schliessen

Filtern

 

Bibliotheken

Speedup of Fuzzy Clustering Through Stream Processing on Graphics Processing Units

As the number of data points, feature dimensionality, and number of centers for clustering algorithms increase, computational tractability becomes a problem. The fuzzy c-means has a large degree of inherent algorithmic parallelism that modern CPU architectures do not exploit. Many pattern recognitio... Full description

Journal Title: IEEE Transactions on Fuzzy Systems 2008, Vol.16(4)
Main Author: Anderson, D
Other Authors: Luke, R , Keller, J
Format: Electronic Article Electronic Article
Language: English
Subjects:
ID: ISSN: 1063-6706 ; DOI: 10.1109/TFUZZ.2008.924203
Link: http://search.proquest.com/docview/869572711/?pq-origsite=primo
Zum Text:
SendSend as email Add to Book BagAdd to Book Bag
Staff View
recordid: proquest869572711
title: Speedup of Fuzzy Clustering Through Stream Processing on Graphics Processing Units
format: Article
creator:
  • Anderson, D
  • Luke, R
  • Keller, J
subjects:
  • Algorithms
  • Architecture
  • General (551.5)
ispartof: IEEE Transactions on Fuzzy Systems, 2008, Vol.16(4)
description: As the number of data points, feature dimensionality, and number of centers for clustering algorithms increase, computational tractability becomes a problem. The fuzzy c-means has a large degree of inherent algorithmic parallelism that modern CPU architectures do not exploit. Many pattern recognition algorithms can be sped up on a graphics processing unit (GPU) as long as the majority of computation at various stages and the components are not dependent on each other. We present a generalized method for offloading fuzzy clustering to a GPU, while maintaining control over the number of data points, feature dimensionality, and the number of cluster centers. GPU-based clustering is a high-performance low-cost solution that frees up the CPU. Our results show a speed increase of over two orders of magnitude for particular clustering configurations and platforms.
language: eng
source:
identifier: ISSN: 1063-6706 ; DOI: 10.1109/TFUZZ.2008.924203
fulltext: fulltext
issn:
  • 10636706
  • 1063-6706
url: Link


@attributes
ID1032923887
RANK0.07
NO1
SEARCH_ENGINEprimo_central_multiple_fe
SEARCH_ENGINE_TYPEPrimo Central Search Engine
LOCALfalse
PrimoNMBib
record
control
sourcerecordid869572711
sourceidproquest
recordidTN_proquest869572711
sourcesystemPC
pqid869572711
display
typearticle
titleSpeedup of Fuzzy Clustering Through Stream Processing on Graphics Processing Units
creatorAnderson, D ; Luke, R ; Keller, J
contributorAnderson, D (correspondence author)
ispartofIEEE Transactions on Fuzzy Systems, 2008, Vol.16(4)
identifierISSN: 1063-6706 ; DOI: 10.1109/TFUZZ.2008.924203
subjectAlgorithms ; Architecture ; General (551.5)
descriptionAs the number of data points, feature dimensionality, and number of centers for clustering algorithms increase, computational tractability becomes a problem. The fuzzy c-means has a large degree of inherent algorithmic parallelism that modern CPU architectures do not exploit. Many pattern recognition algorithms can be sped up on a graphics processing unit (GPU) as long as the majority of computation at various stages and the components are not dependent on each other. We present a generalized method for offloading fuzzy clustering to a GPU, while maintaining control over the number of data points, feature dimensionality, and the number of cluster centers. GPU-based clustering is a high-performance low-cost solution that frees up the CPU. Our results show a speed increase of over two orders of magnitude for particular clustering configurations and platforms.
languageeng
source
version4
lds50peer_reviewed
links
openurl$$Topenurl_article
openurlfulltext$$Topenurlfull_article
backlink$$Uhttp://search.proquest.com/docview/869572711/?pq-origsite=primo$$EView_record_in_ProQuest_(subscribers_only)
search
creatorcontrib
0Anderson, D
1Luke, R
2Keller, J
titleSpeedup of Fuzzy Clustering Through Stream Processing on Graphics Processing Units
descriptionAs the number of data points, feature dimensionality, and number of centers for clustering algorithms increase, computational tractability becomes a problem. The fuzzy c-means has a large degree of inherent algorithmic parallelism that modern CPU architectures do not exploit. Many pattern recognition algorithms can be sped up on a graphics processing unit (GPU) as long as the majority of computation at various stages and the components are not dependent on each other. We present a generalized method for offloading fuzzy clustering to a GPU, while maintaining control over the number of data points, feature dimensionality, and the number of cluster centers. GPU-based clustering is a high-performance low-cost solution that frees up the CPU. Our results show a speed increase of over two orders of magnitude for particular clustering configurations and platforms.
subject
0Algorithms
1Architecture
2General (551.5)
3M2 551.5
general
0English
110.1109/TFUZZ.2008.924203
2ProQuest Atmospheric Science Collection
3ProQuest Natural Science Collection
4ProQuest SciTech Collection
5Earth, Atmospheric & Aquatic Science Database
6Natural Science Collection
7SciTech Premium Collection
sourceidproquest
recordidproquest869572711
issn
010636706
11063-6706
rsrctypearticle
creationdate2008
addtitleIEEE Transactions on Fuzzy Systems
searchscope
01007529
11007944
210000032
310000037
410000050
510000120
610000200
710000202
810000209
910000244
1010000253
1110000260
12proquest
scope
01007529
11007944
210000032
310000037
410000050
510000120
610000200
710000202
810000209
910000244
1010000253
1110000260
12proquest
lsr43
01007529false
11007944false
210000032false
310000037false
410000050false
510000120false
610000200false
710000202false
810000209false
910000244false
1010000253false
1110000260false
contributorAnderson, D
startdate20080101
enddate20080101
citationvol 16 issue 4
lsr30VSR-Enriched:[pqid, pages, eissn]
sort
titleSpeedup of Fuzzy Clustering Through Stream Processing on Graphics Processing Units
authorAnderson, D ; Luke, R ; Keller, J
creationdate20080101
lso0120080101
facets
frbrgroupid6253864655415541902
frbrtype5
languageeng
creationdate2008
topic
0Algorithms
1Architecture
2General (551.5)
collection
0ProQuest Atmospheric Science Collection
1ProQuest Natural Science Collection
2ProQuest SciTech Collection
3Earth, Atmospheric & Aquatic Science Database
4Natural Science Collection
5SciTech Premium Collection
prefilterarticles
rsrctypearticles
creatorcontrib
0Anderson, D
1Luke, R
2Keller, J
jtitleIEEE Transactions on Fuzzy Systems
toplevelpeer_reviewed
delivery
delcategoryRemote Search Resource
fulltextfulltext
addata
aulast
0Anderson
1Luke
2Keller
aufirst
0D
1R
2J
au
0Anderson, D
1Luke, R
2Keller, J
addauAnderson, D
atitleSpeedup of Fuzzy Clustering Through Stream Processing on Graphics Processing Units
jtitleIEEE Transactions on Fuzzy Systems
risdate20080101
volume16
issue4
issn1063-6706
formatjournal
genrearticle
ristypeJOUR
abstractAs the number of data points, feature dimensionality, and number of centers for clustering algorithms increase, computational tractability becomes a problem. The fuzzy c-means has a large degree of inherent algorithmic parallelism that modern CPU architectures do not exploit. Many pattern recognition algorithms can be sped up on a graphics processing unit (GPU) as long as the majority of computation at various stages and the components are not dependent on each other. We present a generalized method for offloading fuzzy clustering to a GPU, while maintaining control over the number of data points, feature dimensionality, and the number of cluster centers. GPU-based clustering is a high-performance low-cost solution that frees up the CPU. Our results show a speed increase of over two orders of magnitude for particular clustering configurations and platforms.
doi10.1109/TFUZZ.2008.924203
urlhttp://search.proquest.com/docview/869572711/
pages1101-1106
eissn19410034
date2008-01-01