Energy-Efficient Mapping and Scheduling for DVS Enabled Distributed Embedded Systems

  • Authors:
  • M. Schmitz;B. Al-Hashimi;P. Eles

  • Affiliations:
  • Dept. of Electronics and Computer Science, University of Southampton, Southampton, SO17 1BJ, United Kingdom;Dept. of Electronics and Computer Science, University of Southampton, Southampton, SO17 1BJ, United Kingdom;Dept. of Computer and Information Science, Linköping University, S-58183 Linköping, Sweden

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

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we present an efficient two-step iterative synthesisapproach for distributed embedded systems containing dynamicvoltage scalable processing elements (DVS-PEs), basedon genetic algorithms. The approach partitions, schedules, andvoltage scales multi-rate specifications given as task graphswith multiple deadlines. A distinguishing feature of the proposedsynthesis is the utilisation of a generalised DVS method.In contrast to previous techniques, which "simply" exploitavailable slack time, this generalised technique additionallyconsiders the PE power profile during a refined voltage selectionto further increase the energy savings. Extensive experimentsare conducted to demonstrate the efficiency of the proposedapproach. We report up to 43.2% higher energy reductionscompared to previous DVS scheduling approaches basedon constructive techniques and total energy savings of up to82.9% for mapping and scheduling optimised DVS systems.