Torta: Extending applications' capability to select heterogeneous computing resources adaptively to the PC usage

Tetsuro Horikawa, Jin Nakazawa, Kazunori Takashio, Hideyuki Tokuda

Research output: Contribution to journalArticle

Abstract

Spread of GPU-accelerated applications on PCs can cause serious degradation of the user experience such as frame dropping on the video playback, due to applications' resource competition on the same GPU due to arbitrary processors selection. In this paper, we propose a processors assignment system for real applications that achieves processors assignment according to condition based rules without modifying applications. To demonstrate the feasibility of our concept, we implemented a prototype of the centralized processors assignment mechanism called Torta. Our experiment using eight practical applications has shown that Torta achieves binary-compatible processors switching with an average performance penalty on only 0.2%. In a particular case where a video playback application is executed with three other GPU-intensive applications, our method enables users to enjoy the video playback with 60 frames per second (FPS) while the FPS decreases to 14 without the mechanism. This paper shows the design and the implementation of Torta on Windows 7 and concludes that our mechanism increases the efficiency of computational resource usage on PCs, thus improves the overall user experiences.

Original languageEnglish
Pages (from-to)128-140
Number of pages13
JournalIPSJ Online Transactions
Volume6
Issue number2013
DOIs
Publication statusPublished - 2013

Fingerprint

Degradation
Graphics processing unit
Experiments

Keywords

  • Binarycompatibility
  • GPU
  • OpenCL
  • Processors assignment
  • Resource management
  • Users' context
  • Users' preference

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Torta : Extending applications' capability to select heterogeneous computing resources adaptively to the PC usage. / Horikawa, Tetsuro; Nakazawa, Jin; Takashio, Kazunori; Tokuda, Hideyuki.

In: IPSJ Online Transactions, Vol. 6, No. 2013, 2013, p. 128-140.

Research output: Contribution to journalArticle

@article{239a9f10abac4921b781eddf794b4a8b,
title = "Torta: Extending applications' capability to select heterogeneous computing resources adaptively to the PC usage",
abstract = "Spread of GPU-accelerated applications on PCs can cause serious degradation of the user experience such as frame dropping on the video playback, due to applications' resource competition on the same GPU due to arbitrary processors selection. In this paper, we propose a processors assignment system for real applications that achieves processors assignment according to condition based rules without modifying applications. To demonstrate the feasibility of our concept, we implemented a prototype of the centralized processors assignment mechanism called Torta. Our experiment using eight practical applications has shown that Torta achieves binary-compatible processors switching with an average performance penalty on only 0.2{\%}. In a particular case where a video playback application is executed with three other GPU-intensive applications, our method enables users to enjoy the video playback with 60 frames per second (FPS) while the FPS decreases to 14 without the mechanism. This paper shows the design and the implementation of Torta on Windows 7 and concludes that our mechanism increases the efficiency of computational resource usage on PCs, thus improves the overall user experiences.",
keywords = "Binarycompatibility, GPU, OpenCL, Processors assignment, Resource management, Users' context, Users' preference",
author = "Tetsuro Horikawa and Jin Nakazawa and Kazunori Takashio and Hideyuki Tokuda",
year = "2013",
doi = "10.2197/ipsjtrans.6.128",
language = "English",
volume = "6",
pages = "128--140",
journal = "IPSJ Online Transactions",
issn = "1882-6660",
publisher = "Information Processing Society of Japan",
number = "2013",

}

TY - JOUR

T1 - Torta

T2 - Extending applications' capability to select heterogeneous computing resources adaptively to the PC usage

AU - Horikawa, Tetsuro

AU - Nakazawa, Jin

AU - Takashio, Kazunori

AU - Tokuda, Hideyuki

PY - 2013

Y1 - 2013

N2 - Spread of GPU-accelerated applications on PCs can cause serious degradation of the user experience such as frame dropping on the video playback, due to applications' resource competition on the same GPU due to arbitrary processors selection. In this paper, we propose a processors assignment system for real applications that achieves processors assignment according to condition based rules without modifying applications. To demonstrate the feasibility of our concept, we implemented a prototype of the centralized processors assignment mechanism called Torta. Our experiment using eight practical applications has shown that Torta achieves binary-compatible processors switching with an average performance penalty on only 0.2%. In a particular case where a video playback application is executed with three other GPU-intensive applications, our method enables users to enjoy the video playback with 60 frames per second (FPS) while the FPS decreases to 14 without the mechanism. This paper shows the design and the implementation of Torta on Windows 7 and concludes that our mechanism increases the efficiency of computational resource usage on PCs, thus improves the overall user experiences.

AB - Spread of GPU-accelerated applications on PCs can cause serious degradation of the user experience such as frame dropping on the video playback, due to applications' resource competition on the same GPU due to arbitrary processors selection. In this paper, we propose a processors assignment system for real applications that achieves processors assignment according to condition based rules without modifying applications. To demonstrate the feasibility of our concept, we implemented a prototype of the centralized processors assignment mechanism called Torta. Our experiment using eight practical applications has shown that Torta achieves binary-compatible processors switching with an average performance penalty on only 0.2%. In a particular case where a video playback application is executed with three other GPU-intensive applications, our method enables users to enjoy the video playback with 60 frames per second (FPS) while the FPS decreases to 14 without the mechanism. This paper shows the design and the implementation of Torta on Windows 7 and concludes that our mechanism increases the efficiency of computational resource usage on PCs, thus improves the overall user experiences.

KW - Binarycompatibility

KW - GPU

KW - OpenCL

KW - Processors assignment

KW - Resource management

KW - Users' context

KW - Users' preference

UR - http://www.scopus.com/inward/record.url?scp=84937041275&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84937041275&partnerID=8YFLogxK

U2 - 10.2197/ipsjtrans.6.128

DO - 10.2197/ipsjtrans.6.128

M3 - Article

AN - SCOPUS:84937041275

VL - 6

SP - 128

EP - 140

JO - IPSJ Online Transactions

JF - IPSJ Online Transactions

SN - 1882-6660

IS - 2013

ER -