• DocumentCode
    2052898
  • Title

    Improving the performance of thermally constrained multi core processors using DTM techniques

  • Author

    Lawrence, W.S. ; Kumar, P. Roshan

  • Author_Institution
    K.S. Rangasamy Coll. of Technol., Tiruchengode, India
  • fYear
    2013
  • fDate
    21-22 Feb. 2013
  • Firstpage
    1035
  • Lastpage
    1040
  • Abstract
    Extracting high performance from multi-core processors requires increased use of dynamic thermal management techniques. This paper presents performance optimal online thermal management techniques for multi-core processors. The techniques include dynamic voltage and frequency scaling and task-to-core allocation or task migration. The problem formulation includes accurate power and thermal models, as well as leakage dependence on temperature. The effectiveness of our DVFS and task-to-core allocation techniques are demonstrated by numerical simulations. The proposed task to-core allocation method showed a 20.2% improvement in performance over a power-based thread migration approach.
  • Keywords
    multiprocessing systems; power aware computing; DTM techniques; DVFS; dynamic thermal management techniques; dynamic voltage; frequency scaling; leakage dependence; power-based thread migration approach; task migration; task-to-core allocation techniques; thermally constrained multi core processors; Computational modeling; Multicore processing; Power demand; Program processors; Resource management; Thermal management; Threshold voltage; dynamic voltage and frequency scaling; leakage dependence on temperature; makespan minimization; multi-core; performance optimization; task migration;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Information Communication and Embedded Systems (ICICES), 2013 International Conference on
  • Conference_Location
    Chennai
  • Print_ISBN
    978-1-4673-5786-9
  • Type

    conf

  • DOI
    10.1109/ICICES.2013.6508292
  • Filename
    6508292