Research on deterministic networking requirements and technologies for RDMA-WAN

Junfeng ZHAO; Fang LI; Xiaofeng YE; Song JIANG

doi:10.11959/j.issn.1000-0801.2023248

您当前的位置：

首页 >

文章列表页 >

Research on deterministic networking requirements and technologies for RDMA-WAN

Topic: Deterministic Networking | 更新时间：2024-06-05

- Research on deterministic networking requirements and technologies for RDMA-WAN
- Telecommunications Science Vol. 39, Issue 11, Pages: 39-51(2023)
- 作者机构：
  
  1. 中国信息通信研究院，北京 100191
  2. 江苏为是科技有限公司，江苏苏州 215000
  3. 国网江苏省电力有限公司信息通信分公司，江苏南京 210024
- 作者简介：
- 基金信息：
- DOI：10.11959/j.issn.1000-0801.2023248
  CLC： TP393
- Published Online：2023-11，
  
  Published：20 November 2023
- 稿件说明：
移动端阅览
Junfeng ZHAO, Fang LI, Xiaofeng YE, et al. Research on deterministic networking requirements and technologies for RDMA-WAN[J]. Telecommunications science, 2023, 39(11): 39-51.
DOI：

Junfeng ZHAO, Fang LI, Xiaofeng YE, et al. Research on deterministic networking requirements and technologies for RDMA-WAN[J]. Telecommunications science, 2023, 39(11): 39-51. DOI： 10.11959/j.issn.1000-0801.2023248.

摘要

随着生成式人工智能（artificial intelligence generated content，AIGC）大模型和智能计算应用的高速发展，面向广域远程直接存储器访问（remote direct memory access，RDMA）的确定性网络技术成为近期智算中心互联的研究热点。首先分析了广域 RDMA技术发展路线、实现方案和技术挑战，然后研究广域 RDMA部署时面临的各类网络问题，通过搭建实验室和外场测试环境开展了一系列测试验证，分析提出了广域RDMA的确定性需求特性和边界性能指标，最后对面向广域RDMA的确定性网络技术进行了总结和展望。

Abstract

With the rapid development of AIGC large models and intelligent computing applications

deterministic networking technology for wide-area RDMA has become a recent research hotspot.The development path

implementation solutions and technical challenges of RDMA-WAN were analyzed

by setting up a laboratory and an outfield testing environment

with a series of testing and verifications.The deterministic demand characteristics and boundary performance indicators of RDMA-WAN were proposed.Finally

the deterministic networking technologies for RDMA-WAN were summarized and prospected.

关键词

Keywords

references

CARTER S , MINICH M , RAO N S . Experimental evaluation of in finiband transport over local-and wide-area networks [C ] // Proceedings of the 2007 Spring Simulation Multi Conference-Volume 2 .[S.l.:s.n. ] , 2007 : 419 - 426 .

YU W K , RAO N S V , VETTER J S . Experimental analysis of InfiniBand transport services on WAN [C ] // Proceedings of 2008 International Conference on Networking,Architecture,and Storage . Piscataway:IEEE Press , 2008 : 233 - 240 .

LU Y , CHEN G , LI B , et al . Multi-path transport for RDMA in datacenters [C ] // Proceedings of 15th USENIX Symposium on Networked Systems Design and Implementation (NSDI 18) .[S.l.:s.n. ] , 2018 : 357 - 371 .

ALIZADEH M , EDSALL T , DHARMAPURIKAR S , et al . Conga:distributed congestion-aware load balancing for datacenters [J ] . ACM SIGCOMM Computer Communication Review , 2014 , 44 ( 4 ): 503 - 514 .

RAICIU C , BARRE S , PLUNTKE C , et al . Improving datacenter performance and robustness with multipath TCP [C ] // Proceedings of the ACM SIGCOMM 2011 Conference . New York:ACM Press , 2011 .

GUO C X , WU H T , DENG Z , et al . RDMA over commodity Ethernet at scale [C ] // Proceedings of the 2016 ACM SIGCOMM Conference . New York:ACM Press , 2016 : 202 - 215 .

SHALEV L , AYOUB H , BSHARA N , et al . A cloud-optimized transport protocol for elastic and scalable HPC [J ] . IEEE Micro , 2020 , 40 ( 6 ): 67 - 73 .

BAI W , ABDEEN S S , AGRAWAL A , et al . Empowering azure storage with RDMA [C ] // Proceedings of 20th USENIX Symposium on Networked Systems Design and Implementation (NSDI 23) .[S.l.:s.n. ] , 2023 : 49 - 67 .

CHEN Y Q , TIAN C , DONG J Q , et al . Swing:providing long-range lossless RDMA via PFC-relay [J ] . IEEE Transactions on Parallel and Distributed Systems , 2023 , 34 ( 1 ): 63 - 75 .

SHPINER A , ZAHAVI E , DAHLEY O , et al . RoCE rocks without PFC:detailed evaluation [C ] // Proceedings of the Workshop on Kernel-Bypass Networks . New York:ACM Press , 2017 : 25 - 30 .

MATHIS M , SEMKE J , MAHDAVI J , et al . The macroscopic behavior of the TCP congestion avoidance algorithm [J ] . ACM SIGCOMM Computer Communication Review , 1997 , 27 ( 3 ): 67 - 82 .

MITTAL R , SHPINER A , PANDA A , et al . Revisiting network support for RDMA [C ] // Proceedings of the 2018 Conference of the ACM Special Interest Group on Data Communication . New York:ACM Press , 2018 : 313 - 326 .

ZHU Y B , ERAN H , FIRESTONE D , et al . Congestion control for large-scale RDMA deployments [J ] . ACM SIGCOMM Computer Communication Review , 2015 , 45 ( 4 ): 523 - 536 .

MITTAL R , LAM V T , DUKKIPATI N , et al . Timely:RTT-based congestion control for the data-center [J ] . ACM SIGCOMM Computer Communication Review , 2015 , 45 ( 4 ): 537 - 550 .

刘运渠 . 2023算力互联互通大会贡献示范单位“为是科技”展示长途RDMA [EB ] . 2023 .

LIU Y Q . Viscore Tech.demonstrates long-haul RDMA in the 2023 Computility Interconnection Conference [EB ] . 2023 .

陆璐 . 网络与AI融合的思考和实践 [R ] . 2023 .

LU L . Thinking and practicing network and AI convergence [R ] . 2023 .

YU Z L , SU B W , BAI W , et al . Understanding the micro-behaviors of hardware offloaded network stacks with lumina [C ] // Proceedings of the ACM SIGCOMM 2023 Conference . New York:ACM Press , 2023 : 1074 - 1087 .

李彤 . 广域确定性网络中的传输优化 [R ] . 2023 .

LI T . Transmission optimization in wide-area deterministic networks [R ] . 2023 .

Views

442

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Research on deterministic networking technology for space-ground integrated

Research on control plane technology for wide-area deterministic networking

Auto-learning TSN: improvements on TSN usability

Research on space-air-ground integrated deterministic networking

Review of deterministic networking technology

Related Author

Xia ZHU

Guoxin LIANG

Jiaen ZHOU

Yafei ZHAO

Mugen PENG

Yuman ZHANG

Hongwei SHI

Tao HUANG

Related Institution

State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications

State Grid Siji Location based Service Co., Ltd.

School of Cyber Science and Engineering, Southeast University

Purple Mountain Laboratories

⁰