自智算力网络：架构、技术与展望

doi:10.11959/j.issn.2096-3750.2023.00350

物联网学报 ›› 2023, Vol. 7 ›› Issue (4): 1-12.doi: 10.11959/j.issn.2096-3750.2023.00350

• 理论与技术 •

自智算力网络：架构、技术与展望

周晓茂¹, 贾庆民¹, 胡玉姣¹, 郭凯¹, 马千飘¹, 刘辉¹, 谢人超¹^,²

¹ 网络通信与安全紫金山实验室，江苏南京 211111
² 北京邮电大学，北京 100876

修回日期:2023-06-16 出版日期:2023-12-01 发布日期:2023-12-01
作者简介:周晓茂（1993- ），男，博士，网络通信与安全紫金山实验室研究员，主要研究方向为边缘计算、人工智能、算力网络、工业互联网等
贾庆民（1990- ），男，博士，网络通信与安全紫金山实验室研究员，主要研究方向为算力网络、工业互联网、确定性网络、人工智能等
胡玉姣（1993- ），女，博士，网络通信与安全紫金山实验室研究员，主要研究方向为信息物理融合系统、算力网络、人工智能、云边协同等
郭凯（1991- ），男，博士，网络通信与安全紫金山实验室研究员，主要研究方向为算力网络、工业互联网、确定性网络、人工智能等
马千飘（1993- ），男，博士，网络通信与安全紫金山实验室研究员，主要研究方向为边缘计算、联邦学习、分布式机器学习等
刘辉（1983- ），男，博士，网络通信与安全紫金山实验室高级工程师，主要研究方向为边缘智能体系架构、算力网络、云计算等
谢人超（1984- ），男，博士，北京邮电大学教授、博士生导师，主要研究方向为未来网络体系架构设计、信息中心网络、移动网络内容分发等
基金资助:
国家自然科学基金资助项目(92267301);紫金山英才-江宁百家湖计划项目(74072203-3)

Autonomous computing and network convergence:architecture, technologies, and prospects

Xiaomao ZHOU¹, Qingmin JIA¹, Yujiao HU¹, Kai GUO¹, Qianpiao MA¹, Hui LIU¹, Renchao XIE¹^,²

¹ Purple Mountain Laboratories, Nanjing 211111, China
² Beijing University of Posts and Telecommunications, Beijing 100876, China

Revised:2023-06-16 Online:2023-12-01 Published:2023-12-01
Supported by:
The National Natural Science Foundation of China(92267301);Purple Mountain Talents-Jiangning Baijia Lake Plan Program(74072203-3)

摘要/Abstract

摘要：

针对算力网络（CNC, computing and network convergence）中的新型业务场景和网络高度智能化的需求，阐述了自智算力网络（Auto-CNC, autonomous CNC）的技术发展理念，即通过将智能引入算网全生命流程的方式实现资源一体化、流程自动化、系统智能化，简要分析了算力网络的研究现状和现存问题，总结了自智算力网络应具备意图驱动的算网融合、算网闭环自治、网-算-智协同自适演进的关键特征，设计了自智算力网络的参考架构，分析了其关键技术，介绍了相关的初阶探索工作，并对自智算力网络的未来发展进行了展望。

关键词: 自智算力网络, 意图驱动, 协同自适演进, 自智运行

Abstract:

In view of the new service scenarios and the demand for high intelligence network in computing and network convergence (CNC), the concept of autonomous CNC (Auto-CNC) is elaborated, where intelligence was introduced into all the aspects of CNC, including resource integration, process automation, and system intelligence.The current research directions and remaining challenges of CNC were introduced, and three key features, i.e., intent-driven computing network, the autonomous system operation and the adaptive co-evolution of communication, computing intelligence, were summarized from the proposed Auto-CNC.Meanwhile, the reference architecture and key technologies of Auto-CNC were described, which were followed by several preliminary exploration cases.Finally, future research trends and technical advice were discussed and recommended.

Key words: autonomous CNC, intent-driven, adaptive co-evolution, autonomous operation

中图分类号:

TN914

周晓茂, 贾庆民, 胡玉姣, 郭凯, 马千飘, 刘辉, 谢人超. 自智算力网络：架构、技术与展望[J]. 物联网学报, 2023, 7(4): 1-12.

Xiaomao ZHOU, Qingmin JIA, Yujiao HU, Kai GUO, Qianpiao MA, Hui LIU, Renchao XIE. Autonomous computing and network convergence:architecture, technologies, and prospects[J]. Chinese Journal on Internet of Things, 2023, 7(4): 1-12.

图/表 6

表1

表2

图1

图2

图3

图4

参考文献 45

[1]	贾庆民, 丁瑞, 刘辉 ,等. 算力网络研究进展综述[J]. 网络与信息安全学报, 2021,7(5): 1-12.
	JIA Q M , DING R , LIU H ,et al. Survey on research progress for compute first networking[J]. Chinese Journal of Network and Information Security, 2021,7(5): 1-12.
[2]	KROEL M , MASTORAKIS S , ORAN D ,et al. Compute first networking:distributed computing meets ICN[C]// The 6th ACM Conference on Information-Centric Networking. New York:ACM Press, 2019: 67-77.
[3]	NOUR B , MASTORAKIS S , MTIBAA A . Compute-less networking:perspectives,challenges,and opportunities[J]. IEEE Network, 2020,34(6): 259-265.
[4]	曹畅, 唐雄燕 . 算力网络关键技术及发展挑战分析[J]. 信息通信技术与政策, 2021,47(3): 6-11.
	CAO C , TANG X Y . Analysis on key technologies and development challenges of computing network[J]. Information and Communications Technology and Policy, 2021,47(3): 6-11.
[5]	中国联通网络技术研究院. 中国联通算力网络白皮书[R]. 2019.
	China Unicom Network Technology Research Institute. White paper on China Unicom computing power network[R]. 2019.
[6]	中国移动研究院. 算力感知网络技术白皮书[R]. 2019.
	China Mobile Research Institute. White paper on computing-aware network technology[R]. 2019.
[7]	中兴通信. 自智网络白皮书[R]. 2022.
	ZTE. White paper on autonomous network[R]. 2022.
[8]	IETF. Computing in the network research group (coinrg)[EB]. 2019.
[9]	IETF. Routing area working group (rtgwg)[EB]. 2019.
[10]	IETF. Computing in network research group,requirement of computing in network draft-liu-coinrg-requirement-02[R]. 2020.
[11]	DI Z , CAO Y , QIU C ,et al. New computing power network architecture and application case analysis[J]. Journal of Computer Applications, 2022,42(6): 1656-1662.
[12]	HUANG G , SHI W , TAN B . Computing power network resources based on SRv6 and its service arrangement and scheduling[J]. ZTE Technology Journal, 2021,27(3): 23-28.
[13]	TANG X Y , CAO C , WANG Y X ,et al. Computing power network:the architecture of convergence of computing and networking towards 6G requirement[J]. China Communications, 2021,18(2): 175-185.
[14]	ZHANG S , CAO C , TANG X . Computing power network technology architecture based on SRv6[J]. ZTE Technology Journal, 2022,28(1): 11-15.
[15]	梁冰, 纪雯 . 基于次模优化的边云协同多用户计算任务迁移方法[J]. 通信学报, 2020,41(10): 25-36.
	LIANG B , JI W . Multiuser computation offloading for edge-cloud collaboration using submodular optimization[J]. Journal on Communications, 2020,41(10): 25-36.
[16]	胡玉姣, 贾庆民, 孙庆爽 ,等. 融智算力网络及其功能架构[J]. 计算机科学, 2022,49(9): 249-259.
	HU Y J , JIA Q M , SUN Q S ,et al. Functional architecture to intelligent computing power network[J]. Computer Science, 2022,49(9): 249-259.
[17]	通信世界. 中国移动牵头 3GPP 成立自治网络分级标准项目[EB]. 2020.
	Communication World Network. China Mobile leads 3GPP to establish autonomous network classification standard project[EB]. 2020.
[18]	ITU-T. Focus group on machine learning for future networks including 5G[S]. 2020.
[19]	ITU-T. Framework for evaluating intelligence levels of future networks including IMT-2020:ITU-T Y.3173[S]. 2020.
[20]	XU X X , SUN J , WANG C Y ,et al. A novel hybrid CNN-LSTM compensation model against DoS attacks in power system state estimation[J]. Neural Processing Letters, 2022,54(3): 1597-1621.
[21]	CHEN W X , XU H W , LI Z Y ,et al. Unsupervised anomaly detection for intricate KPIs via adversarial training of VAE[C]// Proceedings of IEEE INFOCOM 2019 - IEEE Conference on Computer Communications. Piscataway:IEEE Press, 2019: 1891-1899.
[22]	ZHANG R , MAO Y , ZHAO W . Knowledge graphs completion via probabilistic reasoning[J]. Information Sciences, 2020(521): 144-159.
[23]	GUO S Y , DAI Y , XU S Y ,et al. Trusted cloud-edge network resource management:DRL-driven service function chain orchestration for IoT[J]. IEEE Internet of Things Journal, 2020,7(7): 6010-6022.
[24]	WEI Y , PENG M , LIU Y . Intent-based networks for 6G:insights and challenges[J]. Digital Communications and Networks, 2020,6(3): 270-280.
[25]	Gartner. Innovation insight:intent-based networking systems[EB]. 2017.
[26]	QIAO F F , ZHU X J . Domain intelligent Q＆A user intention recognition based on keyword separation[C]// Proceedings of 2020 International Conference on Culture-oriented Science ＆ Technology (ICCST). Piscataway:IEEE Press, 2020: 224-229.
[27]	PéREZ-SOLER S , GUERRA E , DE LARA J . Model-driven chatbot development[M]. Conceptual Modeling. Cham: Springer International Publishing, 2020: 207-222.
[28]	WU S F . Internet public information an text data mining and intelligence influence analysis for user intent understanding[J]. Journal of Intelligent ＆ Fuzzy Systems, 2020,38(1): 487-494.
[29]	WANG Y , FORBES R , CAVIGIOLI C ,et al. Network management and orchestration using artificial intelligence:overview of ETSI ENI[J]. IEEE Communications Standards Magazine, 2018,2(4): 58-65.
[30]	GU Z Y , SHE C Y , HARDJAWANA W ,et al. Knowledge-assisted deep reinforcement learning in 5G scheduler design:from theoretical framework to implementation[J]. IEEE Journal on Selected Areas in Communications, 2021,39(7): 2014-2028.
[31]	CHEN X , JIA S , XIANG Y . A review:knowledge reasoning over knowledge graph[J]. Expert Systems With Applications, 2020(141): 112948.
[32]	邓灵莉, 顾宁伦, 袁向阳 ,等. 网络智能化标准、开源与产业研究[J]. 电信科学, 2021,37(10): 12-21.
	DENG L L , GU N L , YUAN X Y ,et al. Network intelligence standards,open source and industry research[J]. Telecommunications Science, 2021,37(10): 12-21.
[33]	SARKER I H . Deep learning:a comprehensive overview on techniques,taxonomy,applications and research directions[J]. SN Computer Science, 2021,2(6): 420.
[34]	段晓东, 姚惠娟, 付月霞 ,等. 面向算网一体化演进的算力网络技术[J]. 电信科学, 2021,37(10): 76-85.
	DUAN X D , YAO H J , FU Y X ,et al. Computing force network technologies for computing and network integration evolution[J]. Telecommunications Science, 2021,37(10): 76-85.
[35]	THAKKAR H K , SAHOO P K , VEERAVALLI B . RENDA:resource and network aware data placement algorithm for periodic workloads in cloud[J]. IEEE Transactions on Parallel and Distributed Systems, 2021,32(12): 2906-2920.
[36]	HAGHNEGAHDAR L , JOSHI S S , DAHOTRE N B . From IoT-based cloud manufacturing approach to intelligent additive manufacturing:industrial internet of things—an overview[J]. The International Journal of Advanced Manufacturing Technology, 2022,119(3): 1461-1478.
[37]	SUN Z J , YANG H , LI C ,et al. Cloud-edge collaboration in industrial internet of things:a joint offloading scheme based on resource prediction[J]. IEEE Internet of Things Journal, 2022,9(18): 17014-17025.
[38]	DE S , DEY S , BHATIA S ,et al. An introduction to data mining in social networks[J]. Advanced Data Mining Tools and Methods for Social Computing, 2022: 1-25.
[39]	VIJAYABASKAR KUMANAN T . Link quality and energy-aware metric-based routing strategy in WSNS[M]. Intelligent Computing in Engineering. Singapore: Springer Singapore, 2020: 533-539.
[40]	SAMAL T , KABAT M R . Energy-efficient time-sharing multichannel MAC protocol for wireless body area networks[J]. Arabian Journal for Science and Engineering, 2022,47(2): 1791-1804.
[41]	LI C , TANG J , TANG H , LUO Y . Collaborative cache allocation and task scheduling for data-intensive applications in edge computing environment[J]. Future Generation Computer Systems, 2019(95): 249-264.
[42]	刘泽宁, 李凯, 吴连涛 ,等. 多层次算力网络中代价感知任务调度算法[J]. 计算机研究与发展, 2020,57(9): 1810-1822.
	LIU Z N , LI K , WU L T ,et al. CATS:cost aware task scheduling in multi-tier computing networks[J]. Journal of Computer Research and Development, 2020,57(9): 1810-1822.
[43]	王敬宇, 周铖, 张蕾 ,等. 知识定义的意图网络自治[J]. 电信科学, 2021,37(9): 1-13.
	WANG J Y , ZHOU C , ZHANG L ,et al. Knowledge-defined intent-based network autonomy[J]. Telecommunications Science, 2021,37(9): 1-13.
[44]	ZHOU X , JIA Q , XIE R . NestFL:efficient federated learning through progressive model pruning in heterogeneous edge computing[C]// The 28th Annual International Conference on Mobile Computing and Networking.[S.l.:s.n.], 2022: 817-819.
[45]	SHUAI L , WANG K , ZHANG L ,et al. Global-local integration for GNN-based anomalous device state detection in industrial control systems[J]. Expert Systems With Applications, 2022(209): 118345.

研究方向	维度	内容
行业标准	功能架构、网内需求、传输协议、隐私安全、应用案例	算力网络框架与架构标准、算力网络总体技术要求、算力网络的信令需求、泛在算力调度管理技术和算力度量、算力网络控制技术要求、算力网络标识解析技术要求、算力网络交易平台技术研究
学术研究	系统架构、流程机制、关键技术	基于多维资源融合的虚拟化架构、基于云网边协同的算网融合策略、基于 SRv6 的算网一体编排与路由、算力感知路由分配策略、代价感知的任务调度算法
应用场景	生活场景、行业场景、社会场景	VR 互动、云游戏、新媒体直播、智慧医疗、车联网、工业物联网、智能科学模拟、平台型算力共享

研究方向	维度	内容
行业标准	体系架构、关键技术、智能等级、评估方法	自智网络体系架构、自智网络意图管理技术研究、通信网络智能化分级及评估方法、人工智能在电信网络演进中的应用研究、传输网智能化运维研究
学术研究	智能感知、智能分析、自主决策、自动执行	基于时空LSTM的网络态势感知、基于VAE的网络异常检测、基于知识图谱的网络故障传播推理、基于 DRL 的资源按需分配、基于生成对抗网络（GAN, generative adversarial network）的任务策略生成
应用场景	智能网元类、运维智能类	端到端的全流程自动化、意图驱动的按需服务、以用户为中心的闭环运维

自智算力网络：架构、技术与展望

Autonomous computing and network convergence:architecture, technologies, and prospects

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 6

参考文献 45

相关文章 2

Metrics

推荐阅读 0

[1]	周洋程,闫实,彭木根. 意图驱动的6G无线接入网络[J]. 物联网学报, 2020, 4(1): 72-79.
[2]	张佳鸣,杨春刚,庞磊,李建东. 意图物联网[J]. 物联网学报, 2019, 3(3): 5-10.