流数据驱动的多模态网络转发

吴剑松; 谭小彬; 袁莘智; 琚可欣; 骆汉光

doi:10.11959/j.issn.1000-0801.2025047

您当前的位置：

首页 >

文章列表页 >

流数据驱动的多模态网络转发

研究与开发 | 更新时间：2025-05-13

- 流数据驱动的多模态网络转发
- Streaming data-driven forwarding in polymorphic network
- 电信科学 2025年41卷第4期页码：95-106
- 作者机构：
  
  1.中国科学技术大学信息科学技术学院，安徽合肥 230026
  2.之江实验室，浙江杭州 311121
- 作者简介：
  
  [ "吴剑松（1997- ），男，中国科学技术大学信息科学技术学院硕士生，主要研究方向为计算机网络。" ]
  [ "谭小彬（1973- ），男，博士，中国科学技术大学信息科学技术学院副教授，主要研究方向为未来网络架构、智能网络和多媒体通信。" ]
  [ "袁莘智（1998- ），男，中国科学技术大学信息科学技术学院博士生，主要研究方向为计算机网络。" ]
  [ "琚可欣（2002- ），女，中国科学技术大学信息科学技术学院硕士生，主要研究方向为计算机网络。" ]
  [ "骆汉光（1988- ），男，博士，之江实验室副研究员，主要研究方向为新型网络、网络安全等。" ]
- 基金信息：
  
  国家重点研发计划项目;The National Key Research and Development Program of China(2022YFB2901401)
- DOI：10.11959/j.issn.1000-0801.2025047
  中图分类号： TP393
- 收稿日期：2024-11-14，
  
  修回日期：2025-01-05，
  
  纸质出版日期：2025-04-20
- 稿件说明：
移动端阅览
吴剑松,谭小彬,袁莘智等.流数据驱动的多模态网络转发[J].电信科学,2025,41(04):95-106.

WU Jiansong,TAN Xiaobin,YUAN Shenzhi,et al.Streaming data-driven forwarding in polymorphic network[J].Telecommunications Science,2025,41(04):95-106.
吴剑松,谭小彬,袁莘智等.流数据驱动的多模态网络转发[J].电信科学,2025,41(04):95-106. DOI： 10.11959/j.issn.1000-0801.2025047.

WU Jiansong,TAN Xiaobin,YUAN Shenzhi,et al.Streaming data-driven forwarding in polymorphic network[J].Telecommunications Science,2025,41(04):95-106. DOI： 10.11959/j.issn.1000-0801.2025047.

摘要

近年来通信技术发展迅速，传统传输控制协议/因特网互联协议（transmission control protocol/Internet protocol，TCP/IP）网络瘦腰结构的弊端凸显，新网络架构协议相继被提出，多种模态网络将长期共存。传统的网络转发方案难以适应多模态网络的动态变化特性，且存在时延高、成本高等问题。基于此，提出了一种流数据驱动的多模态网络转发方案和基于强化学习的多模态网络转发算法。针对多模态网络动态变化的特点，设计了基于可编程数据平面的多模态网络转发方案。该方案包含负责网络信息采集的带内网络遥测（in-band network telemetry，INT）采集模块、解析和过滤网络流的网络流处理模块、识别和分流多模态网络流的模态处理模块、提供网络功能的服务功能模块、负责转发的数据转发模块、用于数据缓存的缓存模块，以及生成和下发转发策略的节点控制模块。利用编写协议无关的包处理器（programming protocol-independent packet processor，P4）编程语言协议无关的特性，该方案可以支持多个网络模态的共生共存，且通过可编程交换机硬件可以有效提高转发效率。同时，针对可编程交换机通过表的方式进行数据更新的特点，提出了基于强化学习的多模态网络转发算法，并针对可编程硬件限制进行适配，根据多模态网络流的状态，为流选择合适的转发端口，以实现多模态网络的高效转发。在未来网络试验设施（China Environment for Network Innovation，CENI）中完成了转发算法的实现和测试，验证了所提算法的有效性。

Abstract

In recent years

rapid advancements in communication technology have highlighted the shortcomings of the traditional transmission control protocol/Internet protocol (TCP/IP) network architecture. New network architecture protocols have been proposed in response to the evolving landscape

where multiple modal networks are expected to coexist for an extended period. Traditional network forwarding solutions struggle to adapt to the dynamic nature of polymorphic network

leading to issues such as high latency and costs. Based on this

a flow data-driven polymorphic network forwarding scheme and a reinforcement learning-based algorithm tailored for polymorphic network were proposed. Addressing the dynamic nature of polymorphic network

a polymorphic network forwarding scheme based on programmable data planes was proposed. This scheme comprised the in-band network telemetry (INT) collection module responsible for network information gathering

the network flow processing module for parsing and filtering network flows

the modal processing module for identifying and routing polymorphic network flows

the service function module for providing network functions

the data forwarding module for routing data

the cache module for data storage

and the node control module for generating and deploying forwarding strategies. Leveraging the protocol-independent nature of the programming protocol-independent packet processors (P4) programming language

this scheme facilitated the coexistence of multiple network modes and enhances forwarding efficiency through programmable switch hardware. Furthermore

considering the data update mechanism of programmable switches using tables

a reinforcement learning-based algorithm for polymorphic network forwarding was proposed. Adapting to the constraints of programmable hardware

this algorithm selects appropriate forwarding ports based on the state of polymorphic network flows to achieve efficient forwarding. Finally

the implementation and testing of the forwarding algorithm was conducted in China Environment for Network Innovation (CENI)

validating the effectiveness of the algorithm.

关键词

Keywords

references

邬江兴 , 兰巨龙 , 程东年 . 新型网络体系结构 [M ] . 北京 : 人民邮电出版社 , 2014 .

WU J X , LAN J L , CHENG D N . New network architecture [M ] . Beijing : Posts and Telecom Press , 2014 .

胡宇翔 , 伊鹏 , 孙鹏浩 , 等 . 全维可定义的多模态网络体系研究 [J ] . 通信学报 , 2019 , 40 ( 8 ): 1 - 12 ..

HU Y X , YI P , SUN P H , et al . Research on the full-dimensional defined polymorphic smart network [J ] . Journal on Communications , 2019 , 40 ( 8 ): 1 - 12 .

HLGREN B , DANNEWITZ C , IMBRENDA C , et al . A survey of information-centric networking [J ] . IEEE Communications Magazine , 2012 , 50 ( 7 ): 26 - 36 .

DONG F , CHENG C Q , GUO S D . Design and research on GeoIP [C ] // Proceedings of 2010 14th International Conference on Computer Supported Cooperative Work in Design . Piscataway : IEEE Press , 2010 : 13 - 17 .

ZHANG H K , QUAN W , CHAO H C , et al . Smart identifier network: a collaborative architecture for the future Internet [J ] . IEEE Network , 2016 , 30 ( 3 ): 46 - 51 .

李军飞 , 胡宇翔 , 伊鹏 , 等 . 面向2035的多模态网络技术发展路线图 [J ] . 中国工程科学 , 2020 , 22 ( 3 ): 141 - 147 .

LI J F , HU Y X , YI P , et al . Development roadmap of polymorphic intelligence network technology toward 2035 [J ] . Strategic Study of CAE , 2020 , 22 ( 3 ): 141 - 147 .

MCCAULEY J , HARCHOL Y , PANDA A , et al . Enabling a permanent revolution in Internet architecture [C ] // Proceedings of the ACM Special Interest Group on Data Communication . Piscataway : IEEE Press , 2019 : 1 - 14 .

DASGUPTA S , SHANNIGRAHI S , ZINK M . A hybrid NDN-IP architecture for live video streaming: from host-based to content-based delivery to improve QoE [J ] . International Journal of Semantic Computing , 2022 ( 2 ): 163 - 187 .

ZHANG X , GONG L , XUN Y , et al . Centaur: a evolutionary design of hybrid NDN/IP transport architecture for streaming application [C ] // 2016 IEEE 7th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON) . Piscataway : IEEE Press , 2016 : 1 - 7 .

CAROFIGLIO G , MUSCARIELLO L , JORDAN AUGÉ , et al . Enabling ICN in the Internet protocol: analysis and evaluation of the hybrid-ICN architecture [C ] // Proceedings of the 6th ACM Conference on Information-Centric Networking . Piscataway : IEEE Press , 2019 : 55 – 66 .

REFAEI T , MA J , HA S , et al . Integrating IP and NDN through an extensible IP-NDN gateway [C ] // Proceedings of the 4th ACM Conference on Information-Centric Networking . Piscataway : IEEE Press , 2017 : 224 – 225 .

OLIVER M , ROBERTO B , GÁBOR R , and STEFAN S . The Programmable data plane: abstractions, architectures, algorithms, and applications [J ] . ACM Computing Surveys , 2021 , 54 ( 4 ): 1 - 36 .

BRADATSCH L , KARGL F . Integration of security service functions into network-level access control [J ] . IEEE Access , 2024 ( 12 ): 197783 - 197815 .

WU Q Q , LIU Q , JIA Z Q , et al . P4SQA: a P4 switch-based QoS assurance mechanism for SDN [J ] . IEEE Transactions on Network and Service Management , 2023 , 20 ( 4 ): 4875 - 4886 .

ZHENG C G , HONG X P , DING D M , et al . In-network machine learning using programmable network devices: a survey [J ] . IEEE Communications Surveys & Tutorials , 2024 , 26 ( 2 ): 1171 - 1200 .

BOSSHART P , DALY D , GIBB G , et al . P4: programming protocol-independent packet processors [J ] . ACM SIGCOMM Computer Communication Review , 2014 , 44 ( 3 ): 87 - 95 .

ESTAN C , VARGHESE G . New directions in traffic measurement and accounting: focusing on the elephants, ignoring the mice [J ] . ACM Transactions on Computer Systems , 2003 , 21 ( 3 ): 270 - 313 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

一种面向计算的网络模态——算力网络模态探讨

基于意图的多模态网业协同架构研究

面向智能制造的多模态网络应用技术研究

面向多模态网络的隔离转发技术研究

全维可重构的多模态网络交换芯片架构设计