Detail výsledku

The investigation of the ARMv7 and Intel Haswell architectures suitability for performance and energy-aware computing

NIKL, V.; HRADECKÝ, M.; JAROŠ, J.; KELEČÉNI, J. The investigation of the ARMv7 and Intel Haswell architectures suitability for performance and energy-aware computing. In High Performance Computing. Lecture Notes in Computer Science. Cham: Springer International Publishing, 2017. p. 377-393. ISBN: 978-3-319-58666-3.
Typ
článek ve sborníku konference
Jazyk
anglicky
Autoři
Nikl Vojtěch, Ing., UPSY (FIT)
Hradecký Michal, Ing.
Jaroš Jiří, prof. Ing., Ph.D., UPSY (FIT)
Kelečéni Jakub, Ing.
Abstrakt

Reducing CPU frequency and voltage is a well-known approach to improve energy consumption of memory-bound applications. This is based on the conception that the performance of the main memory sees little or no degradation at reduced processor clock speeds, while power consumption decreases significantly and the overall energy efficiency improves. We study this effect on the Haswell generation of Intel Xeon processors as well as the ARMv7 generation of the 32-bit ARM big.LITTLE architecture. The goal is to analyze and compare performance, energy consumption and resulting energy efficiency on a series of tasks, each focusing on different parts of the system, provide analysis and generalization to other similar architectures. The benchmark suit consists of compute and memory intensive benchmarks as well as both single and multi-threaded scientific applications. The results show that frequency and voltage scaling can significantly improve algorithms' energy efficiency, up to 2.5x on ARM and 1.5x on Intel compared to the maximum frequency. ARM is up to 2x more efficient than Intel. The results are valuable for scientists in the areas of computer architecture, performance and power analysis as well as application developers seeking to optimize their Haswell x86_64 and ARMv7 hardware setup to improve energy efficiency.

Klíčová slova

Haswell, ARMv7, Odroid XU4, k-Wave, LAMMPS, Energy Efficiency.

Rok
2017
Strany
377–393
Sborník
High Performance Computing
Řada
Lecture Notes in Computer Science
Svazek
10266
Konference
International Supercomputing Conference 2017
ISBN
978-3-319-58666-3
Vydavatel
Springer International Publishing
Místo
Cham
DOI
UT WoS
000431846900020
EID Scopus
BibTeX
@inproceedings{BUT144403,
  author="Vojtěch {Nikl} and Michal {Hradecký} and Jiří {Jaroš} and Jakub {Kelečéni}",
  title="The investigation of the ARMv7 and Intel Haswell architectures suitability for performance and energy-aware computing",
  booktitle="High Performance Computing",
  year="2017",
  series="Lecture Notes in Computer Science",
  volume="10266",
  pages="377--393",
  publisher="Springer International Publishing",
  address="Cham",
  doi="10.1007/978-3-319-58667-0\{_}20",
  isbn="978-3-319-58666-3",
  url="https://www.fit.vut.cz/research/publication/11303/"
}
Soubory
Projekty
Architektury paralelních a vestavěných počítačových systémů, VUT, Vnitřní projekty VUT, FIT-S-14-2297, zahájení: 2014-01-01, ukončení: 2016-12-31, ukončen
IT4Innovations excellence in science, MŠMT, Národní program udržitelnosti II, LQ1602, zahájení: 2016-01-01, ukončení: 2020-12-31, ukončen
Výzkumné skupiny
Pracoviště
Nahoru