Dynamic Power Management for Dark Silicon Multicore Processors

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Multicore processors in the dark silicon era will be over-provisioned with multiple cores of different types and will have to satisfy stringent power constraints while maximizing performance. As such, a critical research problem is that of developing new, effective, and scalable run-time dynamic power management techniques and algorithms to effectively make use of the over-provisioned but power-constrained computational resources on a chip. This chapter surveys the state of the art on this subject and dives in-depth into three specific dynamic power management solutions for multicore processors in the dark silicon era.

Original languageEnglish (US)
Title of host publicationAdvances in Computers
EditorsAli R. Hurson, Hamid Sarbazi-Azad
PublisherAcademic Press Inc.
Pages171-216
Number of pages46
ISBN (Print)9780128153581
DOIs
StatePublished - Jan 1 2018

Publication series

NameAdvances in Computers
Volume110
ISSN (Print)0065-2458

Fingerprint

Silicon
Power management

Keywords

  • Dark silicon
  • Dynamic power management
  • Voltage and frequency scaling

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Garg, S. (2018). Dynamic Power Management for Dark Silicon Multicore Processors. In A. R. Hurson, & H. Sarbazi-Azad (Eds.), Advances in Computers (pp. 171-216). (Advances in Computers; Vol. 110). Academic Press Inc.. https://doi.org/10.1016/bs.adcom.2018.03.016

Dynamic Power Management for Dark Silicon Multicore Processors. / Garg, Siddharth.

Advances in Computers. ed. / Ali R. Hurson; Hamid Sarbazi-Azad. Academic Press Inc., 2018. p. 171-216 (Advances in Computers; Vol. 110).

Research output: Chapter in Book/Report/Conference proceedingChapter

Garg, S 2018, Dynamic Power Management for Dark Silicon Multicore Processors. in AR Hurson & H Sarbazi-Azad (eds), Advances in Computers. Advances in Computers, vol. 110, Academic Press Inc., pp. 171-216. https://doi.org/10.1016/bs.adcom.2018.03.016
Garg S. Dynamic Power Management for Dark Silicon Multicore Processors. In Hurson AR, Sarbazi-Azad H, editors, Advances in Computers. Academic Press Inc. 2018. p. 171-216. (Advances in Computers). https://doi.org/10.1016/bs.adcom.2018.03.016
Garg, Siddharth. / Dynamic Power Management for Dark Silicon Multicore Processors. Advances in Computers. editor / Ali R. Hurson ; Hamid Sarbazi-Azad. Academic Press Inc., 2018. pp. 171-216 (Advances in Computers).
@inbook{ceec62894bd446e498d66a032a1d2a62,
title = "Dynamic Power Management for Dark Silicon Multicore Processors",
abstract = "Multicore processors in the dark silicon era will be over-provisioned with multiple cores of different types and will have to satisfy stringent power constraints while maximizing performance. As such, a critical research problem is that of developing new, effective, and scalable run-time dynamic power management techniques and algorithms to effectively make use of the over-provisioned but power-constrained computational resources on a chip. This chapter surveys the state of the art on this subject and dives in-depth into three specific dynamic power management solutions for multicore processors in the dark silicon era.",
keywords = "Dark silicon, Dynamic power management, Voltage and frequency scaling",
author = "Siddharth Garg",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/bs.adcom.2018.03.016",
language = "English (US)",
isbn = "9780128153581",
series = "Advances in Computers",
publisher = "Academic Press Inc.",
pages = "171--216",
editor = "Hurson, {Ali R.} and Hamid Sarbazi-Azad",
booktitle = "Advances in Computers",
address = "United States",

}

TY - CHAP

T1 - Dynamic Power Management for Dark Silicon Multicore Processors

AU - Garg, Siddharth

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Multicore processors in the dark silicon era will be over-provisioned with multiple cores of different types and will have to satisfy stringent power constraints while maximizing performance. As such, a critical research problem is that of developing new, effective, and scalable run-time dynamic power management techniques and algorithms to effectively make use of the over-provisioned but power-constrained computational resources on a chip. This chapter surveys the state of the art on this subject and dives in-depth into three specific dynamic power management solutions for multicore processors in the dark silicon era.

AB - Multicore processors in the dark silicon era will be over-provisioned with multiple cores of different types and will have to satisfy stringent power constraints while maximizing performance. As such, a critical research problem is that of developing new, effective, and scalable run-time dynamic power management techniques and algorithms to effectively make use of the over-provisioned but power-constrained computational resources on a chip. This chapter surveys the state of the art on this subject and dives in-depth into three specific dynamic power management solutions for multicore processors in the dark silicon era.

KW - Dark silicon

KW - Dynamic power management

KW - Voltage and frequency scaling

UR - http://www.scopus.com/inward/record.url?scp=85047351759&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85047351759&partnerID=8YFLogxK

U2 - 10.1016/bs.adcom.2018.03.016

DO - 10.1016/bs.adcom.2018.03.016

M3 - Chapter

SN - 9780128153581

T3 - Advances in Computers

SP - 171

EP - 216

BT - Advances in Computers

A2 - Hurson, Ali R.

A2 - Sarbazi-Azad, Hamid

PB - Academic Press Inc.

ER -