PowMon: Accurate and Stable CPU Power Estimation

PowMon Power Modelling Tools

Being able to accurately estimate CPU power consumption is a key requirement for both controlling online CPU energy-saving techniques and design-space exploration. Models built and validated using measured data from an actual device are extremely valuable as their accuracy is known and trusted. PowMon produces power models that are both accurate and stable.

The open source PowMon software tools consists of a number of distinct parts:

  1. Software tools which allow researchers to implement our automated model building methodology, and produce power models of a platform that are both accurate and stable.
  2. Downloadable power models for existing mobile CPUs (quad-core Cortex-A7 and quad-core Cortex-A15), that have been created using our methodology, and which can be used directly in situations where an accurate reference model is required.
  3. Obtaining accurate data from mobile devices can be challenging and more time-consuming that using a simulator or desktop/server devices. For this reason, we make available our experimental platform software tools which allows workloads to be automatically run on a mobile device and Performance Monitoring Counters (PMCs), temperature, CPU utilisation, CPU power and CPU voltage to be collected.

More details of our methodology can be found in the following publications:

Walker, Matthew, Diestelhorst, Stephan, Hansson, Andreas, Das, Anup, Yang, Sheng, Al-Hashimi, Bashir M. and Merrett, Geoff V. (2016) Accurate and stable run-time power modeling for mobile and embedded CPUs. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 1-14.

For more information, and to download the tool, visit www.powmon.ecs.soton.ac.uk.

Related Projects

Related Publications

Drag

Related News

Related Theses