EagerCC: An ultra-low latency congestion control mechanism in datacenter networks

被引:0
|
作者
Lu, Yuan [1 ]
Yuan, Guoyuan [1 ]
Bai, Yang [1 ]
Dong, Dezun [1 ]
Zhou, Renjie [2 ]
机构
[1] Natl Univ Def Technol, Changsha 410000, Peoples R China
[2] Northwest Inst Nucl Technol, Xian 710000, Peoples R China
关键词
Datacenter; HPDC; Congestion control; In-network-telemetry; Switch calculation; Probabilistic feedback;
D O I
10.1016/j.comnet.2023.110009
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
With the rapid development of cloud applications, the workload pattern of datacenters presents the characteristics of mixed long and short flows and frequent micro-burst traffic, which puts forward new requirements for network transmission performance, including ultra-low latency, high throughput, and strong stability. At the same time, datacenters begin to deploy the low-diameter topology to accommodate these new requirements, and the high-performance datacenter (HPDC) comes into being. However, due to the complexity of the load, existing congestion control mechanisms cannot control dynamic delay in the network well, which significantly restricts the development of the HPDC. Therefore, it is necessary to deploy the congestion control mechanism for the HPDC. So we propose EagerCC, an ultra-low latency, low-overhead, and accurate congestion control mechanism based on In-Network-Telemetry (INT) information for various datacenter scenarios, especially for HPDCs. EagerCC uses switch-feedback, ACK-padding, ACK-first to reduce feedback delay and uses switch calculation, probabilistic ACK-padding to reduce the overhead of congestion signals. We conduct a lot of experiments and the result shows that EagerCC performs well in various datacenter scenarios, especially for HPDCs. Specifically, EagerCC reduces the 99th-FCT and avg-FCT by 52.3% and 13.3% for the HPC workload NPB-CG compared to HPCC in Dragonfly. In addition, EagerCC significantly reduces the network's feedback delay and queue occupancy.
引用
收藏
页数:12
相关论文
共 50 条
  • [1] Bolt: Sub-RTT Congestion Control for Ultra-Low Latency
    Arslan, Serhat
    Li, Yuliang
    Kumar, Gautam
    Dukkipati, Nandita
    PROCEEDINGS OF THE 20TH USENIX SYMPOSIUM ON NETWORKED SYSTEMS DESIGN AND IMPLEMENTATION, NSDI 2023, 2023, : 219 - 236
  • [2] LHCC: Low-Latency and Hi-Precision Congestion Control in RDMA Datacenter Networks
    Yan, Bodong
    Zhao, Yangming
    Xu, Sun
    Liu, Jianchun
    Xu, Hongli
    2024 IEEE/ACM 32ND INTERNATIONAL SYMPOSIUM ON QUALITY OF SERVICE, IWQOS, 2024,
  • [3] A new ultra-low latency message transfer mechanism
    Froening, Holger
    Litz, Heiner
    Bruening, Ulrich
    PROCEEDINGS OF THE SIXTH IASTED INTERNATIONAL CONFERENCE ON COMMUNICATION SYSTEMS AND NETWORKS, 2007, : 108 - +
  • [4] OneSpike: Ultra-low latency spiking neural networks
    Tang, Kaiwen
    Yan, Zhanglu
    Wong, Weng-Fai
    2024 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN 2024, 2024,
  • [5] ProCAM: A Proactive Coordinating Mechanism for Low-Congestion Datacenter Networks
    Xu, Xin
    Zhou, Wei
    Yao, Jianguo
    PROCEEDINGS OF 2016 IEEE 18TH INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPUTING AND COMMUNICATIONS; IEEE 14TH INTERNATIONAL CONFERENCE ON SMART CITY; IEEE 2ND INTERNATIONAL CONFERENCE ON DATA SCIENCE AND SYSTEMS (HPCC/SMARTCITY/DSS), 2016, : 695 - 702
  • [6] Active control and management system for providing the ultra-low latency serve on deterministic networks
    Kim, Eungha
    Ryoo, Yeoncheol
    Yoon, Binyeong
    Cheung, Taesik
    12TH INTERNATIONAL CONFERENCE ON UBIQUITOUS AND FUTURE NETWORKS (ICUFN 2021), 2021, : 70 - 74
  • [7] Intelligent Routing Orchestration for Ultra-Low Latency Transport Networks
    Meng, Qingmin
    Wei, Jingcheng
    Wang, Xiaoming
    Guo, Haiyan
    IEEE ACCESS, 2020, 8 : 128324 - 128336
  • [8] HAECN: Hierarchical Automatic ECN Tuning with Ultra-Low Overhead in Datacenter Networks
    Hu, Jinbin
    Wang, Youyang
    Zhou, Zikai
    Rao, Shuying
    Xin, Rundong
    Wang, Jing
    He, Shiming
    ALGORITHMS AND ARCHITECTURES FOR PARALLEL PROCESSING, ICA3PP 2023, PT III, 2024, 14489 : 324 - 343
  • [9] Congestion Control for Cross-Datacenter Networks
    Zeng, Gaoxiong
    Bai, Wei
    Chen, Ge
    Chen, Kai
    Han, Dongsu
    Zhu, Yibo
    Cui, Lei
    IEEE-ACM TRANSACTIONS ON NETWORKING, 2022, 30 (05) : 2074 - 2089
  • [10] Congestion Control for Cross-Datacenter Networks
    Zeng, Gaoxiong
    Bai, Wei
    Chen, Ge
    Chen, Kai
    Han, Dongsu
    Zhu, Yibo
    Cui, Lei
    2019 IEEE 27TH INTERNATIONAL CONFERENCE ON NETWORK PROTOCOLS (IEEE ICNP), 2019,