• DocumentCode
    2954164
  • Title

    A scalability study of interconnect architectures for System-on-Chip

  • Author

    Suboh, Suboh A. ; Narayana, Vikram K. ; Bakhouya, Mohamed ; El-Ghazawi, Tarek

  • Author_Institution
    George Washington Univ., Washington, DC, USA
  • fYear
    2012
  • fDate
    2-6 July 2012
  • Firstpage
    300
  • Lastpage
    306
  • Abstract
    Network-on-Chip (NoC) architectures were proposed to solve scalability issues experienced in bus-based SoCs. They incorporate a communication infrastructure defined by topology, routers and switches, in order to provide a scalable and high performance network for the SoC resources while satisfying the constraints of embedded platforms. The choice of appropriate NoC topology depends on the desired network size for the required performance. Simulation of NoCs based on real application traffic is time consuming, and therefore not a feasible approach for rapid design space exploration. In this paper, a methodology to study the scalability of three on-chip interconnect architectures, WK-recursive, Mesh and Spidergon, is presented. Simulation results are presented for different cases, demonstrating the potential of our approach for selecting the most scalable on-chip interconnect architecture.
  • Keywords
    logic design; network topology; network-on-chip; Mesh; NoC topology; Spidergon; WK-recursive; bus-based SoC; embedded platform; network-on-chip; on-chip interconnect architecture; real application traffic; router; scalability study; switches; system-on-chip; Computer architecture; Measurement; Network topology; Power demand; Scalability; System-on-a-chip; Topology; On-chip interconnect; Performance evaluation; Scalability; System-on-Chip;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    High Performance Computing and Simulation (HPCS), 2012 International Conference on
  • Conference_Location
    Madrid
  • Print_ISBN
    978-1-4673-2359-8
  • Type

    conf

  • DOI
    10.1109/HPCSim.2012.6266928
  • Filename
    6266928