智算数据中心光电交换技术综述（特邀）

唐雄燕; 魏步征; 沈世奎; 王创业; 王泽林; 张贺; 王光全; 张晨芳

doi:10.13756/j.gtxyj.2024.240062

您当前的位置：

首页 >

文章列表页 >

智算数据中心光电交换技术综述（特邀）

专题：数据中心内光交换 | 更新时间：2024-10-08

- 智算数据中心光电交换技术综述（特邀）
- Overview of Optoelectronic Switching Technology in Artificial Intelligent Data Centers
- 光通信研究 2024年第5期页码：24006201-24006213
- 作者机构：
  
  中国联通研究院，北京　100048
- 作者简介：
  
  唐雄燕（1967－），男，湖南永州人。教授级高工，博士，主要研究方向为光通信。
  魏步征（1991－），高级工程师，博士。E-mail：weibz5@chinaunicom.cn
- 基金信息：
  
  国家重点研发计划资助项目(2022YFB2804102)
- DOI：10.13756/j.gtxyj.2024.240062
  中图分类号： TN29
- 纸质出版日期：2024-10-10，
  
  收稿日期：2024-03-20，
  
  修回日期：2024-04-10，
- 稿件说明：
移动端阅览
唐雄燕，魏步征，沈世奎，等. 智算数据中心光电交换技术综述[J].光通信研究，2024(5)：240062.

Tang X Y, Wei B Z, Shen S K，et al. Overview of Optoelectronic Switching Technology in Intelligent Computing Data Centers[J]. Study on Optical Communications，2024(5)：240062.
唐雄燕，魏步征，沈世奎，等. 智算数据中心光电交换技术综述[J].光通信研究，2024(5)：240062. DOI： 10.13756/j.gtxyj.2024.240062.

Tang X Y, Wei B Z, Shen S K，et al. Overview of Optoelectronic Switching Technology in Intelligent Computing Data Centers[J]. Study on Optical Communications，2024(5)：240062. DOI： 10.13756/j.gtxyj.2024.240062.

摘要

近年来，由于人工智能现象级应用的出现，智算数据中心（AIDC）和超算数据中心（SCDC）网络演进成为研究热点。传统的3层全连接电交换架构已经不能很好地满足AIDC或SCDC高带宽、低时延、低功耗和低成本的需求，需要一种更为高效和易于扩容的电光融合交换或光交换方案来逐步替代传统纯电路包交换的策略。文章对近年来基于数据中心的光电融合交换技术方案进行了综述性介绍，对行业中率先进行光交换技术尝试的头部互联网企业的部署经验进行了分析，并结合电信行业需求，给出了AIDC和SCDC交换节点的推荐部署和演进方式。

Abstract

In recent years

due to the emergence of phenomenal applications of artificial intelligence

the evolution of Artificial Intelligent Data Center (AIDC) and Super Computing Data Center (SCDC) network has become a hot research topic. The traditional three-layer fully connected electrical switching architecture can not well meet the requirement of high bandwidth

low delay

low power consumption and low cost of the AIDC or SCDC

which requires a more efficient and easy expansion way of the electrical and optical fusion switching or optical switching scheme to gradually replace the traditional pure circuit package switching strategy. In this paper

the photoelectric fusion switch technology solutions based on data center in recent years are summarized. The deployment experience of leading Internet enterprises that take the lead in optical switch technology is analyzed. Combined with the needs of the telecom industry

the recommended deployment and evolution strategy of the switch nodes of AIDC and SCDC and super computing center are given.

关键词

智算数据中心交换架构光电融合交换光电路交换

Keywords

AIDC switching architectureoptoelectronic convergence switchingoptical circuit switching

references

Al-Fares M, Loukissas A, Vahdat A. A Scalable, Commodity Data Center Network Architecture[C]//Proceedings of the ACM SIGCOMM 2008 Conference on Data Communication. Seattle, WA, USA: ACM, 2008: 63-74.

Cisco Systems, Inc. Cisco Data Center Infrastructure 2.5 Design Guide[EB/OL]. (2011-11-02)[2024-03-20]. https://www.cisco.com/c/en/us/td/docs/solutions/Enterprise/Data_Center/DC_Infra2_5/DCI_SRND_2_5a_book/DCInfra_3a.htmlhttps://www.cisco.com/c/en/us/td/docs/solutions/Enterprise/Data_Center/DC_Infra2_5/DCI_SRND_2_5a_book/DCInfra_3a.html.

王甫涵, 郝祥勇, 蔡轶, 等. 全光数据中心互联的混合放大技术研究[J]. 光通信研究, 2023(3): 6-9.

Wang F H, Hao X Y, Cai Y, et al. Research on Hybrid Amplification Technology of All-optical Data Center Interconnection[J]. Study on Optical Communications, 2023(3): 6-9.

Juniper Networks, Inc. Documentation Archives (Portable Libraries)[EB/OL]. (2024-02-08) [2024-03-20]. http://www.juniper.net/techpubs/software/erx/junose61/swconfig-routing-vol1/html/ip-jflow-stats-config2.htmlhttp://www.juniper.net/techpubs/software/erx/junose61/swconfig-routing-vol1/html/ip-jflow-stats-config2.html.

Sun Microsystems, Inc. SUN™ Datacenter Switch 3456 System Architecture Massively Scalable InfiniBand Switch Architecture for Petascale Computing White Paper[EB/OL]. (2007-11-01)[2024-03-20]. https://www.cs.rpi.edu/~chrisc/COURSES/HPDC/SPRING-2008/papers/ds3456_wp.pdfhttps://www.cs.rpi.edu/~chrisc/COURSES/HPDC/SPRING-2008/papers/ds3456_wp.pdf.

工业和信息化部. 新型数据中心发展三年行动计划(2021-2023年)[EB/OL]. (2021-07-04) [2024-03-20]. https://www.gov.cn/zhengce/zhengceku/2021-07/14/content_5624964.htmhttps://www.gov.cn/zhengce/zhengceku/2021-07/14/content_5624964.htm.

Ministry of Industry and Information Technology. Three-year Action Plan for the Development of New data Centers (2021-2023)[EB/OL]. (2021-07-04) [2024-03-20]. https://www.gov.cn/zhengce/zhengceku/2021-07/14/content_5624964.htmhttps://www.gov.cn/zhengce/zhengceku/2021-07/14/content_5624964.htm.

Abadi M, Barham P, Chen J, et al. TensorFlow: A System for Large-scale Machine Learning[DB/OL]. (2016-05-27) [2024-03-20]. https://arxiv.org/abs/1605.08695https://arxiv.org/abs/1605.08695.

张镭. 深度学习在自然语言处理中的应用：从词表征到ChatGPT[M]. 北京: 人民邮电出版社, 2023.

Zhang L. Application of Deep Learning in Natural Language Processing: From Word Representation to ChatGPT[M]. Beijin: Posts and Telecom Press, 2023.

Nath S, Marie A, Ellershaw S, et al. New Meaning for NLP: The Trials and Tribulations of Natural Language Processing with GPT-3 in Ophthalmology[J]. The British Journal of Ophthalmology, 2022, 106(7): 889-892.

Asim M, Wang Y, Wang K, et al. A Review on Computational Intelligence Techniques in Cloud and Edge Computing[J]. IEEE Transactions on Emerging Topics in Computational Intelligence, 2020, 4(6): 742-763.

Farrington N, Porter G, Radhakrishnan S, et al. Helios: a Hybrid Electrical/Optical Switch Architecture for Modular Data Centers[C]//Proceedings of the ACM SIGCOMM 2010 Conference. New Delhi, India: ACM, 2010: 339-350.

Wang G, Andersen D G, Kaminsky M, et al. C-Through: Part-Time Optics in Data Centers[J]. ACM SIGCOMM Computer Communication Review, 2010, 40(4): 327-338.

Farrington N, Forencich A, Porter G, et al. A Multiport Microsecond Optical Circuit Switch for Data Center Networking[J]. IEEE Photonics Technology Letters, 2013, 25(16): 1589-1592.

Mellette W M, McGuinness R, Roy A, et al. RotorNet: a Scalable, Low-complexity, Optical Datacenter Network[C]//Proceedings of the Conference of the ACM Special Interest Group on Data Communication. Los Angeles, CA, USA: ACM, 2017: 267-280.

Ghobadi M, Mahajan R, Phanishayee A, et al. ProjecToR: Agile Reconfigurable Data Center Interconnect[C]//Proceedings of the 2016 ACM SIGCOMM Conference. Florianopolis, Brazil: ACM, 2016: 216-229.

Zhao Z, Xue X, Guo B, et al. ReSAW: a Reconfigurable and Picosecond-synchronized Optical Data Center Network based on an AWGR and the WR Protocol[J]. Journal of Optical Communications and Networking, 2022, 14(9): 702-712.

Dang D, Guo B, Li W, et al. AWGR-based 25 Gb/s Optical-electrical Switching System Featuring Dynamic Bandwidth Allocating[C]//49th European Conference on Optical Communications (ECOC 2023). Hybrid Conference, Glasgow, UK: IET, 2023: 1745-1748.

Hamedazimi N, Gupta H, Sekar V, et al. Patch Panels in the Sky: A Case for Free-space Optics in Data Centers[C]//Proceedings of the Twelfth ACM Workshop on Hot Topics in Networks. College Park Maryland, USA: ACM, 2013: 2535771.

Hamedazimi N, Qazi Z, Gupta H, et al. FireFly[J]. ACM SIGCOMM Computer Communication Review, 2015, 44(4): 319-330.

尹欣, 侯维刚, 郭磊. 一种自由空间光数据中心网络架构、拓扑重构系统和方法：中国，CN114745618A[P]. 2022-07-12.

Yin X, Hou W G, Guo L. Free Space Optical Data Center Network Architecture, Topology Reconstruction System and Method: China, CN114745618A[P]. 2022-07-12.

Ballani H, Costa P, Behrendt R, et al. Sirius: a Flat Datacenter Network with Nanosecond Optical Switching[C]//Proceedings of the Annual Conference of the ACM Special Interest Group on Data Communication on the Applications, Technologies, Architectures, and Protocols for Computer Communication. Virtual Event, USA: ACM, 2020: 782-797.

Singla A, Singh A, Ramachandran K, et al. Proteus: A Topology Malleable Data Center Network[C]//Proceedings of the 9th ACM SIGCOMM Workshop on Hot Topics in Networks. Monterey California, USA: ACM, 2010: 1868447.

Chen K, Singla A, Singh A, et al. OSA: An Optical Switching Architecture for Data Center Networks with Unprecedented Flexibility[J]. IEEE/ACM Transactions on Networking, 2014, 22(2): 498-511.

Lu Y, Gu H, Yu X, et al. Lotus: A New Topology for Large-scale Distributed Machine Learning[J]. ACM Journal on Emerging Technologies in Computing Systems, 2021, 17(1): 1-21.

Alam S, Athanassiadou T, Robinson T W, et al. First 12-cabinets Cray XC30 System at CSCS: Scaling and Performance Efficiencies of Applications[C]//CUG 2013. Napa Valley, CA, USA: CUG, 2013: 285583551.

李泳成, 廖晶晶, 沈纲祥. 面向基于Spanke架构全光交换数据中心的Ring业务部署方法：中国，CN115175027B[P]. 2023-06-23.

Li Y C, Liao J J, Shen G X. Ring Service Deployment Method for All-optical Switching Data Center based on Spanke Architecture: China, CN115175027B[P]. 2023-06-23.

Anderson E, González J, Gazman A, et al. Optically Connected and Reconfigurable GPU Architecture for Optimized Peer-to-peer Access[C]//Proceedings of the International Symposium on Memory Systems. Alexandria Virginia, USA: ACM, 2018: 257-258.

浏览量

下载量

CSCD

CNKI被引量

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据