• DocumentCode
    649208
  • Title

    FLNR: A fast light-weight NoC router for FPGAs

  • Author

    Imbewa, Abdelrazag ; Khalid, Mohammed A. S.

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Univ. of Windsor, Windsor, ON, Canada
  • fYear
    2013
  • fDate
    4-7 Aug. 2013
  • Firstpage
    445
  • Lastpage
    448
  • Abstract
    Currently, FPGAs serve as Field-Programmable-Systems-on-Chip (FPSoCs) and are widely used to implement computationally intensive real world applications. As the number of components in FPSoCs increase, the interconnect schemes based on Network-on-Chip (NoC) approach are increasingly being used to overcome the deficiencies of the traditional bus- based and point-to-point interconnect schemes. The router is a key component that greatly impacts the performance and cost of an NoC. In this paper we present FLNR, a fast light-weight NoC router designed for FPGAs. It is a 5-port packet switched wormhole router that uses a deterministic XY Routing algorithm and round robin arbitration scheme. The size of the input buffers and the flit size are parameterizable. We used novel techniques to achieve good speed performance while minimizing the area used. The number of control fields in a packet is minimized to reduce the size of buffers used. Credit based flow control is used to reduce the number of clock cycles required for transferring each flit. Both edges of the clock are used to implement router operations thereby speeding up the router. FLNR is compared to other proposed routers based on three metrics: area, frequency and zero load latency. Synthesis results and zero load latency evaluations show that our router is significantly superior to widely referenced, previously proposed routers.
  • Keywords
    field programmable gate arrays; integrated circuit interconnections; network routing; network-on-chip; 5-port packet switched wormhole router; FLNR; FPGA; FPSoC; credit based flow control; deterministic XY routing; fast light-weight NoC router; field programmable gate arrays; field programmable systems-on-chip; network-on-chip; point-to-point interconnect schemes; round robin arbitration scheme; zero load latency evaluations;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Circuits and Systems (MWSCAS), 2013 IEEE 56th International Midwest Symposium on
  • Conference_Location
    Columbus, OH
  • ISSN
    1548-3746
  • Type

    conf

  • DOI
    10.1109/MWSCAS.2013.6674681
  • Filename
    6674681