An out-of-order vector processing mechanism for multimedia applications

Ye Gao; Ryusuke Egawa; Hiroyuki Takizawa; Hiroaki Kobayashi

doi:10.1145/2212908.2212941

An out-of-order vector processing mechanism for multimedia applications

Ye Gao, Ryusuke Egawa, Hiroyuki Takizawa, Hiroaki Kobayashi

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Next generation multimedia applications (MMAs) will become a heavy workload for general-purpose processors (GPPs). Although the current single instruction multiple data (SIMD) extensions enable to improve the performance of MMAs, their limited parallel-processing capability is a drawback to execute next generation MMAs in the era of data explosion. This paper proposes an out-of-order vector processing mechanism named OVPM, in order to expand the potential of GPPs on next generation media processing. OVPM can more efficiently utilize memory bandwidth and vector function units than a conventional in-order vector processing mechanism (IVPM), especially for the MMAs with short vectors. The results of performance evaluations using kernels of MMAs show that OVPM achieves a higher performance than that with IVPM.

Original language	English
Title of host publication	CF '12 - Proceedings of the ACM Computing Frontiers Conference
Pages	233-235
Number of pages	3
DOIs	https://doi.org/10.1145/2212908.2212941
Publication status	Published - 2012 Jun 28
Event	ACM Computing Frontiers Conference, CF '12 - Cagliari, Italy Duration: 2012 May 15 → 2012 May 17

Publication series

Name	CF '12 - Proceedings of the ACM Computing Frontiers Conference

Other

Other	ACM Computing Frontiers Conference, CF '12
Country	Italy
City	Cagliari
Period	12/5/15 → 12/5/17

Keywords

SIMD extension
multimedia application
vector architecture

ASJC Scopus subject areas

Software

Access to Document

10.1145/2212908.2212941

Fingerprint Dive into the research topics of 'An out-of-order vector processing mechanism for multimedia applications'. Together they form a unique fingerprint.

Cite this

An out-of-order vector processing mechanism for multimedia applications. / Gao, Ye; Egawa, Ryusuke; Takizawa, Hiroyuki ; Kobayashi, Hiroaki.

CF '12 - Proceedings of the ACM Computing Frontiers Conference. 2012. p. 233-235 (CF '12 - Proceedings of the ACM Computing Frontiers Conference).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Gao, Y, Egawa, R, Takizawa, H & Kobayashi, H 2012, An out-of-order vector processing mechanism for multimedia applications. in CF '12 - Proceedings of the ACM Computing Frontiers Conference. CF '12 - Proceedings of the ACM Computing Frontiers Conference, pp. 233-235, ACM Computing Frontiers Conference, CF '12, Cagliari, Italy, 12/5/15. https://doi.org/10.1145/2212908.2212941

@inproceedings{15e352a6effe45beb8434574d068a6b0,

title = "An out-of-order vector processing mechanism for multimedia applications",

abstract = "Next generation multimedia applications (MMAs) will become a heavy workload for general-purpose processors (GPPs). Although the current single instruction multiple data (SIMD) extensions enable to improve the performance of MMAs, their limited parallel-processing capability is a drawback to execute next generation MMAs in the era of data explosion. This paper proposes an out-of-order vector processing mechanism named OVPM, in order to expand the potential of GPPs on next generation media processing. OVPM can more efficiently utilize memory bandwidth and vector function units than a conventional in-order vector processing mechanism (IVPM), especially for the MMAs with short vectors. The results of performance evaluations using kernels of MMAs show that OVPM achieves a higher performance than that with IVPM.",

keywords = "SIMD extension, multimedia application, vector architecture",

author = "Ye Gao and Ryusuke Egawa and Hiroyuki Takizawa and Hiroaki Kobayashi",

year = "2012",

month = jun,

day = "28",

doi = "10.1145/2212908.2212941",

language = "English",

isbn = "9781450312158",

series = "CF '12 - Proceedings of the ACM Computing Frontiers Conference",

pages = "233--235",

booktitle = "CF '12 - Proceedings of the ACM Computing Frontiers Conference",

note = "ACM Computing Frontiers Conference, CF '12 ; Conference date: 15-05-2012 Through 17-05-2012",

}

TY - GEN

T1 - An out-of-order vector processing mechanism for multimedia applications

AU - Gao, Ye

AU - Egawa, Ryusuke

AU - Takizawa, Hiroyuki

AU - Kobayashi, Hiroaki

PY - 2012/6/28

Y1 - 2012/6/28

N2 - Next generation multimedia applications (MMAs) will become a heavy workload for general-purpose processors (GPPs). Although the current single instruction multiple data (SIMD) extensions enable to improve the performance of MMAs, their limited parallel-processing capability is a drawback to execute next generation MMAs in the era of data explosion. This paper proposes an out-of-order vector processing mechanism named OVPM, in order to expand the potential of GPPs on next generation media processing. OVPM can more efficiently utilize memory bandwidth and vector function units than a conventional in-order vector processing mechanism (IVPM), especially for the MMAs with short vectors. The results of performance evaluations using kernels of MMAs show that OVPM achieves a higher performance than that with IVPM.

AB - Next generation multimedia applications (MMAs) will become a heavy workload for general-purpose processors (GPPs). Although the current single instruction multiple data (SIMD) extensions enable to improve the performance of MMAs, their limited parallel-processing capability is a drawback to execute next generation MMAs in the era of data explosion. This paper proposes an out-of-order vector processing mechanism named OVPM, in order to expand the potential of GPPs on next generation media processing. OVPM can more efficiently utilize memory bandwidth and vector function units than a conventional in-order vector processing mechanism (IVPM), especially for the MMAs with short vectors. The results of performance evaluations using kernels of MMAs show that OVPM achieves a higher performance than that with IVPM.

KW - SIMD extension

KW - multimedia application

KW - vector architecture

UR - http://www.scopus.com/inward/record.url?scp=84862671997&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862671997&partnerID=8YFLogxK

U2 - 10.1145/2212908.2212941

DO - 10.1145/2212908.2212941

M3 - Conference contribution

AN - SCOPUS:84862671997

SN - 9781450312158

T3 - CF '12 - Proceedings of the ACM Computing Frontiers Conference

SP - 233

EP - 235

BT - CF '12 - Proceedings of the ACM Computing Frontiers Conference

T2 - ACM Computing Frontiers Conference, CF '12

Y2 - 15 May 2012 through 17 May 2012

ER -