNeT++来生成数据并进行实验,实验结果表明本文提出的模型能够针对延迟抖动等端到端性能指标进行准确预估,预估平均相对误差不超过4.1%.实验也对比了传统最短路径路由算法与基于该预测模型给出的最优路由方案下的端到端性能,相比传统最短路径路由算法,平均延迟和平均抖动分别降低了19.8%和33.52%,最大延迟和最大抖动降低了36.18%和35.45%.

 

关键词: 软件定义网络, 端到端性能预测, 图神经网络, SDN路由优化

Abstract: As the development direction of future network architectures,Software Defined Networks can efficiently set routing schemes by separating the data plane and the control plane.In the process of optimizing a routing scheme,it is the key to accurately predict the network performance under a given routing scheme.This paper uses graph neural networks to model the relationship between physical links and routing scheme paths,so that the model can predict various end-to-end performance indicators (such as delay and jitter) in the network under a given routing scheme and network traffic.This paper uses OMNeT ++ to generate datasets.The experimental results show that the model proposed in this paper can accurately predict end-to-end performance indicators such as delay and jitter.The average relative error of the estimate does not exceed 4.1%.The experiment also compares the end-to-end performance of the traditional shortest path routing algorithm with the optimal routing scheme based on the prediction model proposed in this paper.Compared to the traditional shortest path routing algorithm,the average delay and average jitter are reduced by 19.8% and 33.52%,and the maximum delay and maximum jitter are reduced by 36.18% and 35.45%.

Key words: software development network(SDN), end-to-end performance prediction, graph neural network, SDN routing optimization

中图分类号: