• DocumentCode
    2366063
  • Title

    SCTP Throughput Improvement with Best Load Sharing Based on Multihoming

  • Author

    Joe, Inwhee ; Yan, Sijia

  • Author_Institution
    Dept. of Electron. & Comput. Eng., Hanyang Univ., Seoul, South Korea
  • fYear
    2009
  • fDate
    25-27 Aug. 2009
  • Firstpage
    138
  • Lastpage
    142
  • Abstract
    The Stream Control Transmission Protocol (SCTP) is a new transport layer protocol. Multihoming is one of the most important features in SCTP, which can be used by the data sender to send data to a receiver through different paths. Although SCTP provides support for multihoming, the basic reason for such a provision was to improve reliability of associations. Currently, simultaneous transfer of new data to multiple paths is not allowed in SCTP. We can gain significant throughput improvement by simultaneously transferring new data across multiple paths to the receiver. In this paper, the best load sharing using SCTP multihoming is discussed and the techniques are analyzed. In order to achieve load sharing, original congestion control mechanism should be modified to implement load sharing. The new algorithm, split fast retransmission is the key of SCTP load sharing. Finally, through NS-2 simulations we prove that the proposed scheme is effective.
  • Keywords
    telecommunication congestion control; transport protocols; SCTP load sharing; SCTP multihoming; SCTP throughput improvement; best load sharing; congestion control; split fast retransmission; stream control transmission protocol; transport layer protocol; Condition monitoring; Floods; Flowcharts; Heart beat; Redundancy; Research and development; Throughput; Transport protocols; Wireless application protocol; Wireless networks; Best Load Sharing; Multihoming; SCTP; Split Fast Retransmission; Throughput;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    INC, IMS and IDC, 2009. NCM '09. Fifth International Joint Conference on
  • Conference_Location
    Seoul
  • Print_ISBN
    978-1-4244-5209-5
  • Electronic_ISBN
    978-0-7695-3769-6
  • Type

    conf

  • DOI
    10.1109/NCM.2009.170
  • Filename
    5331737