Multicore power proxies using least-angle regression

Rupesh Raj Karn, Ibrahim Abe M. Elfadel

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations


The use of performance counters (PCs) to develop per-core power proxies for multicore processors is now well established. These proxies are typically obtained using traditional linear regression techniques. These techniques have the disadvantage of requiring the full PC set regardless of the workload run by the multicore processor. Typically a computationally expensive principal component analysis is conducted to find the PCs most correlated with each workload. In this paper, we use the more recent algorithm of least-angle regression to efficiently develop power proxies that include only PCs most relevant to the workload. Such PCs can be considered workload signatures and used to categorize the workload and to trigger specific power management action. Our new power proxies are trained and tested on workloads from the PARSEC and SPEC CPU 2006 benchmarks with an average error of less than 3%.

Original languageBritish English
Title of host publication2015 IEEE International Symposium on Circuits and Systems, ISCAS 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages4
ISBN (Electronic)9781479983919
StatePublished - 27 Jul 2015
EventIEEE International Symposium on Circuits and Systems, ISCAS 2015 - Lisbon, Portugal
Duration: 24 May 201527 May 2015

Publication series

NameProceedings - IEEE International Symposium on Circuits and Systems
ISSN (Print)0271-4310


ConferenceIEEE International Symposium on Circuits and Systems, ISCAS 2015


  • Core
  • Correlation
  • DVFS(Dynamic Voltage Frequency Scaling)
  • Modeling
  • Multicore
  • Power
  • Regression


Dive into the research topics of 'Multicore power proxies using least-angle regression'. Together they form a unique fingerprint.

Cite this