University of Sussex
Browse
mmptcp_kheirkhah.pdf (1.16 MB)

MMPTCP: a multipath transport protocol for data centers

Download (1.16 MB)
conference contribution
posted on 2023-06-08, 23:40 authored by Morteza Kheirkhah, Ian WakemanIan Wakeman, George ParisisGeorge Parisis
Modern data centres provide large aggregate network capacity and multiple paths among servers. Traffic is very diverse; most of the data is produced by long, bandwidth hungry flows but the large majority of flows, which commonly come with strict deadlines regarding their completion time, are short. It has been shown that TCP is not efficient for any of these types of traffic in modern data centres. More recent protocols such MultiPath TCP (MPTCP) are very efficient for long flows, but are ill-suited for short flows. In this paper, we present MMPTCP, a novel transport protocol which, compared to TCP and MPTCP, reduces short flows' completion times, while providing excellent goodput to long flows. MMPTCP runs in two phases; initially, it randomly scatters packets in the network under a single congestion window exploiting all available paths. This is beneficial to latency-sensitive flows. After a specific amount of data is sent, MMPTCP switches to a regular MPTCP mode. MMPTCP is incrementally deployable in existing data centres as it does not require any modifications outside the transport layer and behaves well when competing with legacy TCP and MPTCP flows. Our extensive experimental evaluation shows that all design objectives for MMPTCP are met.

History

Publication status

  • Published

File Version

  • Accepted version

Journal

Proceedings of the 35th Annual IEEE International Conference on Computer Communications, IEEE INFOCOM 2016; San Francisco, CA, USA; 10-15 April 2016

Publisher

Institute of Electrical and Electronics Engineers

Page range

1-9

Event name

IEEE INFOCOM 2016 - The 35th Annual IEEE International Conference on Computer Communications

Event location

San Francisco, CA, USA

Event type

conference

Event date

10-14 April 2016

ISBN

9781467399548

Department affiliated with

  • Informatics Publications

Full text available

  • Yes

Peer reviewed?

  • Yes

Legacy Posted Date

2015-12-06

First Open Access (FOA) Date

2017-07-10

First Compliant Deposit (FCD) Date

2015-12-05

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC