OSCAR compiler controlled multicore power reduction on android platform

Hideo Yamamoto, Tomohiro Hirano, Kohei Muto, Hiroki Mikami, Takashi Goto1, Dominic Hillenbrand, Moriyuki Takamura, Keiji Kimura, Hironori Kasahara

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

In recent years, smart devices are transitioning from single core processors to multicore processors to satisfy the growing demands of higher performance and lower power consumption. However, power consumption of multicore processors is increasing, as usage of smart devices become more intense. This situation is one of the most fundamental and important obstacle that the mobile device industries face, to extend the battery life of smart devices. This paper evaluates the power reduction control by the OSCAR Automatic Parallelizing Compiler on an Android platform with the newly developed precise power measurement environment on the ODROID-X2, a development platform with the Samsung Exynos4412 Prime, which consists of 4 ARM Cortex-A9 cores. The OSCAR Compiler enables automatic exploitation of multigrain parallelism within a sequential program, and automatically generates a parallelized code with the OSCAR Multi-Platform API power reduction directives for the purpose of DVFS (Dynamic Voltage and Frequency Scaling), clock gating, and power gating. The paper also introduces a newly developed micro second order pseudo clock gating method to reduce power consumption using WFI (Wait For Interrupt). By inserting GPIO (General Purpose Input Output) control functions into programs, signals appear on the power waveform indicating the point of where the GPIO control was inserted and provides a precise power measurement of the specified program area. The results of the power evaluation for realtime Mpeg2 Decoder show 86.7% power reduction, namely from 2.79[W] to 0.37[W] and for real-time Optical Flow show 86.5% power reduction, namely from 2.23[W] to 0.36[W] on 3 core execution.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages155-168
Number of pages14
Volume8664
ISBN (Print)9783319099668
DOIs
Publication statusPublished - 2014
Event26th Workshop on Languages and Compilers for Parallel Computing, LCPC 2013 - San Jose
Duration: 2013 Sep 252013 Sep 27

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8664
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other26th Workshop on Languages and Compilers for Parallel Computing, LCPC 2013
CitySan Jose
Period13/9/2513/9/27

Fingerprint

Compiler
Power Consumption
Electric power utilization
Multi-core Processor
Clocks
Parallelizing Compilers
Real-time
Optical flows
Output
Optical Flow
Control Function
Cortex
Application programming interfaces (API)
Mobile devices
Mobile Devices
Battery
Waveform
Exploitation
Parallelism
High Performance

Keywords

  • Android
  • API
  • Automatic parallelization
  • Multicore processor
  • Power control
  • Power reduction
  • Smart device
  • WFI

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Yamamoto, H., Hirano, T., Muto, K., Mikami, H., Goto1, T., Hillenbrand, D., ... Kasahara, H. (2014). OSCAR compiler controlled multicore power reduction on android platform. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8664, pp. 155-168). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8664). Springer Verlag. https://doi.org/10.1007/978-3-319-09967-5_9

OSCAR compiler controlled multicore power reduction on android platform. / Yamamoto, Hideo; Hirano, Tomohiro; Muto, Kohei; Mikami, Hiroki; Goto1, Takashi; Hillenbrand, Dominic; Takamura, Moriyuki; Kimura, Keiji; Kasahara, Hironori.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8664 Springer Verlag, 2014. p. 155-168 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8664).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Yamamoto, H, Hirano, T, Muto, K, Mikami, H, Goto1, T, Hillenbrand, D, Takamura, M, Kimura, K & Kasahara, H 2014, OSCAR compiler controlled multicore power reduction on android platform. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 8664, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8664, Springer Verlag, pp. 155-168, 26th Workshop on Languages and Compilers for Parallel Computing, LCPC 2013, San Jose, 13/9/25. https://doi.org/10.1007/978-3-319-09967-5_9
Yamamoto H, Hirano T, Muto K, Mikami H, Goto1 T, Hillenbrand D et al. OSCAR compiler controlled multicore power reduction on android platform. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8664. Springer Verlag. 2014. p. 155-168. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-09967-5_9
Yamamoto, Hideo ; Hirano, Tomohiro ; Muto, Kohei ; Mikami, Hiroki ; Goto1, Takashi ; Hillenbrand, Dominic ; Takamura, Moriyuki ; Kimura, Keiji ; Kasahara, Hironori. / OSCAR compiler controlled multicore power reduction on android platform. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8664 Springer Verlag, 2014. pp. 155-168 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{2c0f69f2b871465980217f0d852baf4e,
title = "OSCAR compiler controlled multicore power reduction on android platform",
abstract = "In recent years, smart devices are transitioning from single core processors to multicore processors to satisfy the growing demands of higher performance and lower power consumption. However, power consumption of multicore processors is increasing, as usage of smart devices become more intense. This situation is one of the most fundamental and important obstacle that the mobile device industries face, to extend the battery life of smart devices. This paper evaluates the power reduction control by the OSCAR Automatic Parallelizing Compiler on an Android platform with the newly developed precise power measurement environment on the ODROID-X2, a development platform with the Samsung Exynos4412 Prime, which consists of 4 ARM Cortex-A9 cores. The OSCAR Compiler enables automatic exploitation of multigrain parallelism within a sequential program, and automatically generates a parallelized code with the OSCAR Multi-Platform API power reduction directives for the purpose of DVFS (Dynamic Voltage and Frequency Scaling), clock gating, and power gating. The paper also introduces a newly developed micro second order pseudo clock gating method to reduce power consumption using WFI (Wait For Interrupt). By inserting GPIO (General Purpose Input Output) control functions into programs, signals appear on the power waveform indicating the point of where the GPIO control was inserted and provides a precise power measurement of the specified program area. The results of the power evaluation for realtime Mpeg2 Decoder show 86.7{\%} power reduction, namely from 2.79[W] to 0.37[W] and for real-time Optical Flow show 86.5{\%} power reduction, namely from 2.23[W] to 0.36[W] on 3 core execution.",
keywords = "Android, API, Automatic parallelization, Multicore processor, Power control, Power reduction, Smart device, WFI",
author = "Hideo Yamamoto and Tomohiro Hirano and Kohei Muto and Hiroki Mikami and Takashi Goto1 and Dominic Hillenbrand and Moriyuki Takamura and Keiji Kimura and Hironori Kasahara",
year = "2014",
doi = "10.1007/978-3-319-09967-5_9",
language = "English",
isbn = "9783319099668",
volume = "8664",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "155--168",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - OSCAR compiler controlled multicore power reduction on android platform

AU - Yamamoto, Hideo

AU - Hirano, Tomohiro

AU - Muto, Kohei

AU - Mikami, Hiroki

AU - Goto1, Takashi

AU - Hillenbrand, Dominic

AU - Takamura, Moriyuki

AU - Kimura, Keiji

AU - Kasahara, Hironori

PY - 2014

Y1 - 2014

N2 - In recent years, smart devices are transitioning from single core processors to multicore processors to satisfy the growing demands of higher performance and lower power consumption. However, power consumption of multicore processors is increasing, as usage of smart devices become more intense. This situation is one of the most fundamental and important obstacle that the mobile device industries face, to extend the battery life of smart devices. This paper evaluates the power reduction control by the OSCAR Automatic Parallelizing Compiler on an Android platform with the newly developed precise power measurement environment on the ODROID-X2, a development platform with the Samsung Exynos4412 Prime, which consists of 4 ARM Cortex-A9 cores. The OSCAR Compiler enables automatic exploitation of multigrain parallelism within a sequential program, and automatically generates a parallelized code with the OSCAR Multi-Platform API power reduction directives for the purpose of DVFS (Dynamic Voltage and Frequency Scaling), clock gating, and power gating. The paper also introduces a newly developed micro second order pseudo clock gating method to reduce power consumption using WFI (Wait For Interrupt). By inserting GPIO (General Purpose Input Output) control functions into programs, signals appear on the power waveform indicating the point of where the GPIO control was inserted and provides a precise power measurement of the specified program area. The results of the power evaluation for realtime Mpeg2 Decoder show 86.7% power reduction, namely from 2.79[W] to 0.37[W] and for real-time Optical Flow show 86.5% power reduction, namely from 2.23[W] to 0.36[W] on 3 core execution.

AB - In recent years, smart devices are transitioning from single core processors to multicore processors to satisfy the growing demands of higher performance and lower power consumption. However, power consumption of multicore processors is increasing, as usage of smart devices become more intense. This situation is one of the most fundamental and important obstacle that the mobile device industries face, to extend the battery life of smart devices. This paper evaluates the power reduction control by the OSCAR Automatic Parallelizing Compiler on an Android platform with the newly developed precise power measurement environment on the ODROID-X2, a development platform with the Samsung Exynos4412 Prime, which consists of 4 ARM Cortex-A9 cores. The OSCAR Compiler enables automatic exploitation of multigrain parallelism within a sequential program, and automatically generates a parallelized code with the OSCAR Multi-Platform API power reduction directives for the purpose of DVFS (Dynamic Voltage and Frequency Scaling), clock gating, and power gating. The paper also introduces a newly developed micro second order pseudo clock gating method to reduce power consumption using WFI (Wait For Interrupt). By inserting GPIO (General Purpose Input Output) control functions into programs, signals appear on the power waveform indicating the point of where the GPIO control was inserted and provides a precise power measurement of the specified program area. The results of the power evaluation for realtime Mpeg2 Decoder show 86.7% power reduction, namely from 2.79[W] to 0.37[W] and for real-time Optical Flow show 86.5% power reduction, namely from 2.23[W] to 0.36[W] on 3 core execution.

KW - Android

KW - API

KW - Automatic parallelization

KW - Multicore processor

KW - Power control

KW - Power reduction

KW - Smart device

KW - WFI

UR - http://www.scopus.com/inward/record.url?scp=84910025793&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84910025793&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-09967-5_9

DO - 10.1007/978-3-319-09967-5_9

M3 - Conference contribution

SN - 9783319099668

VL - 8664

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 155

EP - 168

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -