一种基于深度学习的SRv6网络流量调度优化算法

doi:10.3969/j.issn.1671-1122.2024.02.010

信息网络安全 ›› 2024, Vol. 24 ›› Issue (2): 272-281.doi: 10.3969/j.issn.1671-1122.2024.02.010

一种基于深度学习的SRv6网络流量调度优化算法

赵鹏程¹, 于俊清¹^,²(), 李冬²

1.华中科技大学网络空间安全学院，武汉 430074
2.华中科技大学网络与计算中心，武汉 430074

收稿日期:2023-10-18 出版日期:2024-02-10 发布日期:2024-03-06
通讯作者: 于俊清 E-mail:yjqing@hust.edu.cn
作者简介:赵鹏程（1996—），男，湖北，硕士研究生，主要研究方向为网络安全、软件定义网络|于俊清（1975—），男，内蒙古，教授，博士，CCF会员，主要研究方向为数字媒体处理与检索、网络安全、多核计算与流编译|李冬（1979—），男，湖北，高级工程师，博士，主要研究方向为计算机网络、软件定义网络、网络安全
基金资助:
国家重点研发计划(2020YFB1805601);中国高校产学研创新基金(2021FNA02005)

An Optimal Algorithm for Traffic Scheduling in SRv6 Network Based on Deep Learning

ZHAO Pengcheng¹, YU Junqing¹^,²(), LI Dong²

1. School of Cyber Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, China
2. Network and Computation Center, Huazhong University of Science and Technology, Wuhan 430074, China

Received:2023-10-18 Online:2024-02-10 Published:2024-03-06
Contact: YU Junqing E-mail:yjqing@hust.edu.cn

摘要/Abstract

摘要：

目前SRv6网络中的流量调度方法主要是基于固定或启发式规则的方法，缺乏灵活调度整体网络流量的能力，难以适应动态的网络环境变化。针对SRv6网络缺乏关键流识别能力的问题，文章提出一种基于深度强化学习的关键流识别算法，建立适应网络动态变化的关键流学习模型，在不同的流量矩阵中识别出对网络性能影响最大的关键流集合。针对SRv6网络流量调度问题，文章提出一种基于关键流的流量调度优化算法，采用线性规划求解出每一条关键流的最优显式路径，并采用不同的路由方式对普通流和关键流进行负载均衡。实验结果表明，该算法可显著提升SRv6网络流量负载均衡能力，降低网络端到端传输延迟。

关键词: 深度学习, 软件定义网络, 分段路由, 流量调度

Abstract:

Current traffic scheduling methods in SRv6 network are mainly based on fixed or heuristic rules, which lack of the ability to schedule overall network traffic flexibly and are difficult to adapt to dynamic network environment changes. To address the deficiency in key flow identification within SRv6 network, the article introduced a key flow identification algorithm based on deep reinforcement learning. This approach established a key flow learning model adapted to the dynamic changes of the network, identifying sets of key flows that significantly impact network performance across various traffic matrices. In response to the challenges of traffic scheduling in SRv6 network, the article developed an optimization algorithm for traffic scheduling, rooted in key flow analysis. This algorithm employed linear programming to determine the optimal explicit path for each key flow and utilized different routing methods for ordinary flows and key flows, effectively enhancing network performance. The experimental results demonstrate that the proposed traffic scheduling algorithm leads to a significant improvement in network load balancing and a substantial reduction in network end-to-end transmission delay.

Key words: deep learning, SDN, segment routing, traffic scheduling

中图分类号:

TP309

赵鹏程, 于俊清, 李冬. 一种基于深度学习的SRv6网络流量调度优化算法[J]. 信息网络安全, 2024, 24(2): 272-281.

ZHAO Pengcheng, YU Junqing, LI Dong. An Optimal Algorithm for Traffic Scheduling in SRv6 Network Based on Deep Learning[J]. Netinfo Security, 2024, 24(2): 272-281.

图/表 13

表1

图1

图2

图3

图4

表2

图5

图6

图7

图8

图9

图10

图11

参考文献 22

[1]	KREUTZ D, RAMOS F M V, VERISSIMO P E, et al. Software-Defined Networking: A Comprehensive Survey[J]. Proceedings of the IEEE, 2014, 103(1): 14-76. doi: 10.1109/JPROC.2014.2371999 URL
[2]	VENTRE P L, SALSANO S, POLVERINI M, et al. Segment Routing: A Comprehensive Survey of Research Activities, Standardization Efforts, and Implementation Results[J]. IEEE Communications Surveys & Tutorials, 2020, 23(1): 182-221.
[3]	CHENG Weiqiang, LIU Yisong, JIANG Wenying, et al. Research and Application of G-SRv6 Head Compressions Optimization Technology[J]. Telecommunications Science, 2020, 36(8): 22-27. doi: 10.11959/j.issn.1000-0801.2020254
	程伟强, 刘毅松, 姜文颖, 等. G-SRv6 头压缩优化技术研究与应用[J]. 电信科学, 2020, 36(8):22-27. doi: 10.11959/j.issn.1000-0801.2020254
[4]	WU Wei, ZHANG Wenqiang, YANG Guangming, et al. SRv6 +EVPN Technology Research and Scale Deployment of 5G Bearer Network[J]. Telecommunications Science, 2020, 36(8): 43-52. doi: 10.11959/j.issn.1000-0801.2020253
	吴伟, 张文强, 杨广铭, 等. 5G 承载网的“SRv6+ EVPN”技术研究与规模部署[J]. 电信科学, 2020, 36(8):43-52. doi: 10.11959/j.issn.1000-0801.2020253
[5]	CHEN Kefan, ZHAO Shanghou, LV Na, et al. Segment Routing Based Traffic Scheduling for the Software-Defined Airborne Backbone Network[J]. IEEE Access, 2019, 7: 106162-106178. doi: 10.1109/ACCESS.2019.2930229
[6]	STAMPA G, ARIAS M, SÁNCHEZ-CHARLES D, et al. A Deep-Reinforcement Learning Approach for Software-Defined Networking Routing Optimization[EB/OL]. (2017-09-20)[2023-08-10]. https://arxiv.org/abs/1709.07080.pdf.
[7]	CASAS N. Deep Deterministic Policy Gradient for Urban Traffic Light Control[EB/OL]. (2017-03-27)[2023-08-10]. https://arxiv.org/pdf/1703.09035.pdf.
[8]	TENG Yuantao, XIA Zhengyou. A Traffic Engineering Technology Based on Segment Routing in SDN[C]// IEEE. 2020 16th International Conference on Mobility, Sensing and Networking(MSN). New York:IEEE, 2020: 636-641.
[9]	LI Cheng, MAO Jianwen, PENG Shuping, et al. Application-Aware G-SRv6 Network Enabling 5G Services[C]// IEEE. 2021 IEEE Conference on Computer Communications Workshops(INFOCOM WKSHPS). New York:IEEE, 2021: 1-2.
[10]	REN Bangbang, GUO Deke, YUAN Yali, et al. Optimal Deployment of SRv6 to Enable Network Interconnection Service[J]. IEEE/ACM Transactions on Networking, 2021, 30(1): 120-133. doi: 10.1109/TNET.2021.3105959 URL
[11]	ANAND R, AGGARWAL D, KUMAR V. A Comparative Analysis of Optimization Solvers[J]. Journal of Statistics and Management Systems, 2017, 20(4): 623-635. doi: 10.1080/09720510.2017.1395182 URL
[12]	MAO Hongzi, ALIZADEH M, MENACHE I, et al. Resource Management with Deep Reinforcement Learning[C]// ACM. 15th ACM Workshop on Hot Topics in Networks. New York: ACM, 2016: 50-56.
[13]	WILLIAMS R J. Simple Statistical Gradient-Following Algorithms for Connectionist Reinforcement Learning[J]. Reinforcement Learning, 1992, 8: 5-32.
[14]	JIANG J R, HUANG H W, LIAO J H, et al. Extending Dijkstra’s Shortest Path Algorithm for Software Defined Networking[C]// IEEE. The 16th Asia-Pacific Network Operations and Management Symposium. New York: IEEE, 2014: 1-4.
[15]	CIANFRANI A, LISTANTI M, POLVERIN M. Incremental Deployment of Segment Routing into an ISP Network: A Traffic Engineering Perspective[J]. IEEE/ACM Transactions on Networking, 2017, 25(5): 3146-3160. doi: 10.1109/TNET.2017.2731419 URL
[16]	SPRING N, MAHAJAN R, WETHERALL D. Measuring ISP Topologies with Rocketfuel[J]. ACM SIGCOMM Computer Communication Review, 2002, 32(4): 133-145. doi: 10.1145/964725.633039 URL
[17]	KODIALAM M, LAKSHMAN T V, ORLIN J, et al. Oblivious Routing of Highly Variable Traffic in Service Overlays and IP Backbones[J]. IEEE/ACM Transactions on Networking, 2008, 17(2): 459-472. doi: 10.1109/TNET.2008.927257 URL
[18]	TUNE P, ROUGHAN M. Spatiotemporal Traffic Matrix Synthesis[C]// ACM. 2015 ACM Conference on Special Interest Group on Data Communication. New York: ACM, 2015: 579-592.
[19]	ZHANG Junjie, XI Kang, ZHANG Liren, et al. Optimizing Network Performance Using Weighted Multipath Routing[C]// IEEE. 21st International Conference on Computer Communications and Networks(ICCCN). New York:IEEE, 2012: 1-7.
[20]	ZHANG Hailong, GUO Xiao, YAN Jinyao, et al. SDN-Based ECMP Algorithm for Data Center Networks[C]// IEEE. 2014 IEEE Computers, Communications and IT Applications Conference. New York: IEEE, 2014: 13-18.
[21]	LIU Jing, LI Jie, SHOU Guochu, et al. SDN Based Load Balancing Mechanism for Elephant Flow in Data Center Networks[C]// IEEE. 2014 International Symposium on Wireless Personal Multimedia Communications(WPMC). New York:IEEE, 2014: 486-490.
[22]	YAO Guohao, WU Muqing, XU Yang. A Data Center Load Balancing Algorithm Based on Artificial Bee Colony Algorithm[C]// IEEE. 2020 IEEE 6th International Conference on Computer and Communications(ICCC). New York:IEEE, 2020: 1770-1775.

符号	含义
${{e}_{i,j}}$	属于集合E中的链路$\left( i,j \right)$
${{c}_{i,j}}$	链路${{e}_{i,j}}$的链路容量
${{l}_{i,j}}$	链路${{e}_{i,j}}$上的流量负载
${{\mathbf{D}}^{s,d}}$	源节点s到目的节点d的流量需求
$\sigma _{i,j}^{s,d}$	源节点s到目的节点d在链路${{e}_{i,j}}$上的流量需求百分比

网络拓扑	节点数/个	直连链路数/条
Abilene	12	30
Atlanta	15	44
Ebone	23	72

一种基于深度学习的SRv6网络流量调度优化算法

An Optimal Algorithm for Traffic Scheduling in SRv6 Network Based on Deep Learning

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 22

相关文章 15

编辑推荐

Metrics

本文评价

[1]	冯光升, 蒋舜鹏, 胡先浪, 马明宇. 面向物联网的入侵检测技术研究新进展[J]. 信息网络安全, 2024, 24(2): 167-178.
[2]	金志刚, 丁禹, 武晓栋. 融合梯度差分的双边校正联邦入侵检测算法[J]. 信息网络安全, 2024, 24(2): 293-302.
[3]	薛羽, 张逸轩. 深层神经网络架构搜索综述[J]. 信息网络安全, 2023, 23(9): 58-74.
[4]	刘宇啸, 陈伟, 张天月, 吴礼发. 基于稀疏自动编码器的可解释性异常流量检测[J]. 信息网络安全, 2023, 23(7): 74-85.
[5]	蒋英肇, 陈雷, 闫巧. 基于双通道特征融合的分布式拒绝服务攻击检测算法[J]. 信息网络安全, 2023, 23(7): 86-97.
[6]	赵彩丹, 陈璟乾, 吴志强. 基于多通道联合学习的自动调制识别网络[J]. 信息网络安全, 2023, 23(4): 20-29.
[7]	谭柳燕, 阮树骅, 杨敏, 陈兴蜀. 基于深度学习的教育数据分类方法[J]. 信息网络安全, 2023, 23(3): 96-102.
[8]	徐占洋, 程洛飞, 程建春, 许小龙. 一种使用Bi-ADMM优化深度学习模型的方案[J]. 信息网络安全, 2023, 23(2): 54-63.
[9]	陈得鹏, 刘肖, 崔杰, 仲红. 一种基于双阈值函数的成员推理攻击方法[J]. 信息网络安全, 2023, 23(2): 64-75.
[10]	文伟平, 朱一帆, 吕子晗, 刘成杰. 针对品牌的网络钓鱼扩线与检测方案[J]. 信息网络安全, 2023, 23(12): 1-9.
[11]	刘军, 武志超, 吴建, 谭振华. 一种融合图像空间特征注意力机制的恶意代码识别模型[J]. 信息网络安全, 2023, 23(12): 29-37.
[12]	廖丽云, 张伯雷, 吴礼发. 基于代价敏感学习的物联网异常检测模型[J]. 信息网络安全, 2023, 23(11): 94-103.
[13]	王智, 张浩, 顾建军. SDN网络中基于联合熵与多重聚类的DDoS攻击检测[J]. 信息网络安全, 2023, 23(10): 1-7.
[14]	贾凡, 康舒雅, 江为强, 王光涛. 基于NLP及特征融合的漏洞相似性算法评估[J]. 信息网络安全, 2023, 23(1): 18-27.
[15]	高博, 陈琳, 严迎建. 基于CNN-MGU的侧信道攻击研究[J]. 信息网络安全, 2022, 22(8): 55-63.