• DocumentCode
    2343282
  • Title

    Leveraging torus topology with deadlock recovery for cost-efficient on-chip network

  • Author

    Shin, Minjeong ; Kim, John

  • Author_Institution
    Dept. of Comput. Sci., KAIST, Daejeon, South Korea
  • fYear
    2011
  • fDate
    9-12 Oct. 2011
  • Firstpage
    25
  • Lastpage
    30
  • Abstract
    On-chip networks are becoming more important as the number of on-chip components continue to increase. 2D mesh topology is a commonly assumed topology for on-chip networks but in this work, we make the argument that 2D torus can provide a more cost-efficient on-chip network since the on-chip network datapath is reduced by 2× while providing the same bisection bandwidth as a mesh network. Our results show that 2D torus can achieve an improvement of up to 1.9× over a 2D mesh in performance per watt metric. However, routing deadlock can occur in a torus network with the wrap-around channel and requires additional virtual channels for deadlock avoidance. In this work, we propose deadlock recovery with tokens (DRT) in on-chip networks that exploits on-chip networks - exploiting the abundant wires available while minimizing the need for additional buffers. As a result, deadlocks can be exactly detected without having to rely on a timeout mechanism and when needed, recover from the deadlock. We show how DRT results in minimal loss in performance, compared with deadlock avoidance using virtual channels, while reducing the on-chip network complexity.
  • Keywords
    microprocessor chips; multiprocessing systems; network topology; 2D mesh topology; chip multiprocessor system; cost-efficient on-chip network; deadlock avoidance; deadlock recovery with tokens; on-chip network datapath; routing deadlock; torus topology; virtual channels; Bandwidth; Network topology; Routing; System recovery; System-on-a-chip; Token networks; Topology;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Design (ICCD), 2011 IEEE 29th International Conference on
  • Conference_Location
    Amherst, MA
  • ISSN
    1063-6404
  • Print_ISBN
    978-1-4577-1953-0
  • Type

    conf

  • DOI
    10.1109/ICCD.2011.6081371
  • Filename
    6081371