Adaptive online heuristic performance estimation and power optimization for reconfigurable embedded systems

  • Authors:
  • Jingqing Mu;Roman Lysecky

  • Affiliations:
  • University of Arizona, Tucson, AZ, USA;University of Arizona, Tucson, AZ, USA

  • Venue:
  • Proceedings of the eighth IEEE/ACM/IFIP international conference on Hardware/software codesign and system synthesis
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

For dynamically adaptable systems, accurate online estimation methods can significantly improve the performance and power consumption of using reconfigurable hardware coprocessors implemented within an FPGA. Complex interactions between multiple application tasks, non-deterministic execution behavior in response to varying system inputs, and effects of operating system scheduling introduce significant challenges. We present an adaptive online performance and power estimation framework and heuristic power optimization method that monitors and adapts to dynamically changing application behavior. We further demonstrate the power savings that can be achieved using this approach for dynamic optimization of multitasked applications.