Congestion Control Using In-Network Telemetry for Lossless Datacenters

被引:6
作者
Wang, Jin [1 ]
Yuan, Dongzhi [1 ]
Luo, Wangqing [1 ]
Rao, Shuying [1 ]
Hu, Jinbin [1 ]
Sherratt, R. Simon [2 ]
机构
[1] Changsha Univ Sci & Technol, Sch Comp & Commun Engn, Changsha 410004, Peoples R China
[2] Univ Reading, Sch Syst Engn, Reading RG6 6AY, England
来源
CMC-COMPUTERS MATERIALS & CONTINUA | 2023年 / 75卷 / 01期
基金
中国国家自然科学基金; 美国国家科学基金会;
关键词
Data center; lossless networks; congestion control; head of line; blocking; in-network telemetry;
D O I
10.32604/cmc.2023.035932
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In the Ethernet lossless Data Center Networks (DCNs) deployed with Priority-based Flow Control (PFC), the head-of-line blocking problem is still difficult to prevent due to PFC triggering under burst traffic scenarios even with the existing congestion control solutions. To address the head-of-line blocking problem of PFC, we propose a new congestion control mechanism. The key point of Congestion Control Using In-Network Telemetry for Lossless Datacenters (ICC) is to use In-Network Telemetry (INT) technology to obtain comprehensive congestion information, which is then fed back to the sender to adjust the sending rate timely and accurately. It is possible to control congestion in time, converge to the target rate quickly, and maintain a near-zero queue length at the switch when using ICC. We conducted Network Simulator-3 (NS-3) simulation experiments to test the ICC's performance. When compared to Congestion Control for Large-Scale RDMA Deployments (DCQCN), TIMELY: RTT-based Congestion Control for the Datacenter (TIMELY), and Re-architecting Congestion Management in Lossless Ethernet (PCN), ICC effectively reduces PFC pause messages and Flow Completion Time (FCT) by 47%, 56%, 34%, and 15.3x, 14.8x, and 11.2x, respectively.
引用
收藏
页码:1195 / 1212
页数:18
相关论文
共 35 条
  • [1] Data Center TCP (DCTCP)
    Alizadeh, Mohammad
    Greenberg, Albert
    Maltz, David A.
    Padhye, Jitendra
    Patel, Parveen
    Prabhakar, Balaji
    Sengupta, Sudipta
    Sridharan, Murari
    [J]. ACM SIGCOMM COMPUTER COMMUNICATION REVIEW, 2010, 40 (04) : 63 - 74
  • [2] [Anonymous], 2010, C OPTICAL INTERNET
  • [3] Design and Implementation of a Network Coding Platform based on NetFPGA and ns-3
    Chan, Shu-Min
    Xie, Meng-Hua
    Chang, Han-Sheng
    Huang, Chin-Ya
    [J]. 2021 IEEE CONFERENCE ON NETWORK FUNCTION VIRTUALIZATION AND SOFTWARE DEFINED NETWORKS (IEEE NFV-SDN), 2021, : 108 - 109
  • [4] Cheng WX, 2020, PROCEEDINGS OF THE 17TH USENIX SYMPOSIUM ON NETWORKED SYSTEMS DESIGN AND IMPLEMENTATION, P19
  • [5] Remote Direct Memory Access over the Converged Enhanced Ethernet Fabric: Evaluating the Options
    Cohen, David
    Talpey, Thomas
    Kanevsky, Arkady
    Cummings, Uri
    Krause, Michael
    Recio, Renato
    Crupnicoff, Diego
    Dickman, Lloyd
    Grun, Paul
    [J]. 2009 17TH IEEE SYMPOSIUM ON HIGH-PERFORMANCE INTERCONNECTS (HOTI 2009), 2009, : 123 - +
  • [6] Curtis AR, 2011, IEEE INFOCOM SER, P1629, DOI 10.1109/INFCOM.2011.5934956
  • [7] Danfeng Shan, 2015, 2015 IEEE Conference on Computer Communications (INFOCOM). Proceedings, P118, DOI 10.1109/INFOCOM.2015.7218374
  • [8] Criso: An Incremental Scalable and Cost-Effective Network Architecture for Data Centers
    Feng, Hao
    Deng, Yuhui
    Qin, Xiao
    Min, Geyong
    [J]. IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT, 2021, 18 (02): : 2016 - 2029
  • [9] Hu Jiangqi., 2021, 2021 18th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON), P1
  • [10] Adjusting Switching Granularity of Load Balancing for Heterogeneous Datacenter Traffic
    Hu, Jinbin
    Huang, Jiawei
    Lyu, Wenjun
    Li, Weihe
    Li, Zhaoyi
    Jiang, Wenchao
    Wang, Jianxin
    He, Tian
    [J]. IEEE-ACM TRANSACTIONS ON NETWORKING, 2021, 29 (05) : 2367 - 2384