PACUE: Processor allocator considering user experience

Tetsuro Horikawa, Michio Honda, Jin Nakazawa, Kazunori Takashio, Hideyuki Tokuda

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

1 Citation (Scopus)

Abstract

GPU accelerated applications including GPGPU ones are commonly seen in modern PCs. If many applications compete on the same GPU, the performance will decrease significantly. Some applications have a large impact on user experience. Therefore, for such applications, we have to limit GPU utilization by the other applications. It might be straightforward to modify applications to switch compute device dynamically for intelligent resources allocation. Unfortunately, we cannot do so due to software distribution policy or the other reasons. In this paper, we propose PACUE, which allows the end system to allocate compute devices arbitrary to applications. In addition, PACUE guesses optimal compute device for each application according to user preference. We implemented the dynamic compute device redirector of PACUE including OpenCL API hooking and device camouflaging features. We also implemented the frame of the resource manager of PACUE. We demonstrate PACUE achieves dynamic compute device redirecting on one out of two real applications and on all of 20 sample codes.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages335-344
Number of pages10
Volume7156 LNCS
EditionPART 2
DOIs
Publication statusPublished - 2012
Event17th Parallel Processing Workshops, Euro-Par 2011: CCPI 2011, CGWS 2011, HeteroPar 2011, HiBB 2011, HPCVirt 2011, HPPC 2011, HPSS 2011, MDGS 2011, ProPer 2011, Resilience 2011, UCHPC 2011, VHPC 2011 - Bordeaux, France
Duration: 2011 Aug 292011 Sep 2

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 2
Volume7156 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other17th Parallel Processing Workshops, Euro-Par 2011: CCPI 2011, CGWS 2011, HeteroPar 2011, HiBB 2011, HPCVirt 2011, HPPC 2011, HPSS 2011, MDGS 2011, ProPer 2011, Resilience 2011, UCHPC 2011, VHPC 2011
CountryFrance
CityBordeaux
Period11/8/2911/9/2

Fingerprint

User Experience
GPGPU
User Preferences
Guess
Application programming interfaces (API)
Resource Allocation
Resource allocation
Switch
Managers
Switches
Decrease
Resources
Software
Arbitrary

Keywords

  • binary compatibility
  • GPGPU
  • GPU
  • OpenCL
  • PC
  • Resource management
  • user experience

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Horikawa, T., Honda, M., Nakazawa, J., Takashio, K., & Tokuda, H. (2012). PACUE: Processor allocator considering user experience. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (PART 2 ed., Vol. 7156 LNCS, pp. 335-344). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7156 LNCS, No. PART 2). https://doi.org/10.1007/978-3-642-29740-3-38

PACUE : Processor allocator considering user experience. / Horikawa, Tetsuro; Honda, Michio; Nakazawa, Jin; Takashio, Kazunori; Tokuda, Hideyuki.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7156 LNCS PART 2. ed. 2012. p. 335-344 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7156 LNCS, No. PART 2).

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

Horikawa, T, Honda, M, Nakazawa, J, Takashio, K & Tokuda, H 2012, PACUE: Processor allocator considering user experience. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 2 edn, vol. 7156 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 2, vol. 7156 LNCS, pp. 335-344, 17th Parallel Processing Workshops, Euro-Par 2011: CCPI 2011, CGWS 2011, HeteroPar 2011, HiBB 2011, HPCVirt 2011, HPPC 2011, HPSS 2011, MDGS 2011, ProPer 2011, Resilience 2011, UCHPC 2011, VHPC 2011, Bordeaux, France, 11/8/29. https://doi.org/10.1007/978-3-642-29740-3-38
Horikawa T, Honda M, Nakazawa J, Takashio K, Tokuda H. PACUE: Processor allocator considering user experience. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 2 ed. Vol. 7156 LNCS. 2012. p. 335-344. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2). https://doi.org/10.1007/978-3-642-29740-3-38
Horikawa, Tetsuro ; Honda, Michio ; Nakazawa, Jin ; Takashio, Kazunori ; Tokuda, Hideyuki. / PACUE : Processor allocator considering user experience. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7156 LNCS PART 2. ed. 2012. pp. 335-344 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2).
@inproceedings{8f212b677d8740829d0625a9771f2037,
title = "PACUE: Processor allocator considering user experience",
abstract = "GPU accelerated applications including GPGPU ones are commonly seen in modern PCs. If many applications compete on the same GPU, the performance will decrease significantly. Some applications have a large impact on user experience. Therefore, for such applications, we have to limit GPU utilization by the other applications. It might be straightforward to modify applications to switch compute device dynamically for intelligent resources allocation. Unfortunately, we cannot do so due to software distribution policy or the other reasons. In this paper, we propose PACUE, which allows the end system to allocate compute devices arbitrary to applications. In addition, PACUE guesses optimal compute device for each application according to user preference. We implemented the dynamic compute device redirector of PACUE including OpenCL API hooking and device camouflaging features. We also implemented the frame of the resource manager of PACUE. We demonstrate PACUE achieves dynamic compute device redirecting on one out of two real applications and on all of 20 sample codes.",
keywords = "binary compatibility, GPGPU, GPU, OpenCL, PC, Resource management, user experience",
author = "Tetsuro Horikawa and Michio Honda and Jin Nakazawa and Kazunori Takashio and Hideyuki Tokuda",
year = "2012",
doi = "10.1007/978-3-642-29740-3-38",
language = "English",
isbn = "9783642297397",
volume = "7156 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
number = "PART 2",
pages = "335--344",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
edition = "PART 2",

}

TY - GEN

T1 - PACUE

T2 - Processor allocator considering user experience

AU - Horikawa, Tetsuro

AU - Honda, Michio

AU - Nakazawa, Jin

AU - Takashio, Kazunori

AU - Tokuda, Hideyuki

PY - 2012

Y1 - 2012

N2 - GPU accelerated applications including GPGPU ones are commonly seen in modern PCs. If many applications compete on the same GPU, the performance will decrease significantly. Some applications have a large impact on user experience. Therefore, for such applications, we have to limit GPU utilization by the other applications. It might be straightforward to modify applications to switch compute device dynamically for intelligent resources allocation. Unfortunately, we cannot do so due to software distribution policy or the other reasons. In this paper, we propose PACUE, which allows the end system to allocate compute devices arbitrary to applications. In addition, PACUE guesses optimal compute device for each application according to user preference. We implemented the dynamic compute device redirector of PACUE including OpenCL API hooking and device camouflaging features. We also implemented the frame of the resource manager of PACUE. We demonstrate PACUE achieves dynamic compute device redirecting on one out of two real applications and on all of 20 sample codes.

AB - GPU accelerated applications including GPGPU ones are commonly seen in modern PCs. If many applications compete on the same GPU, the performance will decrease significantly. Some applications have a large impact on user experience. Therefore, for such applications, we have to limit GPU utilization by the other applications. It might be straightforward to modify applications to switch compute device dynamically for intelligent resources allocation. Unfortunately, we cannot do so due to software distribution policy or the other reasons. In this paper, we propose PACUE, which allows the end system to allocate compute devices arbitrary to applications. In addition, PACUE guesses optimal compute device for each application according to user preference. We implemented the dynamic compute device redirector of PACUE including OpenCL API hooking and device camouflaging features. We also implemented the frame of the resource manager of PACUE. We demonstrate PACUE achieves dynamic compute device redirecting on one out of two real applications and on all of 20 sample codes.

KW - binary compatibility

KW - GPGPU

KW - GPU

KW - OpenCL

KW - PC

KW - Resource management

KW - user experience

UR - http://www.scopus.com/inward/record.url?scp=84882628602&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84882628602&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-29740-3-38

DO - 10.1007/978-3-642-29740-3-38

M3 - Conference contribution

AN - SCOPUS:84882628602

SN - 9783642297397

VL - 7156 LNCS

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

SP - 335

EP - 344

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

ER -