• DocumentCode
    1796451
  • Title

    The SDN controller placement problem for WAN

  • Author

    Peng Xiao ; Wenyu Qu ; Heng Qi ; Zhiyang Li ; Yujie Xu

  • Author_Institution
    Coll. of Inf. Sci. & Technol., Dalian Maritime Univ., Dalian, China
  • fYear
    2014
  • fDate
    13-15 Oct. 2014
  • Firstpage
    220
  • Lastpage
    224
  • Abstract
    Large networks are always partitioned into several small networks when deploying software defined networking (SDN). The network partitioning with SDN control planes open many unanswered questions such as latency, reliability, and load balancing. This paper opens the investigation by focusing on two specific questions: given a wide-area network topology, how to partition it into several small SDN domains, and where should the controller go in each SDN domain? To answer these questions, we propose a novel approach to efficiently and accurately evaluate SDN controller placement problem for WAN. This approach uses the Spectral Clustering placement algorithm to partition a large network into several small SDN domains. After presenting our metrics of the SDN controller placement problem, we evaluate our approach using the Internet2 topology and other available WAN topologies. For testing purposes we developed a new framework to test the SDN controller in WAN. The results show its effectiveness for the SDN controller placement problem.
  • Keywords
    controllers; pattern clustering; resource allocation; software defined networking; wide area networks; Internet2 topology; SDN control plane; SDN controller placement problem; SDN domain; WAN; load balancing; network partitioning; software defined networking; spectral clustering placement algorithm; wide-area network topology; Clustering algorithms; Measurement; Partitioning algorithms; Reliability; Switches; Topology; Wide area networks; Controller Placement; SDN; Spectral Clustering;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Communications in China (ICCC), 2014 IEEE/CIC International Conference on
  • Conference_Location
    Shanghai
  • Type

    conf

  • DOI
    10.1109/ICCChina.2014.7008275
  • Filename
    7008275