Techniques and tools for measuring energy efficiency of scientific software applications

David Abdurachmanov, Peter Elmer, Giulio Eulisse, Robert Knight, Tapio Niemi, Jukka K. Nurminen, Filip Nyback, Gonçalo Pestana, Zhonghong Ou, Kashif Khan

Research output: Contribution to journalArticleScientificpeer-review

5 Citations (Scopus)

Abstract

The scale of scientific High Performance Computing (HPC) and High Throughput Computing (HTC) has increased significantly in recent years, and is becoming sensitive to total energy use and cost. Energy-efficiency has thus become an important concern in scientific fields such as High Energy Physics (HEP). There has been a growing interest in utilizing alternate architectures, such as low power ARM processors, to replace traditional Intel x86 architectures. Nevertheless, even though such solutions have been successfully used in mobile applications with low I/O and memory demands, it is unclear if they are suitable and more energy-efficient in the scientific computing environment. Furthermore, there is a lack of tools and experience to derive and compare power consumption between the architectures for various workloads, and eventually to support software optimizations for energy efficiency. To that end, we have performed several physical and software-based measurements of workloads from HEP applications running on ARM and Intel architectures, and compare their power consumption and performance. We leverage several profiling tools (both in hardware and software) to extract different characteristics of the power use. We report the results of these measurements and the experience gained in developing a set of measurement techniques and profiling tools to accurately assess the power consumption for scientific workloads.
Original languageEnglish
Article number012032
JournalJournal of Physics: Conference Series
Volume608
DOIs
Publication statusPublished - 22 May 2015
MoE publication typeA1 Journal article-refereed

Fingerprint

computer programs
energy
physics
central processing units
hardware
costs
optimization

Cite this

Abdurachmanov, D., Elmer, P., Eulisse, G., Knight, R., Niemi, T., Nurminen, J. K., ... Khan, K. (2015). Techniques and tools for measuring energy efficiency of scientific software applications. Journal of Physics: Conference Series, 608, [012032]. https://doi.org/10.1088/1742-6596/608/1/012032
Abdurachmanov, David ; Elmer, Peter ; Eulisse, Giulio ; Knight, Robert ; Niemi, Tapio ; Nurminen, Jukka K. ; Nyback, Filip ; Pestana, Gonçalo ; Ou, Zhonghong ; Khan, Kashif. / Techniques and tools for measuring energy efficiency of scientific software applications. In: Journal of Physics: Conference Series. 2015 ; Vol. 608.
@article{c5882f34ed0147678b3b7a07510f776c,
title = "Techniques and tools for measuring energy efficiency of scientific software applications",
abstract = "The scale of scientific High Performance Computing (HPC) and High Throughput Computing (HTC) has increased significantly in recent years, and is becoming sensitive to total energy use and cost. Energy-efficiency has thus become an important concern in scientific fields such as High Energy Physics (HEP). There has been a growing interest in utilizing alternate architectures, such as low power ARM processors, to replace traditional Intel x86 architectures. Nevertheless, even though such solutions have been successfully used in mobile applications with low I/O and memory demands, it is unclear if they are suitable and more energy-efficient in the scientific computing environment. Furthermore, there is a lack of tools and experience to derive and compare power consumption between the architectures for various workloads, and eventually to support software optimizations for energy efficiency. To that end, we have performed several physical and software-based measurements of workloads from HEP applications running on ARM and Intel architectures, and compare their power consumption and performance. We leverage several profiling tools (both in hardware and software) to extract different characteristics of the power use. We report the results of these measurements and the experience gained in developing a set of measurement techniques and profiling tools to accurately assess the power consumption for scientific workloads.",
author = "David Abdurachmanov and Peter Elmer and Giulio Eulisse and Robert Knight and Tapio Niemi and Nurminen, {Jukka K.} and Filip Nyback and Gon{\cc}alo Pestana and Zhonghong Ou and Kashif Khan",
year = "2015",
month = "5",
day = "22",
doi = "10.1088/1742-6596/608/1/012032",
language = "English",
volume = "608",
journal = "Journal of Physics: Conference Series",
issn = "1742-6588",
publisher = "Institute of Physics IOP",

}

Abdurachmanov, D, Elmer, P, Eulisse, G, Knight, R, Niemi, T, Nurminen, JK, Nyback, F, Pestana, G, Ou, Z & Khan, K 2015, 'Techniques and tools for measuring energy efficiency of scientific software applications', Journal of Physics: Conference Series, vol. 608, 012032. https://doi.org/10.1088/1742-6596/608/1/012032

Techniques and tools for measuring energy efficiency of scientific software applications. / Abdurachmanov, David; Elmer, Peter; Eulisse, Giulio; Knight, Robert; Niemi, Tapio; Nurminen, Jukka K.; Nyback, Filip; Pestana, Gonçalo; Ou, Zhonghong; Khan, Kashif.

In: Journal of Physics: Conference Series, Vol. 608, 012032, 22.05.2015.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Techniques and tools for measuring energy efficiency of scientific software applications

AU - Abdurachmanov, David

AU - Elmer, Peter

AU - Eulisse, Giulio

AU - Knight, Robert

AU - Niemi, Tapio

AU - Nurminen, Jukka K.

AU - Nyback, Filip

AU - Pestana, Gonçalo

AU - Ou, Zhonghong

AU - Khan, Kashif

PY - 2015/5/22

Y1 - 2015/5/22

N2 - The scale of scientific High Performance Computing (HPC) and High Throughput Computing (HTC) has increased significantly in recent years, and is becoming sensitive to total energy use and cost. Energy-efficiency has thus become an important concern in scientific fields such as High Energy Physics (HEP). There has been a growing interest in utilizing alternate architectures, such as low power ARM processors, to replace traditional Intel x86 architectures. Nevertheless, even though such solutions have been successfully used in mobile applications with low I/O and memory demands, it is unclear if they are suitable and more energy-efficient in the scientific computing environment. Furthermore, there is a lack of tools and experience to derive and compare power consumption between the architectures for various workloads, and eventually to support software optimizations for energy efficiency. To that end, we have performed several physical and software-based measurements of workloads from HEP applications running on ARM and Intel architectures, and compare their power consumption and performance. We leverage several profiling tools (both in hardware and software) to extract different characteristics of the power use. We report the results of these measurements and the experience gained in developing a set of measurement techniques and profiling tools to accurately assess the power consumption for scientific workloads.

AB - The scale of scientific High Performance Computing (HPC) and High Throughput Computing (HTC) has increased significantly in recent years, and is becoming sensitive to total energy use and cost. Energy-efficiency has thus become an important concern in scientific fields such as High Energy Physics (HEP). There has been a growing interest in utilizing alternate architectures, such as low power ARM processors, to replace traditional Intel x86 architectures. Nevertheless, even though such solutions have been successfully used in mobile applications with low I/O and memory demands, it is unclear if they are suitable and more energy-efficient in the scientific computing environment. Furthermore, there is a lack of tools and experience to derive and compare power consumption between the architectures for various workloads, and eventually to support software optimizations for energy efficiency. To that end, we have performed several physical and software-based measurements of workloads from HEP applications running on ARM and Intel architectures, and compare their power consumption and performance. We leverage several profiling tools (both in hardware and software) to extract different characteristics of the power use. We report the results of these measurements and the experience gained in developing a set of measurement techniques and profiling tools to accurately assess the power consumption for scientific workloads.

U2 - 10.1088/1742-6596/608/1/012032

DO - 10.1088/1742-6596/608/1/012032

M3 - Article

VL - 608

JO - Journal of Physics: Conference Series

JF - Journal of Physics: Conference Series

SN - 1742-6588

M1 - 012032

ER -