Task allocation for minimum system power in a homogenous multi-core processor

Yang Ge, Qinru Qiu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

11 Scopus citations

Abstract

In this paper we address the impact of task allocation to the system power consumption of a homogenous multi-core processor with a main focus on its impact on the leakage power and fan power. Although the leakage power is determined by the average die temperature and the fan power is determined by the peak temperature, our analysis shows that the overall power can be minimized if a task allocation with minimum peak temperature is adopted together with an intelligent fan speed adjustment technique that finds the optimal tradeoff between fan power and leakage power. We further propose a multi-agent distributed task migration technique that searches for the best task allocation during runtime. By choosing only those migration requests that will result chip maximum temperature reduction, the proposed framework achieves large fan power savings as well as overall power reduction. Experimental results show that, our agent-based distributed task migration policy can save up to 37.2% fan power and 17.9% system overall power compared to the random mapping policy when the temperature constraint is tight. When the temperature constraint is loose, the overall system power is insensitive to the task allocation.

Original languageEnglish (US)
Title of host publication2010 International Conference on Green Computing, Green Comp 2010
Pages299-306
Number of pages8
DOIs
StatePublished - 2010
Event2010 International Conference on Green Computing, Green Comp 2010 - Chicago, IL, United States
Duration: Aug 15 2010Aug 18 2010

Publication series

Name2010 International Conference on Green Computing, Green Comp 2010

Other

Other2010 International Conference on Green Computing, Green Comp 2010
CountryUnited States
CityChicago, IL
Period8/15/108/18/10

Keywords

  • Low power
  • Multiagent distributed framework
  • Task allocation
  • Thermal aware

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Computer Science Applications

Fingerprint Dive into the research topics of 'Task allocation for minimum system power in a homogenous multi-core processor'. Together they form a unique fingerprint.

  • Cite this

    Ge, Y., & Qiu, Q. (2010). Task allocation for minimum system power in a homogenous multi-core processor. In 2010 International Conference on Green Computing, Green Comp 2010 (pp. 299-306). [5598299] (2010 International Conference on Green Computing, Green Comp 2010). https://doi.org/10.1109/GREENCOMP.2010.5598299