Fast and accurate prediction of the steady-state throughput of multicore processors under thermal constraints

Ravishankar Rao, Sarma Vrudhula

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

This paper describes a fast and accurate technique to predict the steady-state throughput and the corresponding power consumption of a homogeneous multicore processor for a given benchmark workload while accounting for speed reduction due to thermal constraints. The expressions contain several parameters of interest to a system designer, like the static and dynamic-power consumptions (for hottest block and for full chip), the vertical thermal resistance of the hottest block, the leakage sensitivity to temperature, the chip threshold temperature, the ambient temperature, etc. Their computational complexity is independent of the number of cores. These are incorporated in a system-level multicore power/thermal simulator that uses the PTScalar power model and the Hotspot thermal model. The analytical throughput and power predictions were within 1.7% of that predicted by the system-level simulator. However, the analytical technique takes less than 0.2 s for a given set of design parameters, making it well suited for early design-space exploration. In contrast, the numerical technique takes anywhere from a minute (for 4 cores) up to a few hours (for 25 cores).

Original languageEnglish (US)
Pages (from-to)1559-1572
Number of pages14
JournalIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Volume28
Issue number1
StatePublished - Jan 2009

Fingerprint

Throughput
Electric power utilization
Simulators
Heat resistance
Temperature
Computational complexity
Hot Temperature

Keywords

  • Dynamic thermal management (DTM)
  • Frequency scaling
  • Multicore processors
  • Power management
  • Throughput optimization

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Graphics and Computer-Aided Design
  • Software

Cite this

@article{e54857ea9a074cceb30a9c6a66e3637b,
title = "Fast and accurate prediction of the steady-state throughput of multicore processors under thermal constraints",
abstract = "This paper describes a fast and accurate technique to predict the steady-state throughput and the corresponding power consumption of a homogeneous multicore processor for a given benchmark workload while accounting for speed reduction due to thermal constraints. The expressions contain several parameters of interest to a system designer, like the static and dynamic-power consumptions (for hottest block and for full chip), the vertical thermal resistance of the hottest block, the leakage sensitivity to temperature, the chip threshold temperature, the ambient temperature, etc. Their computational complexity is independent of the number of cores. These are incorporated in a system-level multicore power/thermal simulator that uses the PTScalar power model and the Hotspot thermal model. The analytical throughput and power predictions were within 1.7{\%} of that predicted by the system-level simulator. However, the analytical technique takes less than 0.2 s for a given set of design parameters, making it well suited for early design-space exploration. In contrast, the numerical technique takes anywhere from a minute (for 4 cores) up to a few hours (for 25 cores).",
keywords = "Dynamic thermal management (DTM), Frequency scaling, Multicore processors, Power management, Throughput optimization",
author = "Ravishankar Rao and Sarma Vrudhula",
year = "2009",
month = "1",
language = "English (US)",
volume = "28",
pages = "1559--1572",
journal = "IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems",
issn = "0278-0070",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "1",

}

TY - JOUR

T1 - Fast and accurate prediction of the steady-state throughput of multicore processors under thermal constraints

AU - Rao, Ravishankar

AU - Vrudhula, Sarma

PY - 2009/1

Y1 - 2009/1

N2 - This paper describes a fast and accurate technique to predict the steady-state throughput and the corresponding power consumption of a homogeneous multicore processor for a given benchmark workload while accounting for speed reduction due to thermal constraints. The expressions contain several parameters of interest to a system designer, like the static and dynamic-power consumptions (for hottest block and for full chip), the vertical thermal resistance of the hottest block, the leakage sensitivity to temperature, the chip threshold temperature, the ambient temperature, etc. Their computational complexity is independent of the number of cores. These are incorporated in a system-level multicore power/thermal simulator that uses the PTScalar power model and the Hotspot thermal model. The analytical throughput and power predictions were within 1.7% of that predicted by the system-level simulator. However, the analytical technique takes less than 0.2 s for a given set of design parameters, making it well suited for early design-space exploration. In contrast, the numerical technique takes anywhere from a minute (for 4 cores) up to a few hours (for 25 cores).

AB - This paper describes a fast and accurate technique to predict the steady-state throughput and the corresponding power consumption of a homogeneous multicore processor for a given benchmark workload while accounting for speed reduction due to thermal constraints. The expressions contain several parameters of interest to a system designer, like the static and dynamic-power consumptions (for hottest block and for full chip), the vertical thermal resistance of the hottest block, the leakage sensitivity to temperature, the chip threshold temperature, the ambient temperature, etc. Their computational complexity is independent of the number of cores. These are incorporated in a system-level multicore power/thermal simulator that uses the PTScalar power model and the Hotspot thermal model. The analytical throughput and power predictions were within 1.7% of that predicted by the system-level simulator. However, the analytical technique takes less than 0.2 s for a given set of design parameters, making it well suited for early design-space exploration. In contrast, the numerical technique takes anywhere from a minute (for 4 cores) up to a few hours (for 25 cores).

KW - Dynamic thermal management (DTM)

KW - Frequency scaling

KW - Multicore processors

KW - Power management

KW - Throughput optimization

UR - http://www.scopus.com/inward/record.url?scp=77955205881&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955205881&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:77955205881

VL - 28

SP - 1559

EP - 1572

JO - IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems

JF - IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems

SN - 0278-0070

IS - 1

ER -