A Dynamic Voltage Scaling Algorithm for Dynamic-Priority Hard Real-Time Systems Using Slack Time Analysis

  • Authors:
  • W. Kim;J. Kim;S. Min

  • Affiliations:
  • School of Computer Science and Engineering, Seoul National University ENG4190, Seoul, Korea;-;School of Computer Science and Engineering, Seoul National University ENG4190, Seoul, Korea

  • Venue:
  • Proceedings of the conference on Design, automation and test in Europe
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

Dynamic voltage scaling (DVS), which adjusts the clockspeed and supply voltage dynamically, is an effective techniquein reducing the energy consumption of embedded real-timesystems. The energy efficiency of a DVS algorithm largelydepends on the performance of the slack estimation methodused in it. In this paper, we propose a novel DVS algorithmfor periodic hard real-time tasks based on an improved slackestimation algorithm. Unlike the existing techniques, the proposedmethod takes full advantage of the periodic characteristicsof the real-time tasks under priority-driven schedulingsuch as EDF. Experimental results show that the proposed algorithmreduces the energy consumption by 20~40% over theexisting DVS algorithm. The experiment results also show thatour algorithm based on the improved slack estimation methodgives comparable energy savings to the DVS algorithm basedon the theoretically optimal (but impractical) slack estimationmethod.