• DocumentCode
    1682734
  • Title

    Modeling and analysis of power in multicore network processors

  • Author

    Huang, S. ; Luo, Y. ; Feng, W.

  • Author_Institution
    Dept. of Comput. Sci., Virginia Tech, Blacksburg, VA
  • fYear
    2008
  • Firstpage
    1
  • Lastpage
    8
  • Abstract
    With the emergence of multicore network processors in support of high-performance computing and networking applications, power consumption has become a problem of increasing significance. Lower-power multicore processors, such as the Intel IXP network processors, have been employed in network devices to try to address this problem; however, effective tools are still needed to assist in the exploration of the "performance versus power" design space. In this paper, we present a composite power model that simultaneously leverages three existing power-modeling tools (Cacti, Wattch, and Orion) to model the power consumption of Intel IXP2400. This model is then integrated with an open-source Intel IXP2400 NP simulator called NePSim2 and validated against its datasheet to within 5% of actual power consumption. We utilize the simulator and power model to explore how architectural parameters affect the performance and power consumption of a set of high-performance security applications, thus delivering valuable insights to chip architects for designing energy-efficient multicore network processors.
  • Keywords
    microprocessor chips; multiprocessing systems; power consumption; Intel IXP network processors; high-performance computing; multicore network processors; power consumption; Application software; Computer networks; Cryptography; Energy consumption; Multicore processing; Performance analysis; Performance evaluation; Power dissipation; Space exploration; Virtual private networks;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Parallel and Distributed Processing, 2008. IPDPS 2008. IEEE International Symposium on
  • Conference_Location
    Miami, FL
  • ISSN
    1530-2075
  • Print_ISBN
    978-1-4244-1693-6
  • Electronic_ISBN
    1530-2075
  • Type

    conf

  • DOI
    10.1109/IPDPS.2008.4536224
  • Filename
    4536224