Skip to main content

TS

Koala: A platform for OS-level power management

Authors

Dave Snowdon, Etienne Le Sueur, Stefan Petters and Gernot Heiser

NICTA

UNSW

Open Kernel Labs

Abstract

Managing the power consumption of computing platforms is a prohibitively complicated problem thanks to a multitude of hardware configuration options and characteristics. Much of the academic research is based on unrealistic assumptions, and has, therefore, seen little practical uptake. We present an overview of the difficulties facing power management schemes when used in real systems.

We propose Koala, a platform which uses a pre-characterised model at run-time to predict the performance and energy consumption of a piece of software. An arbitrary policy can then be applied in order to dynamically trade performance and energy consumption. We have implemented this system in a recent Linux kernel, and evaluated it by running a variety of benchmarks on a number of different platforms. We observe energy savings of, at times, 30% for a 4% performance loss.

BibTeX Entry

  @inproceedings{Snowdon_SPH_09,
    author           = {Snowdon, Dave and Le Sueur, Etienne and Petters, Stefan and Heiser, Gernot},
    month            = {apr},
    editor           = {{John Wilkes}},
    year             = {2009},
    keywords         = {energy, dvfs},
    title            = {Koala: A Platform for {OS}-Level Power Management},
    booktitle        = {Eurosys 2009},
    pages            = {289-302},
    address          = {Nuremburg, DE}
  }

Download