• DocumentCode
    3048342
  • Title

    The effect of node size, heterogeneity, and network size on FPGA based NoCs

  • Author

    Lee, Jason ; Shannon, Lesley

  • Author_Institution
    Simon Fraser Univ., Burnaby, BC, Canada
  • fYear
    2009
  • fDate
    9-11 Dec. 2009
  • Firstpage
    479
  • Lastpage
    482
  • Abstract
    Modern FPGAs are used to implement complex Systems-on-Chip (SoCs) and more recently Networks-on-Chip (NoCs). NoCs consist of computing nodes that are connected via switches or routers to a network of point-to-point links that define the topology. Previous work has investigated appropriate topology choices for ASICs as dictated by their electrical characteristics. However, since a FPGA has a prefabricated interconnect, their NoC implementations are not restricted by these concerns. Preliminary work has looked at homogeneous multiprocessor networks-on-chip on FPGAs and suggests that these systems perform differently on FPGAs than ASICs. In this work, we looked at the effects of the number of nodes, node sizes and heterogeneity on NoC performance. Assuming that the network node is not the critical path, we found that NoC performance is only dependent on the number of nodes and is not impacted by the node size or heterogeneity of nodes.
  • Keywords
    field programmable gate arrays; network-on-chip; FPGA based NoCs; field programmable gate arrays; homogeneous multiprocessor networks-on-chip; network node; network size; node heterogeneity; node size; systems-on-chip; topology; Character generation; Computer networks; Fabrics; Field programmable gate arrays; Integrated circuit interconnections; Network interfaces; Network topology; Network-on-a-chip; Routing; Switches;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Field-Programmable Technology, 2009. FPT 2009. International Conference on
  • Conference_Location
    Sydney, NSW
  • Print_ISBN
    978-1-4244-4375-8
  • Electronic_ISBN
    978-1-4244-4377-2
  • Type

    conf

  • DOI
    10.1109/FPT.2009.5377628
  • Filename
    5377628