一种基于图神经网络的SDN路由性能预测模型

doi:10.12263/DZXB.20200120

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

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. 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
- 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. 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

摘要/Abstract

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:

TP393.0

车向北, 康文倩, 邓彬, 杨柯涵, 李剑. 一种基于图神经网络的SDN路由性能预测模型[J]. 电子学报, 2021, 49(3): 484-491.

CHE Xiang-bei, KANG Wen-qian, DENG Bing, YANG Ke-han, LI Jian. A Prediction Model of SDN Routing Performance Based on Graph Neural Network[J]. Acta Electronica Sinica, 2021, 49(3): 484-491.

参考文献

[1] Ciucu F,Schmitt J B.Perspectives on network calculus:no free lunch,but still good value[J].Acm Special Interest Group on Data Communication,2012,42(4):311-322.
[2] Meng D,Sun L.Somenew trends of deep learning research[J].Chinese Journal of Electronics,2019,28(6):1087-1090.
[3] Wang M,Cui Y,Wang X,et al.Machine learning for networking:workflow,advances and opportunities[J].IEEE Network,2018,32(2):92-99.
[4] Mao B,Fadlullah Z M,Tang F,et al.Routing or computing? the paradigm shift towards intelligent computer network packet transmission based on deep learning[J].IEEE Transactions on Computers,2017,66(11):1946-1960.
[5] Chen X,Guo J,Zhu Z,et al.Deep-RMSA:Adeep-reinforcement-learning routing,modulation and spectrum assignment agent for elastic optical networks[A].2018 Optical Fiber Communications Conference and Exposition (OFC)[C]. San Diego:IEEE,2018.1-3
[6] 张德干,葛辉,刘晓欢,等.一种基于Q-Learning策略的自适应移动物联网路由新算法[J].电子学报.2018,46(10):2325-2332. Gan Z D,Hui G E,Huan L X,et al.A kind of new routing algorithm with adaptivity for mobile IOT based on Q-learning[J].Acta Electronica Sinica,2018,46(10):2325-2332.(in Chinese)
[7] Xu Z,Tang J,Meng J,et al.Experience-driven networking:a deep reinforcement learning based approach[A].International Conference on Computer Communications[C].Honolulu:IEEE,2018.1871-1879.
[8] Mestres A,Alarcón E,Ji Y,et al.Understanding the modeling of computer network delays using neural networks[A].Proceedings of the 2018 Workshop on Big Data Analytics and Machine Learning for Data Communication Networks[C].Budapest Hungary:ACM,2018.46-52.
[9] Shihan Xiao,Dongdong He,Zhi bo Gong.2018.Deep-Q:traffic-driven QoS inference using deep generative network[A].Proceedings of the Workshop on Network Meets AI & ML[C].Budapest Hungary:ACM,2018.67-73.
[10] Zhou J,Cui G,Zhang Z,et al.Graph neural networks:A review of methods and applications[EB/OL].https://arxiv.org/pdf/1812.08434.pdf,2018.
[11] Xiao L,Lu R.A fully complex-valued gradient neural network for rapidly computing complex-valued linear matrix equations[J].Chinese Journal of Electronics,2017,26(6):1194-1197.
[12] Gilmer J,Schoenholz S S,Riley P,et al.Neural message passing for quantum chemistry[A].International Conference on Machine Learning[C].Sydney:ICML,2017.1263-1272.
[13] Hamaguchi T,Oiwa H,Shimbo M,et al.Knowledge transfer for out-of-knowledge-base entities:a graph neural network approach[A].International Joint Conference on Artificial Intelligence[C].Melbourne:IJCAI,2017.1802-1808.
[14] A Fout,J Byrd,B Shariat,A Ben-Hur.Protein interface prediction using graph convolutional networks[A].Advances in Neural Information Processing Systems[C].Long Beach:NIPS,2017. 6530-6539.
[15] 胡涛,张建辉,邬江,等.SDN中基于分布式决策的控制器负载均衡机制[J].电子学报,2018,46(10):2316-2324. Tao Hu,Jian-Hui Zhang,Jiang Wu,et al.Controller load balancing mechanism based on distributed policy in SDN[J].Acta Electronica Sinica,2018,46(10):2316-2324.(in Chinese)
[16] Hamilton W L,Ying Z,Leskovec J,et al.Inductive representation learning on large graphs[A].Advances in neural information processing systems.[C].Long Beach:NIPS,2017.1024-1034.
[17] Staudemeyer R C,Morris E R..Understanding LSTM-a tutorial intol long short-term memory recurrent neural networks[EB/OL].https://arxiv.org/pdf/1909.09586.pdf,2019.
[18] András Varga,Rudolf Hornig.An overview of the OMNeT++ simulation environment[A].Proceedings of the 1st International Conference on Simulation Tools and Techniques for Communications,Networks and Systems & Workshops[C].Marseille France:ICST, 2008.1-10.
[19] Hei X,Zhang J,Bensaou B,et al.Wavelength converter placement in least-load-routing based optical networks using genetic algorithms[J].Journal of Optical Networking,2004,3(5):363-378.
[20] Barreto F,Wille E C,Nacamura L,et al.Fast emergency paths schema to overcome transient link failures in OSPF routing[J].International Journal of Computer Networks & Communications,2012,4(2):17-34.
[21] Pedro J,Santos J,Pires J,et al.Performance evaluation of integrated OTN/DWDM networks with single-stage multiplexing of optical channel data units[A].International Conference on Transparent Optical Networks[C].Stockholm:IEEE,2011.1-4.
[22] Qin A K,Huang V L,Suganthan P N,et al.Differential evolution algorithm with strategy adaptation for global numerical a Optimization[J].IEEE Transactions on Evolutionary Computation,2009,13(2):398-417.
[23] 王福才,周鲁苹.混合精英策略的元胞多目标遗传算法及其应用[J].电子学报.2016,(3):709-717 Fucai Wang,Lupin Zhou.Cellular Multi-objective genetic algorithm based on hybrid elite and application[J].Acta Electronica Sinica,2016(3):709-717.(in Chinese)

一种基于图神经网络的SDN路由性能预测模型

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

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	毛国君, 王者浩, 黄山, 王翔. 基于剪边策略的图残差卷积深层网络模型[J]. 电子学报, 2022, 50(9): 2205-2214.
[2]	袁冠, 邴睿, 刘肖, 代伟, 张艳梅, 蔡卓. 基于时空图神经网络的手势识别[J]. 电子学报, 2022, 50(4): 921-931.
[3]	柴蓉, 谢德胜, 陈前斌. 基于成本及功耗联合优化的SDN虚拟网络映射算法[J]. 电子学报, 2021, 49(8): 1615-1624.
[4]	陈亮, 李峰, 任保全, 杨建喜. 软件定义物联网研究综述[J]. 电子学报, 2021, 49(5): 1019-1032.
[5]	徐川, 胡渝, 韩珍珍, 熊郑英, 赵国锋. 基于链路效用的3D-VANET可靠路由算法[J]. 电子学报, 2021, 49(5): 872-878.
[6]	王莅晟, 伊鹏, 胡涛, 江逸茗, 胡静萍, 胡宗魁. SDN中基于全局拓扑感知的自适应流量均衡算法[J]. 电子学报, 2021, 49(5): 964-974.
[7]	孙鹏浩, 兰巨龙, 申涓, 胡宇翔. 一种基于深度增强学习的智能路由技术[J]. 电子学报, 2020, 48(11): 2170-2177.
[8]	黄美根, 郁滨. 软件定义WSN规则一致更新研究[J]. 电子学报, 2019, 47(9): 1965-1971.
[9]	姚蓝, 胡涛, 伊鹏, 胡宇翔, 兰巨龙, 李子勇. SDN中基于效能优化的交换机动态迁移策略[J]. 电子学报, 2019, 47(7): 1482-1489.
[10]	张云, 江勇, 郑靖, 庞春辉, 李琦. SDN跨层回环攻击的检测与防御[J]. 电子学报, 2019, 47(5): 1146-1151.
[11]	韩珍珍, 徐川, 王倩云, 王新恒, 赵国锋. 基于贝叶斯博弈的WLAN节能机制研究[J]. 电子学报, 2019, 47(10): 2083-2088.
[12]	王鹏超, 陈福才, 程国振, 陈扬, 谷允捷. 软件定义的L2/L3地址协同拟态伪装策略研究[J]. 电子学报, 2019, 47(10): 2032-2039.
[13]	黄建洋, 兰巨龙, 胡宇翔, 马腾. 一种基于分段路由的多路径流传输机制[J]. 电子学报, 2018, 46(6): 1488-1495.
[14]	胡涛, 张建辉, 邬江, 何为伟, 江逸茗, 赵伟. SDN中基于分布式决策的控制器负载均衡机制[J]. 电子学报, 2018, 46(10): 2316-2324.
[15]	杨洋, 杨家海, 秦董洪, 王于丁, 凌晓. DraLCD:一种新的数据中心流量工程方法[J]. 电子学报, 2017, 45(5): 1261-1267.