Leveraging process variation for performance and energy: In the perspective of overclocking

Hyung Beom Jang, Junhee Lee, Joonho Kong, Taeweon Suh, Sung Woo Jung

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Process variation is one of the most important factors to be considered in recent microprocessor design, since it negatively affects performance, power, and yield of microprocessors. However, by leveraging process variation, overclocking techniques can improve performance. As microprocessors have substantial clock cycle time margin for yield, there is enough room for performance improvement by overclocking techniques. In this paper, we adopt the F-overclocking technique, which increases clock frequency without changing supply voltage. Our experimental results show that the F-overclocking technique significantly improves performance as well as energy consumption. In addition, the F-overclocking technique is superior to the conventional overclocking technique which increases clock frequency and supply voltage together in the perspective of energy efficiency and reliability, showing similar performance improvement. Furthermore, we propose an adaptive overclocking controller which dynamically applies the F-overclocking technique based on the application characteristics. By adopting our adaptive overclocking controller, we further minimize the reliability loss caused by the F-overclocking technique.

Original languageEnglish
Article number6374616
Pages (from-to)1316-1322
Number of pages7
JournalIEEE Transactions on Computers
Volume63
Issue number5
DOIs
Publication statusPublished - 2014 Jan 1

Fingerprint

Process Variation
Microprocessor chips
Clocks
Energy
Controllers
Microprocessor
Electric potential
Energy efficiency
Energy utilization
Voltage
Controller
Energy Efficiency
Margin
Energy Consumption
Minimise
Experimental Results

Keywords

  • energy efficiency
  • Overclocking
  • performance
  • process variation
  • reliability

ASJC Scopus subject areas

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

Cite this

Leveraging process variation for performance and energy : In the perspective of overclocking. / Jang, Hyung Beom; Lee, Junhee; Kong, Joonho; Suh, Taeweon; Jung, Sung Woo.

In: IEEE Transactions on Computers, Vol. 63, No. 5, 6374616, 01.01.2014, p. 1316-1322.

Research output: Contribution to journalArticle

@article{2aa8a7355dd94c3abfd14ec6ceea3abe,
title = "Leveraging process variation for performance and energy: In the perspective of overclocking",
abstract = "Process variation is one of the most important factors to be considered in recent microprocessor design, since it negatively affects performance, power, and yield of microprocessors. However, by leveraging process variation, overclocking techniques can improve performance. As microprocessors have substantial clock cycle time margin for yield, there is enough room for performance improvement by overclocking techniques. In this paper, we adopt the F-overclocking technique, which increases clock frequency without changing supply voltage. Our experimental results show that the F-overclocking technique significantly improves performance as well as energy consumption. In addition, the F-overclocking technique is superior to the conventional overclocking technique which increases clock frequency and supply voltage together in the perspective of energy efficiency and reliability, showing similar performance improvement. Furthermore, we propose an adaptive overclocking controller which dynamically applies the F-overclocking technique based on the application characteristics. By adopting our adaptive overclocking controller, we further minimize the reliability loss caused by the F-overclocking technique.",
keywords = "energy efficiency, Overclocking, performance, process variation, reliability",
author = "Jang, {Hyung Beom} and Junhee Lee and Joonho Kong and Taeweon Suh and Jung, {Sung Woo}",
year = "2014",
month = "1",
day = "1",
doi = "10.1109/TC.2012.286",
language = "English",
volume = "63",
pages = "1316--1322",
journal = "IEEE Transactions on Computers",
issn = "0018-9340",
publisher = "IEEE Computer Society",
number = "5",

}

TY - JOUR

T1 - Leveraging process variation for performance and energy

T2 - In the perspective of overclocking

AU - Jang, Hyung Beom

AU - Lee, Junhee

AU - Kong, Joonho

AU - Suh, Taeweon

AU - Jung, Sung Woo

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Process variation is one of the most important factors to be considered in recent microprocessor design, since it negatively affects performance, power, and yield of microprocessors. However, by leveraging process variation, overclocking techniques can improve performance. As microprocessors have substantial clock cycle time margin for yield, there is enough room for performance improvement by overclocking techniques. In this paper, we adopt the F-overclocking technique, which increases clock frequency without changing supply voltage. Our experimental results show that the F-overclocking technique significantly improves performance as well as energy consumption. In addition, the F-overclocking technique is superior to the conventional overclocking technique which increases clock frequency and supply voltage together in the perspective of energy efficiency and reliability, showing similar performance improvement. Furthermore, we propose an adaptive overclocking controller which dynamically applies the F-overclocking technique based on the application characteristics. By adopting our adaptive overclocking controller, we further minimize the reliability loss caused by the F-overclocking technique.

AB - Process variation is one of the most important factors to be considered in recent microprocessor design, since it negatively affects performance, power, and yield of microprocessors. However, by leveraging process variation, overclocking techniques can improve performance. As microprocessors have substantial clock cycle time margin for yield, there is enough room for performance improvement by overclocking techniques. In this paper, we adopt the F-overclocking technique, which increases clock frequency without changing supply voltage. Our experimental results show that the F-overclocking technique significantly improves performance as well as energy consumption. In addition, the F-overclocking technique is superior to the conventional overclocking technique which increases clock frequency and supply voltage together in the perspective of energy efficiency and reliability, showing similar performance improvement. Furthermore, we propose an adaptive overclocking controller which dynamically applies the F-overclocking technique based on the application characteristics. By adopting our adaptive overclocking controller, we further minimize the reliability loss caused by the F-overclocking technique.

KW - energy efficiency

KW - Overclocking

KW - performance

KW - process variation

KW - reliability

UR - http://www.scopus.com/inward/record.url?scp=84901036190&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901036190&partnerID=8YFLogxK

U2 - 10.1109/TC.2012.286

DO - 10.1109/TC.2012.286

M3 - Article

AN - SCOPUS:84901036190

VL - 63

SP - 1316

EP - 1322

JO - IEEE Transactions on Computers

JF - IEEE Transactions on Computers

SN - 0018-9340

IS - 5

M1 - 6374616

ER -