Mars: A MapReduce Framework on graphics processors

We design and implement Mars, a MapReduce framework, on graphics processors (GPUs). MapReduce is a distributed programming framework originally proposed by Google for the ease of development of web search applications on a large number of commodity CPUs. Compared with CPUs, GPUs have an order of mag...

Full description

Saved in:
Bibliographic Details
Published in:PACT'08 : proceedings of the Seventeenth International Conference on Parallel Architectures and Compilation Techniques : Toronto, Ontario, Canada, October 25-29, 2008 pp. 260 - 269
Main Authors: He, Bingsheng, Fang, Wenbin, Luo, Qiong, Govindaraju, Naga K., Wang, Tuyong
Format: Conference Proceeding
Language:English
Published: ACM 01.10.2008
Subjects:
Computer architecture
Data parallelism
GPGPU
Graphics processing units
Graphics Processor
Instruction sets
MapReduce
Multi-core processors
Parallel processing
Programming
Runtime
Web Analysis
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract We design and implement Mars, a MapReduce framework, on graphics processors (GPUs). MapReduce is a distributed programming framework originally proposed by Google for the ease of development of web search applications on a large number of commodity CPUs. Compared with CPUs, GPUs have an order of magnitude higher computation power and memory bandwidth, but are harder to program since their architectures are designed as a special-purpose co-processor and their programming interfaces are typically for graphics applications. As the first attempt to harness GPU's power for MapReduce, we developed Mars on an NVIDIA G80 GPU, which contains over one hundred processors, and evaluated it in comparison with Phoenix, the state-of-the-art MapReduce framework on multi-core CPUs. Mars hides the programming complexity of the GPU behind the simple and familiar MapReduce interface. It is up to 16 times faster than its CPU-based counterpart for six common web applications on a quad-core machine.
AbstractList We design and implement Mars, a MapReduce framework, on graphics processors (GPUs). MapReduce is a distributed programming framework originally proposed by Google for the ease of development of web search applications on a large number of commodity CPUs. Compared with CPUs, GPUs have an order of magnitude higher computation power and memory bandwidth, but are harder to program since their architectures are designed as a special-purpose co-processor and their programming interfaces are typically for graphics applications. As the first attempt to harness GPU's power for MapReduce, we developed Mars on an NVIDIA G80 GPU, which contains over one hundred processors, and evaluated it in comparison with Phoenix, the state-of-the-art MapReduce framework on multi-core CPUs. Mars hides the programming complexity of the GPU behind the simple and familiar MapReduce interface. It is up to 16 times faster than its CPU-based counterpart for six common web applications on a quad-core machine.
Author Govindaraju, Naga K.
He, Bingsheng
Fang, Wenbin
Luo, Qiong
Wang, Tuyong
Author_xml – sequence: 1
  givenname: Bingsheng
  surname: He
  fullname: He, Bingsheng
  organization: HKUST, China
– sequence: 2
  givenname: Wenbin
  surname: Fang
  fullname: Fang, Wenbin
  organization: HKUST, China
– sequence: 3
  givenname: Qiong
  surname: Luo
  fullname: Luo, Qiong
  organization: HKUST, China
– sequence: 4
  givenname: Naga K.
  surname: Govindaraju
  fullname: Govindaraju, Naga K.
  organization: Microsoft Corp., USA
– sequence: 5
  givenname: Tuyong
  surname: Wang
  fullname: Wang, Tuyong
  organization: Sina Corp., China
BookMark eNotzEFLAzEQBeAICmrdswcv-QNbk2ySmfVWilWhRRA9l9l0oqt2sySK-O9d1Afvfbd3Kg6HNLAQ51rNtbbucqrV2s1_deZAVC2g9so5NGjcsahKeVVTwDXTnAizoVyu5EJuaHzg3Wdgucq056-U32Qa5HOm8aUPRY45BS4l5XImjiK9F67-nYmn1fXj8rZe39_cLRfrmoyFj9oEUsGrXYsheuMgasuhAybULbbKd4qRIzgfLTZRwyREaztAQxE9NTNx8ffbM_N2zP2e8vcW0LbW6eYHnVRDtA
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1145/1454115.1454152
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Xplore POP ALL
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISBN 9781605582825
1605582824
EndPage 269
ExternalDocumentID 7849451
Genre orig-research
GroupedDBID 6IE
6IL
ACM
ALMA_UNASSIGNED_HOLDINGS
APO
CBEJK
GUFHI
LHSKQ
RIE
RIL
ID FETCH-LOGICAL-a247t-2ca0c60d98cf6257f14ecb7ea8198906b0e8ef756f483f176f47f44b782af86a3
IEDL.DBID RIE
ISICitedReferencesCount 331
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000267053400026&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Aug 27 01:42:04 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-a247t-2ca0c60d98cf6257f14ecb7ea8198906b0e8ef756f483f176f47f44b782af86a3
PageCount 10
ParticipantIDs ieee_primary_7849451
PublicationCentury 2000
PublicationDate 2008-Oct.
PublicationDateYYYYMMDD 2008-10-01
PublicationDate_xml – month: 10
  year: 2008
  text: 2008-Oct.
PublicationDecade 2000
PublicationTitle PACT'08 : proceedings of the Seventeenth International Conference on Parallel Architectures and Compilation Techniques : Toronto, Ontario, Canada, October 25-29, 2008
PublicationTitleAbbrev PACT
PublicationYear 2008
Publisher ACM
Publisher_xml – name: ACM
SSID ssj0000753007
Score 2.3213832
Snippet We design and implement Mars, a MapReduce framework, on graphics processors (GPUs). MapReduce is a distributed programming framework originally proposed by...
SourceID ieee
SourceType Publisher
StartPage 260
SubjectTerms Computer architecture
Data parallelism
GPGPU
Graphics processing units
Graphics Processor
Instruction sets
MapReduce
Multi-core processors
Parallel processing
Programming
Runtime
Web Analysis
Title Mars: A MapReduce Framework on graphics processors
URI https://ieeexplore.ieee.org/document/7849451
WOSCitedRecordID wos000267053400026&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwED6VioGpQIt4ywMjaZPY8YMNISqWVhUCqVvlOGepS1I1Lb8f20kKAwuDH_Ji-aX7bN_3HcCDgwxSZjKPnDGiEXOQJMoz4_aygw-xEVbpVIdgE2I-l8ulWvTg8cCFQcTgfIZjXw1_-UVl9v6pbCIkU8zzpY-E4A1X6_Ce4kwfdVmr3pOwbOISc3hnHEpPLPoVPiVYj-ngf_2ewuiHhkcWBwNzBj0sz2HQxWEg7bEcQjpzt9Mn8kxmevPupViRTDufK1KVJIhSr01NNg0roNrWI_icvn68vEVtLIRIp0zsotTo2PC4UNJYd2URNmFocoFaeqenmOcxSrQi45ZJahPhSmEZyx0A0FZyTS-gX1YlXgLRurAaaYKZ0iznShnKKecpKlMoodUVDP0UrDaN3MWqHf313803cJJ2ErHJLfR32z3ewbH52q3r7X1Yo2_ro5GF
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV27TsMwFL2qChJMBVrEGw-MpE1ixw82hKiKaKsKFalb5TjXUpck6oPvx07TwsDC4NjKEvkR3WP7nnMAHhxkkDKRaeCCEQ2YgyRBmhi3lh18CI2wSse6MpsQ47GczdSkAY97LgwiVsln2PXN6i4_K8zGH5X1hGSKeb70gXfOqtla-xMVF_yoe9T6PRFLeq4wh3i6Ve2pRb8MVKr40W_978sn0Pkh4pHJPsScQgPzM2jtnBhI_WO2IR65_ekTeSYjXX54MVYk_V3WFSlyUslSL8yKlFteQLFcdeCz_zp9GQS1G0KgYybWQWx0aHiYKWms27QIGzE0qUAtfdpTyNMQJVqRcMsktZFwtbCMpQ4CaCu5pufQzIscL4BonVmNNMJEaZZypQzllPMYlcmU0OoS2n4I5uVW8GJe9_7q79f3cDSYjobz4dv4_RqO451gbHQDzfVyg7dwaL7Wi9Xyrpqvb7LGlM4
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=PACT%2708+%3A+proceedings+of+the+Seventeenth+International+Conference+on+Parallel+Architectures+and+Compilation+Techniques+%3A+Toronto%2C+Ontario%2C+Canada%2C+October+25-29%2C+2008&rft.atitle=Mars%3A+A+MapReduce+Framework+on+graphics+processors&rft.au=He%2C+Bingsheng&rft.au=Fang%2C+Wenbin&rft.au=Luo%2C+Qiong&rft.au=Govindaraju%2C+Naga+K.&rft.date=2008-10-01&rft.pub=ACM&rft.spage=260&rft.epage=269&rft_id=info:doi/10.1145%2F1454115.1454152&rft.externalDocID=7849451