Dynamic Voltage Scheduling Using Adaptive Filtering of Workload Traces

  • Authors:
  • Amit Sinha;Anantha P. Chandrakasan

  • Affiliations:
  • -;-

  • Venue:
  • VLSID '01 Proceedings of the The 14th International Conference on VLSI Design (VLSID '01)
  • Year:
  • 2001

Quantified Score

Hi-index 0.01

Visualization

Abstract

An adaptive approach for dynamic voltage scheduling on processors is presented based on workload prediction by filtering a trace history. The effects of update frequency and filtering strategy on the energy savings is analyzed. A performance hit metric is defined and techniques to minimize energy under a given performance requirement are outlined. Our results demonstrate that up to two orders of magnitude energy savings is possible with dynamic voltage scheduling depending on workload statistics.