• DocumentCode
    2255356
  • Title

    LA-XYZ: Low Latency, High Throughput Look-Ahead Routing Algorithm for 3D Network-on-Chip (3D-NoC) Architecture

  • Author

    Ahmed, Achraf ; Abdallah, Abderazek

  • Author_Institution
    Grad. Sch. of Comput. Sci. & Eng., Univ. of Aizu, Aizu-Wakamatsu, Japan
  • fYear
    2012
  • fDate
    20-22 Sept. 2012
  • Firstpage
    167
  • Lastpage
    174
  • Abstract
    Despite the higher scalability and parallelism integration offered by 2D-Network-on-Chip (NoC) over the traditional shared-bus based systems, it is still not an ideal solution for future large scale Systems-on-Chip (SoCs). Recently, merging NoC to the third dimension (3D-NoC) has been proposed as a promising solution offering lower power consumption and higher speed. One of the most important design choices for 3D-NoC implementation is the routing algorithm, as it controls the path decision that a flit has tofollow while traveling along the network. This has a direct impact on the overall system performance. In this paper, we present an efficient routing algorithm for 3D-NoC named Look-Ahead-XYZ (LA-XYZ). This algorithm aims to minimize the communication latency and power consumption while enhancing the system throughput. Comparison results with systems adopting two dimensional routing showed that, using JPEG encoder and Matrix applications, LA-XYZ reduces the communication latency with up to 44.9% and enhances the throughput that can reach the 45.3% while observing an average 15.9% reduction in terms of dynamic power.
  • Keywords
    matrix multiplication; network routing; network-on-chip; three-dimensional integrated circuits; 2D-network-on-chip; 3D network-on-chip architecture; 3D-NoC architecture; JPEG encoder; LA-XYZ routing algorithm; communication latency minimization; look-ahead-XYZ routing algorithm; matrix multiplication application; parallelism integration; path decision control; power consumption minimization; scalability; shared-bus based systems; throughput enhancement; Algorithm design and analysis; Computer architecture; Pipelines; Power demand; Routing; Switches; System performance; 3D NoC; Concurrent; Look-Ahead Routing;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Embedded Multicore Socs (MCSoC), 2012 IEEE 6th International Symposium on
  • Conference_Location
    Aizu-Wakamatsu
  • Print_ISBN
    978-1-4673-2535-6
  • Electronic_ISBN
    978-0-7695-4800-5
  • Type

    conf

  • DOI
    10.1109/MCSoC.2012.24
  • Filename
    6354695