schliessen

Filtern

 

Bibliotheken

Automatic intra-register vectorization for the Intel[R] architecture

Recent extensions to the Intel Architecture feature the SIMD technique to enhance the performance of computational intensive applications that perform the same operation on different elements in a data set. To date, much of the code that exploits these extensions has been hand-coded. The task of the... Full description

Journal Title: International journal of parallel programming 2002-04-01, Vol.30 (2), p.65-98
Main Author: Bik, Aart J.C
Other Authors: Girkar, Milind , Grey, Paul M , Tian, Xinmin
Format: Electronic Article Electronic Article
Language: English
Subjects:
Publisher: New York: Springer
ID: ISSN: 0885-7458
Zum Text:
SendSend as email Add to Book BagAdd to Book Bag
Staff View
recordid: cdi_proquest_miscellaneous_36306461
title: Automatic intra-register vectorization for the Intel[R] architecture
format: Article
creator:
  • Bik, Aart J.C
  • Girkar, Milind
  • Grey, Paul M
  • Tian, Xinmin
subjects:
  • Design and construction
  • Planning
  • Processor architecture
  • Semiconductor industry
  • Studies
ispartof: International journal of parallel programming, 2002-04-01, Vol.30 (2), p.65-98
description: Recent extensions to the Intel Architecture feature the SIMD technique to enhance the performance of computational intensive applications that perform the same operation on different elements in a data set. To date, much of the code that exploits these extensions has been hand-coded. The task of the programmer is substantially simplified, however, if a compiler does this exploitation automatically. The high-performance Intel C++/Fortran compiler supports automatic translation of serial loops into code that uses the SIMD extensions to the Intel Architecture. This paper provides a detailed overview of the automatic vectorization methods used by this compiler together with an experimental validation of their effectiveness.
language: eng
source:
identifier: ISSN: 0885-7458
fulltext: no_fulltext
issn:
  • 0885-7458
  • 1573-7640
url: Link


@attributes
NO1
SEARCH_ENGINEprimo_central_multiple_fe
SEARCH_ENGINE_TYPEPrimo Central Search Engine
RANK2.1043134
LOCALfalse
PrimoNMBib
record
control
sourceidgale_proqu
recordidTN_cdi_proquest_miscellaneous_36306461
sourceformatXML
sourcesystemPC
galeidA84549686
sourcerecordidA84549686
originalsourceidFETCH-LOGICAL-1324t-ea870ab4063ccbda469f1f870393915385b66c08f5c293f0ff058af99f56ce020
addsrcrecordideNpt0EtLAzEQAOAgCtbq2eviwdvq5LmJt6VWLRQE0ZPIkqZJm7LdaDbrwV9vioIiMoeBmY9hZhA6xXCBgdDL-goDZoQCI4qxag-NMK9oWQkG-2gEUvKyYlweoqO-3wCAqqQcoet6SGGrkzeF71LUZbQr3ycbi3drUoj-I_dCV7gQi7S2xaxLtn1-eCl0NGufshmiPUYHTre9PfnOY_R0M32c3JXz-9vZpJ6XmBKWSqtlBXrBQFBjFkvNhHLY5RpVVGFOJV8IYUA6boiiDpwDLrVTynFhLBAYo_Ovua8xvA22T83W98a2re5sGPqGCgqCCZzh2R-4CUPs8m4N2f2HkOoXWunWNr5zId9vdhObWjLOlJAio_IflGNpt96Ezjqf6z_-E-1ldMQ
sourcetypeAggregation Database
isCDItrue
recordtypearticle
pqid204292271
display
typearticle
titleAutomatic intra-register vectorization for the Intel[R] architecture
creatorBik, Aart J.C ; Girkar, Milind ; Grey, Paul M ; Tian, Xinmin
creatorcontribBik, Aart J.C ; Girkar, Milind ; Grey, Paul M ; Tian, Xinmin
descriptionRecent extensions to the Intel Architecture feature the SIMD technique to enhance the performance of computational intensive applications that perform the same operation on different elements in a data set. To date, much of the code that exploits these extensions has been hand-coded. The task of the programmer is substantially simplified, however, if a compiler does this exploitation automatically. The high-performance Intel C++/Fortran compiler supports automatic translation of serial loops into code that uses the SIMD extensions to the Intel Architecture. This paper provides a detailed overview of the automatic vectorization methods used by this compiler together with an experimental validation of their effectiveness.
identifier
0ISSN: 0885-7458
1EISSN: 1573-7640
2DOI: 10.1023/A:1014230429447
3CODEN: IJPPE5
languageeng
publisherNew York: Springer
subjectDesign and construction ; Planning ; Processor architecture ; Semiconductor industry ; Studies
ispartofInternational journal of parallel programming, 2002-04-01, Vol.30 (2), p.65-98
rights
0COPYRIGHT 2002 Springer
1Plenum Publishing Corporation 2002
lds50peer_reviewed
links
openurl$$Topenurl_article
thumbnail$$Usyndetics_thumb_exl
search
creatorcontrib
0Bik, Aart J.C
1Girkar, Milind
2Grey, Paul M
3Tian, Xinmin
title
0Automatic intra-register vectorization for the Intel[R] architecture
1International journal of parallel programming
descriptionRecent extensions to the Intel Architecture feature the SIMD technique to enhance the performance of computational intensive applications that perform the same operation on different elements in a data set. To date, much of the code that exploits these extensions has been hand-coded. The task of the programmer is substantially simplified, however, if a compiler does this exploitation automatically. The high-performance Intel C++/Fortran compiler supports automatic translation of serial loops into code that uses the SIMD extensions to the Intel Architecture. This paper provides a detailed overview of the automatic vectorization methods used by this compiler together with an experimental validation of their effectiveness.
subject
0Design and construction
1Planning
2Processor architecture
3Semiconductor industry
4Studies
issn
00885-7458
11573-7640
fulltextfalse
rsrctypearticle
creationdate2002
recordtypearticle
recordideNpt0EtLAzEQAOAgCtbq2eviwdvq5LmJt6VWLRQE0ZPIkqZJm7LdaDbrwV9vioIiMoeBmY9hZhA6xXCBgdDL-goDZoQCI4qxag-NMK9oWQkG-2gEUvKyYlweoqO-3wCAqqQcoet6SGGrkzeF71LUZbQr3ycbi3drUoj-I_dCV7gQi7S2xaxLtn1-eCl0NGufshmiPUYHTre9PfnOY_R0M32c3JXz-9vZpJ6XmBKWSqtlBXrBQFBjFkvNhHLY5RpVVGFOJV8IYUA6boiiDpwDLrVTynFhLBAYo_Ovua8xvA22T83W98a2re5sGPqGCgqCCZzh2R-4CUPs8m4N2f2HkOoXWunWNr5zId9vdhObWjLOlJAio_IflGNpt96Ezjqf6z_-E-1ldMQ
startdate20020401
enddate20020401
creator
0Bik, Aart J.C
1Girkar, Milind
2Grey, Paul M
3Tian, Xinmin
general
0Springer
1Springer Nature B.V
scope
0BSHEE
13V.
27SC
37WY
47WZ
57XB
687Z
78AL
88FD
98FE
108FG
118FK
128FL
138G5
14ABUWG
15ARAPS
16AZQEC
17BENPR
18BEZIV
19BGLVJ
20DWQXO
21FRNLG
22F~G
23GNUQQ
24GUQSH
25HCIFZ
26JQ2
27K60
28K6~
29K7-
30L.-
31L7M
32L~C
33L~D
34M0C
35M0N
36M2O
37MBDVC
38P5Z
39P62
40PADUT
41PQBIZ
42PQBZA
43PQEST
44PQQKQ
45PQUKI
46PRINS
47PYYUZ
48Q9U
sort
creationdate20020401
titleAutomatic intra-register vectorization for the Intel[R] architecture
authorBik, Aart J.C ; Girkar, Milind ; Grey, Paul M ; Tian, Xinmin
facets
frbrtype5
frbrgroupidcdi_FETCH-LOGICAL-1324t-ea870ab4063ccbda469f1f870393915385b66c08f5c293f0ff058af99f56ce020
rsrctypearticles
prefilterarticles
languageeng
creationdate2002
topic
0Design and construction
1Planning
2Processor architecture
3Semiconductor industry
4Studies
toplevelpeer_reviewed
creatorcontrib
0Bik, Aart J.C
1Girkar, Milind
2Grey, Paul M
3Tian, Xinmin
collection
0Academic OneFile (A&I only)
1ProQuest Central (Corporate)
2Computer and Information Systems Abstracts
3ABI/INFORM Collection
4ABI/INFORM Global (PDF only)
5ProQuest Central (purchase pre-March 2016)
6ABI/INFORM Global (Alumni Edition)
7Computing Database (Alumni Edition)
8Technology Research Database
9ProQuest SciTech Collection
10ProQuest Technology Collection
11ProQuest Central (Alumni) (purchase pre-March 2016)
12ABI/INFORM Collection (Alumni Edition)
13Research Library (Alumni Edition)
14ProQuest Central (Alumni Edition)
15Advanced Technologies & Aerospace Collection
16ProQuest Central Essentials
17ProQuest Central
18Business Premium Collection
19Technology Collection
20ProQuest Central Korea
21Business Premium Collection (Alumni)
22ABI/INFORM Global (Corporate)
23ProQuest Central Student
24Research Library Prep
25SciTech Premium Collection
26ProQuest Computer Science Collection
27ProQuest Business Collection (Alumni Edition)
28ProQuest Business Collection
29Computer Science Database
30ABI/INFORM Professional Advanced
31Advanced Technologies Database with Aerospace
32Computer and Information Systems Abstracts – Academic
33Computer and Information Systems Abstracts Professional
34ABI/INFORM Global
35Computing Database
36Research Library
37Research Library (Corporate)
38Advanced Technologies & Aerospace Database
39ProQuest Advanced Technologies & Aerospace Collection
40Research Library China
41ProQuest One Business
42ProQuest One Business (Alumni)
43ProQuest One Academic Eastern Edition
44ProQuest One Academic
45ProQuest One Academic UKI Edition
46ProQuest Central China
47ABI/INFORM Collection China
48ProQuest Central Basic
jtitleInternational journal of parallel programming
delivery
delcategoryRemote Search Resource
fulltextno_fulltext
addata
au
0Bik, Aart J.C
1Girkar, Milind
2Grey, Paul M
3Tian, Xinmin
formatjournal
genrearticle
ristypeJOUR
atitleAutomatic intra-register vectorization for the Intel[R] architecture
jtitleInternational journal of parallel programming
date2002-04-01
risdate2002
volume30
issue2
spage65
epage98
pages65-98
issn0885-7458
eissn1573-7640
codenIJPPE5
abstractRecent extensions to the Intel Architecture feature the SIMD technique to enhance the performance of computational intensive applications that perform the same operation on different elements in a data set. To date, much of the code that exploits these extensions has been hand-coded. The task of the programmer is substantially simplified, however, if a compiler does this exploitation automatically. The high-performance Intel C++/Fortran compiler supports automatic translation of serial loops into code that uses the SIMD extensions to the Intel Architecture. This paper provides a detailed overview of the automatic vectorization methods used by this compiler together with an experimental validation of their effectiveness.
copNew York
pubSpringer
doi10.1023/A:1014230429447