The BubbleWrap many-core: Popping cores for sequential acceleration

Ulya R. Karpuzcu, Brian Greskamp, Josep Torrellas

Research output: Contribution to journalConference article

Abstract

Many-core scaling now faces a power wall. The gap between the number of cores that fit on a die and the number that can operate simultaneously under the power budget is rapidly increasing with technology scaling. In future designs, many of the cores may have to be dormant at any given time to meet the power budget. To push back the many-core power wall, this paper proposes Dynamic Voltage Scaling for Aging Management (DVSAM) - - a new scheme for managing processor aging to attain higher performance or lower power consumption. In addition, this paper introduces the BubbleWrap many-core, a novel architecture that makes extensive use of DVSAM. BubbleWrap identifies the most power-efficient set of cores in a variation-affected chip - - the largest set that can be simultaneously powered-on - - and designates them as Throughput cores dedicated to parallel-section execution. The rest of the cores are designated as Expendable and are dedicated to accelerating sequential sections. BubbleWrap attains maximum sequential acceleration by sacrificing Expendable cores one at a time, running them at elevated supply voltage for a significantly shorter service life each, until they completely wear-out and are discarded - - figuratively, as if popping bubbles in bubble wrap that protects Throughput cores. In simulated 32-core chips, BubbleWrap provides substantial improvements over a plain chip. For example, on average, one design runs fully-sequential applications at a 16% higher frequency, and fully-parallel ones with a 30% higher throughput.

Original languageEnglish (US)
Pages (from-to)447-458
Number of pages12
JournalProceedings of the Annual International Symposium on Microarchitecture, MICRO
DOIs
StatePublished - Dec 1 2009
Event42nd Annual IEEE/ACM International Symposium on Microarchitecture, Micro-42 - New York, NY, United States
Duration: Dec 12 2009Dec 16 2009

Fingerprint

Aging of materials
Throughput
Service life
Electric power utilization
Wear of materials
Electric potential
Voltage scaling

Keywords

  • Power wall
  • Process scaling
  • Processor aging
  • Voltage scaling

ASJC Scopus subject areas

  • Hardware and Architecture

Cite this

The BubbleWrap many-core : Popping cores for sequential acceleration. / Karpuzcu, Ulya R.; Greskamp, Brian; Torrellas, Josep.

In: Proceedings of the Annual International Symposium on Microarchitecture, MICRO, 01.12.2009, p. 447-458.

Research output: Contribution to journalConference article

@article{2bce50aff7674092920379184803bae5,
title = "The BubbleWrap many-core: Popping cores for sequential acceleration",
abstract = "Many-core scaling now faces a power wall. The gap between the number of cores that fit on a die and the number that can operate simultaneously under the power budget is rapidly increasing with technology scaling. In future designs, many of the cores may have to be dormant at any given time to meet the power budget. To push back the many-core power wall, this paper proposes Dynamic Voltage Scaling for Aging Management (DVSAM) - - a new scheme for managing processor aging to attain higher performance or lower power consumption. In addition, this paper introduces the BubbleWrap many-core, a novel architecture that makes extensive use of DVSAM. BubbleWrap identifies the most power-efficient set of cores in a variation-affected chip - - the largest set that can be simultaneously powered-on - - and designates them as Throughput cores dedicated to parallel-section execution. The rest of the cores are designated as Expendable and are dedicated to accelerating sequential sections. BubbleWrap attains maximum sequential acceleration by sacrificing Expendable cores one at a time, running them at elevated supply voltage for a significantly shorter service life each, until they completely wear-out and are discarded - - figuratively, as if popping bubbles in bubble wrap that protects Throughput cores. In simulated 32-core chips, BubbleWrap provides substantial improvements over a plain chip. For example, on average, one design runs fully-sequential applications at a 16{\%} higher frequency, and fully-parallel ones with a 30{\%} higher throughput.",
keywords = "Power wall, Process scaling, Processor aging, Voltage scaling",
author = "Karpuzcu, {Ulya R.} and Brian Greskamp and Josep Torrellas",
year = "2009",
month = "12",
day = "1",
doi = "10.1145/1669112.1669169",
language = "English (US)",
pages = "447--458",
journal = "Proceedings of the Annual International Symposium on Microarchitecture, MICRO",
issn = "1072-4451",

}

TY - JOUR

T1 - The BubbleWrap many-core

T2 - Popping cores for sequential acceleration

AU - Karpuzcu, Ulya R.

AU - Greskamp, Brian

AU - Torrellas, Josep

PY - 2009/12/1

Y1 - 2009/12/1

N2 - Many-core scaling now faces a power wall. The gap between the number of cores that fit on a die and the number that can operate simultaneously under the power budget is rapidly increasing with technology scaling. In future designs, many of the cores may have to be dormant at any given time to meet the power budget. To push back the many-core power wall, this paper proposes Dynamic Voltage Scaling for Aging Management (DVSAM) - - a new scheme for managing processor aging to attain higher performance or lower power consumption. In addition, this paper introduces the BubbleWrap many-core, a novel architecture that makes extensive use of DVSAM. BubbleWrap identifies the most power-efficient set of cores in a variation-affected chip - - the largest set that can be simultaneously powered-on - - and designates them as Throughput cores dedicated to parallel-section execution. The rest of the cores are designated as Expendable and are dedicated to accelerating sequential sections. BubbleWrap attains maximum sequential acceleration by sacrificing Expendable cores one at a time, running them at elevated supply voltage for a significantly shorter service life each, until they completely wear-out and are discarded - - figuratively, as if popping bubbles in bubble wrap that protects Throughput cores. In simulated 32-core chips, BubbleWrap provides substantial improvements over a plain chip. For example, on average, one design runs fully-sequential applications at a 16% higher frequency, and fully-parallel ones with a 30% higher throughput.

AB - Many-core scaling now faces a power wall. The gap between the number of cores that fit on a die and the number that can operate simultaneously under the power budget is rapidly increasing with technology scaling. In future designs, many of the cores may have to be dormant at any given time to meet the power budget. To push back the many-core power wall, this paper proposes Dynamic Voltage Scaling for Aging Management (DVSAM) - - a new scheme for managing processor aging to attain higher performance or lower power consumption. In addition, this paper introduces the BubbleWrap many-core, a novel architecture that makes extensive use of DVSAM. BubbleWrap identifies the most power-efficient set of cores in a variation-affected chip - - the largest set that can be simultaneously powered-on - - and designates them as Throughput cores dedicated to parallel-section execution. The rest of the cores are designated as Expendable and are dedicated to accelerating sequential sections. BubbleWrap attains maximum sequential acceleration by sacrificing Expendable cores one at a time, running them at elevated supply voltage for a significantly shorter service life each, until they completely wear-out and are discarded - - figuratively, as if popping bubbles in bubble wrap that protects Throughput cores. In simulated 32-core chips, BubbleWrap provides substantial improvements over a plain chip. For example, on average, one design runs fully-sequential applications at a 16% higher frequency, and fully-parallel ones with a 30% higher throughput.

KW - Power wall

KW - Process scaling

KW - Processor aging

KW - Voltage scaling

UR - http://www.scopus.com/inward/record.url?scp=76749156251&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=76749156251&partnerID=8YFLogxK

U2 - 10.1145/1669112.1669169

DO - 10.1145/1669112.1669169

M3 - Conference article

AN - SCOPUS:76749156251

SP - 447

EP - 458

JO - Proceedings of the Annual International Symposium on Microarchitecture, MICRO

JF - Proceedings of the Annual International Symposium on Microarchitecture, MICRO

SN - 1072-4451

ER -