Bristol Ridge: A 28-nm x 86 Performance-Enhanced Microprocessor Through System Power Management

Miguel Rodriguez, Sriram Sundaram, Aaron Grenat, Samuel Naffziger, Tom Burd, Stephen Kosonocky, Steve Liepe, Ravinder Rachala, Michael Austin, Sriram Sambamurthy

Research output: Contribution to journalArticlepeer-review

5 Scopus Citations

Abstract

Power management techniques can be effective at extracting more performance and energy efficiency out of mature systems on chip (SoCs). For instance, the peak performance of microprocessors is often limited by worst case technology (Vmax), infrastructure (thermal/electrical), and microprocessor usage assumptions. Performance/watt of microprocessors also typically suffers from guard bands associated with the test and binning processes as well as worst case aging/lifetime degradation. Similarly, on multicore processors, shared voltage rails tend to limit the peak performance achievable in low thread count workloads. In this paper, we describe five power management techniques that maximize the per-part performance under the before-mentioned constraints. Using these techniques, we demonstrate a net performance increase of up to 15% depending on the application and TDP of the SoC, implemented on 'Bristol Ridge,' a 28-nm CMOS, dual-core x 86 accelerated processing unit.
Original languageAmerican English
Pages (from-to)89-97
Number of pages9
JournalIEEE Journal of Solid-State Circuits
Volume52
Issue number1
DOIs
StatePublished - 2017

NREL Publication Number

  • NREL/JA-5D00-71267

Keywords

  • accelerated processing unit (APU)
  • system on chip (SoC)
  • system power management
  • thermal design power
  • Vmax reliability margin

Fingerprint

Dive into the research topics of 'Bristol Ridge: A 28-nm x 86 Performance-Enhanced Microprocessor Through System Power Management'. Together they form a unique fingerprint.

Cite this