AdBoost: Thermal Aware Performance Boosting Through Dark Silicon Patterning

Anil Kanduri, Mohammad Hashem Haghbayan, Amir M. Rahmani, Muhammad Shafique, Axel Jantsch, Pasi Liljeberg

Research output: Contribution to journalArticlepeer-review

Abstract

Increasing power densities of many-core systems leaves a fraction of on-chip resources inactive, referred to as dark silicon. Efficient management of critical interlinked parameters-power, performance and temperature can improve resource utilization and mitigate dark silicon. In this paper, we present a run-time resource management system for thermal aware performance boosting using a dark silicon aware run-time application mapping strategy. The mapping policy patterns inactive cores among active cores for relatively lower and even distribution of operating temperatures. This provides enough thermal headroom for boosting the frequency of active cores upon performance surges and allows sustained boosting periods, improving the performance further. We design a controller for thermal aware performance boosting that decides on efficient allocation utilization of power budget and thermal headroom obtained from patterning. Our strategy yields up to 37 percent better throughput, 29 percent lower waiting time and up to 2 × longer boosting periods, in comparison with other state-of-the-art run-time mapping policies.

Original languageEnglish (US)
Pages (from-to)1062-1077
Number of pages16
JournalIEEE Transactions on Computers
Volume67
Issue number8
DOIs
StatePublished - Aug 1 2018

Keywords

  • Dark silicon
  • dynamic power and thermal management
  • run-time mapping

ASJC Scopus subject areas

  • Software
  • Theoretical Computer Science
  • Hardware and Architecture
  • Computational Theory and Mathematics

Fingerprint

Dive into the research topics of 'AdBoost: Thermal Aware Performance Boosting Through Dark Silicon Patterning'. Together they form a unique fingerprint.

Cite this