可编程网络数据平面技术进展

doi:10.11959/j.issn.1000-0801.2022131

电信科学 ›› 2022, Vol. 38 ›› Issue (6): 42-50.doi: 10.11959/j.issn.1000-0801.2022131

可编程网络数据平面技术进展

张昕怡¹, 潘恒², 谢高岗¹^,³

¹ 中国科学院计算机网络信息中心，北京 100190
² 中国科学院计算技术研究所，北京 100190
³ 中国科学院大学，北京 100049

修回日期:2022-06-07 出版日期:2022-06-20 发布日期:2022-06-01
作者简介:张昕怡（1992- ），女，博士，中国科学院计算机网络信息中心在站博士后，主要研究方向为计算机网络系统结构、高性能网络数据平面
潘恒（1990- ），男，博士，中国科学院计算技术研究所副研究员，主要研究方向为可编程网络与系统
谢高岗（1974- ），男，博士，中国科学院计算机网络信息中心研究员，中国科学院大学教授、博士生导师，主要研究方向为计算机网络体系结构与系统
基金资助:
国家重点研发计划项目(2019YFB1802800)

Progress in programmable network data plane

Xinyi ZHANG¹, Heng PAN², Gaogang XIE¹^,³

¹ Computer Network Information Center, Chinese Academy of Sciences, Beijing 100190, China
² Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China
³ University of Chinese Academy of Sciences, Beijing 100049, China

Revised:2022-06-07 Online:2022-06-20 Published:2022-06-01
Supported by:
The National Key Research and Development Program of China(2019YFB1802800)

摘要/Abstract

摘要：

网络数据平面执行数据包处理转发，是网络性能的决定性因素之一。大带宽、低时延、可持续演进的网络基础设施需要构建高效可编程的网络数据平面。首先，介绍数据包处理转发模型，并以此为基础概述网络数据平面在性能与可编程性面临的关键挑战。然后，从数据包查找算法理论与软/硬件协同实现机制出发，详细论述其基本思路及关键核心技术进展以应对上述关键挑战。最后，探讨高效可编程数据平面的未来发展趋势与技术演进路线。

关键词: 网络可编程数据平面, 软/硬件协同设计, 数据包处理, 网络内计算

Abstract:

Network data plane performs packet processing and forwarding, which is one of the core factors determining the network performance.High bandwidth, low latency and rapid evolution of the network infrastructure require building an efficient and programmable network data plane.The packet processing and forwarding model was firstly introduced, and on this basis, the challenges of the network data plane in the performance and programmability were outlined.Then the basic ideas and core technology progress were elaborated to deal with the above challenges by analyzing the theory of packet processing algorithms and hardware-software codesign implementations.Finally, the trend and technical route of constructing an efficient programmable data plane were discussed.

Key words: programmable network data plane, hardware-software codesign, packet processing, in-network computing

中图分类号:

TP393

张昕怡, 潘恒, 谢高岗. 可编程网络数据平面技术进展[J]. 电信科学, 2022, 38(6): 42-50.

Xinyi ZHANG, Heng PAN, Gaogang XIE. Progress in programmable network data plane[J]. Telecommunications Science, 2022, 38(6): 42-50.

图/表 1

参考文献 48

[1]	AGIWAL M , ROY A , SAXENA N . Next generation 5G wireless networks:a comprehensive survey[J]. IEEE Communications Surveys ＆ Tutorials, 2016,18(3): 1617-1655.
[2]	NIGHTINGALE J , SALVA-GARCIA P , CALERO J M A ,et al. 5G-QoE:QoE modelling for ultra-HD video streaming in 5G networks[J]. IEEE Transactions on Broadcasting, 2018,64(2): 621-634.
[3]	SIRIWARDHANA Y , PORAMBAGE P , LIYANAGE M ,et al. A survey on mobile augmented reality with 5G mobile edge computing:architectures,applications,and technical aspects[J]. IEEE Communications Surveys ＆ Tutorials, 2021,23(2): 1160-1192.
[4]	MCKEOWN N , ANDERSON T , BALAKRISHNAN H ,et al. OpenFlow[J]. ACM SIGCOMM Computer Communication Review, 2008,38(2): 69-74.
[5]	HONG C Y , MANDAL S , AL-FARES M ,et al. B4 and after:managing hierarchy,partitioning,and asymmetry for availability and scale in Google's software-defined WAN[C]// Proceedings of SIGCOMM '18:Proceedings of the 2018 Conference of the ACM Special Interest Group on Data Communication.[S.l:s.n.], 2018: 74-87.
[6]	HAN B , GOPALAKRISHNAN V , JI L S ,et al. Network function virtualization:challenges and opportunities for innovations[J]. IEEE Communications Magazine, 2015,53(2): 90-97.
[7]	HYAFIL L , RIVEST R L . Constructing optimal binary decision trees is NP-complete[J]. Information Processing Letters, 1976,5(1): 15-17.
[8]	DPDK. Developer quick start guide[EB]. 2022.
[9]	RIZZO L , . Netmap:a novel framework for fast packet I/O[C]// Proceedings of 21st USENIX Security Symposium.[S.l.:s.n.] 2012: 101-112.
[10]	Ntop. PF_RING:high-speed packet capture,filtering and analysis[EB]. 2022.
[11]	Intel. 探索英特尔? 智能结构处理器的强大功能[EB]. 2022.
	INTEI. Discover the power of the Intel? intellight fabric pracessor[EB]. 2022.
[12]	YANG T , XIE G G , LI Y B ,et al. Guarantee IP lookup performance with FIB explosion[C]// SIGCOMM '14:Proceedings of the 2014 ACM Conference on SIGCOMM. New York:ACM Press, 2014: 39-50.
[13]	LUO L Y , XIE G G , SALAMATIAN K ,et al. A trie merging approach with incremental updates for virtual routers[J]. 2013 Proceedings IEEE INFOCOM, 2013: 1222-1230.
[14]	ZHANG C Y , XIE G G . Using XorOffsetTrie for high- performance IPv6 lookup in the backbone network[J]. Computer Communications, 2022(181): 438-445.
[15]	VAMANAN B , VOSKUILEN G , VIJAYKUMAR T . EffiCuts:optimizing packet classification for memory and throughput[J]. ACM SIGCOMM Computer Communication Review, 2010,40(4): 207-218.
[16]	YINGCHAREONTHAWORNCHAI S , DALY J , LIU A X ,et al. A sorted-partitioning approach to fast and scalable dynamic packet classification[J]. IEEE/ACM Transactions on Networking, 2018,26(4): 1907-1920.
[17]	HE P , XIE G G , SALAMATIAN K ,et al. Meta-algorithms for software-based packet classification[C]// Proceedings of 2014 IEEE 22nd International Conference on Network Protocols. Piscataway:IEEE Press, 2014: 308-319.
[18]	SRINIVASAN V , SURI S , VARGHESE G . Packet classification using tuple space search[C]// Proceedings of the Conference on Applications,Technologies,Architectures,and Protocols for Computer Communication - SIGCOMM '99. New York:ACM Press, 1999.
[19]	DALY J , BRUSCHI V , LINGUAGLOSSA L ,et al. TupleMerge:fast software packet processing for online packet classification[J]. IEEE/ACM Transactions on Networking, 2019,27(4): 1417-1431.
[20]	ZHANG X Y , SALAMATIAN K , XIE G G . Improving open virtual switch performance through tuple merge relaxation in software defined networks[J]. IEEE Transactions on Network and Service Management, 255927(11): 1.
[21]	ZHANG X Y , XIE G G , WANG X ,et al. Fast online packet classification with convolutional neural network[J]. IEEE/ACM Transactions on Networking, 2021,29(6): 2765-2778.
[22]	KOHLER E , MORRIS R , CHEN B J ,et al. The click modular router[J]. ACM Transactions on Computer Systems, 2000,18(3): 263-297.
[23]	BREBNER G , JIANG W R . High-speed packet processing using reconfigurable computing[J]. IEEE Micro, 2014,34(1): 8-18.
[24]	BOSSHART P , DALY D , GIBB G ,et al. p4[J]. ACM SIGCOMM Computer Communication Review, 2014,44(3): 87-95.
[25]	CUI P L , PAN H , LI Z Y ,et al. NetFC:enabling accurate floating-point arithmetic on programmable switches[C]// Proceedings of 2021 IEEE 29th International Conference on Network Protocols. Piscataway:IEEE Press, 2021: 1-11.
[26]	CHEN X Q , . Implementing AES encryption on programmable switches via scrambled lookup tables[C]// SPIN '20:Proceedings of the Workshop on Secure Programmable Network Infrastructure.[S.l:s.n. ], 2020: 8-14.
[27]	VAUCHER S , YAZDANI N , FELBER P ,et al. ZipLine:in-network compression at line speed[C]// Proceedings of the 16th International Conference on emerging Networking EXperiments and Technologies.[S.l.:s.n.], 2020: 399-405.
[28]	OLTEANU V , AGACHE A , VOINESCU A ,et al. Stateless datacenter load-balancing with beamer[C]// Proceedings of 15th USENIX Symposium on Networked Systems Design and Implementation.[S.l.:s.n.], 2018: 125-139.
[29]	HE Y C , WU W F , WEN X M ,et al. Scalable on-switch rate limiters for the cloud[C]// Proceedings of IEEE INFOCOM 2021- IEEE Conference on Computer Communications. Piscataway:IEEE Press, 2021: 1-10.
[30]	JIN X , LI X Z , ZHANG H Y ,et al. NetCache:balancing key-value stores with fast in-network caching[C]// Proceedings of the 26th Symposium on Operating Systems Principles. New York:ACM Press, 2017.
[31]	SAPIO A , CANINI M , HO C Y ,et al. Scaling distributed machine learning with in-network aggregation[EB]. 2019.
[32]	LAO C , LE Y F , MAHAJAN K ,et al. ATP:In-network Aggregation for Multi-tenant Learning[C]// Proceedings of 18th USENIX Symposium on Networked Systems Design and Implementation.[S.l.:s.n.], 2021: 741-761.
[33]	ZHANG P H , PAN H , LI Z Y ,et al. Accelerating LSH-based distributed search with in-network computation[C]// Proceedings of IEEE INFOCOM 2021 - IEEE Conference on Computer Communications. Piscataway:IEEE Press, 2021: 1-10.
[34]	INTEL. Development reference guides:_pext_u32/64[EB]. 2022.
[35]	MA X X , YAN F , WANG Z ,et al. Survey on smart network interface card[J]. Journal of Computer Research and Development. 2022,59(1): 1-21.
[36]	INTEL. 700 series network adapters[EB]. 2022.
[37]	NETRONOME. Agilio open vSwitch TC user guide[EB]. 2022.
[38]	NAPATECH. Napatech smart NIC solution for hardware offload[EB]. 2022.
[39]	FRANKEL S , KRISHNAN S . IP security and internet key exchange document roadmap RFC 6071[EB]. 2011.
[40]	FUNGIBLE. The Fungible data processing unit[EB]. 2022.
[41]	NVIDIA. NVIDIA bluefield data processing units[EB]. 2022.
[42]	PFAFF B , PETTIT J , KOPONEN T ,et al. The design and implementation of open vswitch[C]// Proceedings of 12th USENIX Symposium on Networked Systems Design and Implementation.[S.l.:s.n.], 2015: 117-130.
[43]	NVIDIA. OVS offload using ASAP2 direct[EB]. 2022.
[44]	NVIDIA. Virtual switch on bluefield smart NIC[EB]. 2022.
[45]	CORMODE G , MUTHUKRISHNAN S . An improved data stream summary:the count-Min sketch and its applications[J]. Journal of Algorithms, 2005,55(1): 58-75.
[46]	NAMKUNG H , LIU Z X , KIM D ,et al. SketchLib:enabling efficient sketch-based monitoring on programmable switches[C]// Proceedings of 19th USENIX Symposium on Networked Systems Design and Implementation.[S.l.:s.n.], 2022: 743-759.
[47]	KRASKA T , BEUTEL A , CHI E H ,et al. The case for learned index structures[C]// Proceedings of the 2018 International Conference on Management of Data. New York:ACM Press, 2018.
[48]	MARCUS R , ZHANG E , KRASKA T . CDFShop:exploring and optimizing learned index structures[C]// Proceedings of SIGMOD '20:Proceedings of the 2020 ACM SIGMOD International Conference on Management of Data. New York:ACM Press, 2020: 2789-2792.

可编程网络数据平面技术进展

Progress in programmable network data plane

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 1

参考文献 48

相关文章 1

Metrics

推荐阅读 0