skip to main content
article

Why flow-completion time is the right metric for congestion control

Published: 10 January 2006 Publication History
  • Get Citation Alerts
  • Abstract

    Users typically want their flows to complete as quickly as possible. This makes Flow Completion Time (FCT) an im portant - arguably the most important - performance metric for the user. Yet research on congestion control focuses almost entirely on maximizing link throughput, utilization and fairness, which matter more to the operator than the user. In this paper we show that with typical Internet flow sizes, existing (TCP Reno) and newly proposed (XCP) congestion control algorithms make flows last much longer than necessary - often by one or two orders of magnitude. In contrast, we show how a new and practical algorithm - RCP (Rate Control Protocol) - enables flows to complete close to the minimum possible

    References

    [1]
    A.M. Odlyzko, "The Internet and other networks: Utilization rates and their implications," In Information Economic and Policy, 12 (2000), Pages 341--365.
    [2]
    D.A. Patterso, "Latency Lags Bandwidth," In Communications of the ACM, Volume 47, Number 10 (2004), Pages 71--75.
    [3]
    J. Du, J.Y. Leung, G.H. Young, "Minimizing mean flow time with release time constraint," In Theoretical Computer Science archive, Volume 75, Issue 3, October 1990.
    [4]
    D. Katabi, M. Handley, C. Rohrs, "Internet Congestion Control for High Bandwidth-Delay Product Networks," In Proceeding of ACM Sigcomm 2002, Pittsburgh, August, 2002.
    [5]
    N. Dukkipati, M. Kobayashi, R. Zhang-Shen, N. McKeown, "Processor Sharing Flows in the Internet," In Thirteenth International Workshop on Quality of Service (IWQoS), Passau, Germany, June 2005.
    [6]
    N. Dukkipati, N. McKeow, "Processor Sharing Flows in the Internet," In http://yuba.stanford.edu/tr.html, Stanford HPNG Technical Report TR04-HPNG-061604, June 2004.
    [7]
    N. Dukkipati, N. McKeow, "Why Flow-Completion Time is the Right metric for Congestion Control and why this means we need new algorithms," In http://yuba.stanford.edu/tr.html, Staford HPNG Technical Report TR05-HPNG-112102, November 2005.

    Cited By

    View all
    • (2024)Accurate and fast congestion feedback in MEC-enabled RDMA datacentersJournal of Cloud Computing: Advances, Systems and Applications10.1186/s13677-024-00642-813:1Online publication date: 23-Mar-2024
    • (2024)DR-PIFO: A Dynamic Ranking Packet Scheduler Using a Push-In-First-Out QueueIEEE Transactions on Network and Service Management10.1109/TNSM.2023.330489421:1(355-371)Online publication date: 1-Feb-2024
    • (2024)Load Profiling via In-Band Flow Classification and P4 With HowdahIEEE Transactions on Network and Service Management10.1109/TNSM.2023.329972921:1(295-309)Online publication date: 1-Feb-2024
    • Show More Cited By

    Recommendations

    Comments

    Information & Contributors

    Information

    Published In

    cover image ACM SIGCOMM Computer Communication Review
    ACM SIGCOMM Computer Communication Review  Volume 36, Issue 1
    January 2006
    90 pages
    ISSN:0146-4833
    DOI:10.1145/1111322
    Issue’s Table of Contents

    Publisher

    Association for Computing Machinery

    New York, NY, United States

    Publication History

    Published: 10 January 2006
    Published in SIGCOMM-CCR Volume 36, Issue 1

    Check for updates

    Qualifiers

    • Article

    Contributors

    Other Metrics

    Bibliometrics & Citations

    Bibliometrics

    Article Metrics

    • Downloads (Last 12 months)79
    • Downloads (Last 6 weeks)8

    Other Metrics

    Citations

    Cited By

    View all
    • (2024)Accurate and fast congestion feedback in MEC-enabled RDMA datacentersJournal of Cloud Computing: Advances, Systems and Applications10.1186/s13677-024-00642-813:1Online publication date: 23-Mar-2024
    • (2024)DR-PIFO: A Dynamic Ranking Packet Scheduler Using a Push-In-First-Out QueueIEEE Transactions on Network and Service Management10.1109/TNSM.2023.330489421:1(355-371)Online publication date: 1-Feb-2024
    • (2024)Load Profiling via In-Band Flow Classification and P4 With HowdahIEEE Transactions on Network and Service Management10.1109/TNSM.2023.329972921:1(295-309)Online publication date: 1-Feb-2024
    • (2024) xNet : Modeling Network Performance With Graph Neural Networks IEEE/ACM Transactions on Networking10.1109/TNET.2023.332935732:2(1753-1767)Online publication date: Apr-2024
    • (2024)Alleviating Congestion via Switch Design for Fair Buffer Allocation in DatacentersIEEE Transactions on Cloud Computing10.1109/TCC.2024.335759512:1(219-231)Online publication date: Jan-2024
    • (2023)Datacenter Network Deserves Better Traffic ModelsProceedings of the 22nd ACM Workshop on Hot Topics in Networks10.1145/3626111.3628209(124-130)Online publication date: 28-Nov-2023
    • (2023)Green With EnvyProceedings of the 22nd ACM Workshop on Hot Topics in Networks10.1145/3626111.3628200(220-228)Online publication date: 28-Nov-2023
    • (2023)Elephants Sharing the Highway: Studying TCP Fairness in Large Transfers over High Throughput LinksProceedings of the SC '23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis10.1145/3624062.3624594(806-818)Online publication date: 12-Nov-2023
    • (2023)Mistill: Distilling Distributed Network Protocols From ExamplesIEEE Transactions on Network and Service Management10.1109/TNSM.2023.326352920:4(4110-4125)Online publication date: 1-Dec-2023
    • (2023)Threshold-Based Routing-Topology Co-Design for Optical Data CenterIEEE/ACM Transactions on Networking10.1109/TNET.2023.326527631:6(2870-2885)Online publication date: Dec-2023
    • Show More Cited By

    View Options

    Get Access

    Login options

    View options

    PDF

    View or Download as a PDF file.

    PDF

    eReader

    View online with eReader.

    eReader

    Media

    Figures

    Other

    Tables

    Share

    Share

    Share this Publication link

    Share on social media