CuHMMer: A load-balanced CPU-GPU cooperative bioinformatics application

GPUs have recently been used to accelerate data-parallel applications for they provide easier programmability and increased generality while maintaining the tremendous memory bandwidth and computational power. Most of those applications use CPU as a controller who decides when GPUs run the computing...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2010 International Conference on High Performance Computing and Simulation S. 24 - 30
Hauptverfasser: Ping Yao, Hong An, Mu Xu, Gu Liu, Xiaoqiang Li, Yaobin Wang, Wenting Han
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.06.2010
Schlagworte:
Computer architecture
CUDA
data-parallel computation
Graphics processing unit
Hidden Markov models
HMMER
Kernel
load-balanced
multi-threaded programming
ISBN:9781424468270, 1424468272
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract GPUs have recently been used to accelerate data-parallel applications for they provide easier programmability and increased generality while maintaining the tremendous memory bandwidth and computational power. Most of those applications use CPU as a controller who decides when GPUs run the computing-intensive tasks. This CPU-control-GPU-compute pattern wastes much of CPU's computational power. In this paper, we present a new CPU-GPU cooperative pattern for bioinformatics applications which can use both of CPU and GPU to compute. This pattern includes two parts: 1) the load-balanced data structure which manages data to keep the computational efficiency of GPU high enough when the length distribution of sequences in a sequence database is very uneven; 2) multi-threaded code partition which schedules computing on CPU and GPU in a cooperative way. Using this pattern, we develop CuHMMer based on HMMER which is one of the most important algorithms in bioinformatics. The experimental result demonstrates that CuHMMer get 13x to 45x speed up over available CPU implementations and could also outperform the traditional CUDA implementations which use CPU-control-GPU-compute pattern.
AbstractList GPUs have recently been used to accelerate data-parallel applications for they provide easier programmability and increased generality while maintaining the tremendous memory bandwidth and computational power. Most of those applications use CPU as a controller who decides when GPUs run the computing-intensive tasks. This CPU-control-GPU-compute pattern wastes much of CPU's computational power. In this paper, we present a new CPU-GPU cooperative pattern for bioinformatics applications which can use both of CPU and GPU to compute. This pattern includes two parts: 1) the load-balanced data structure which manages data to keep the computational efficiency of GPU high enough when the length distribution of sequences in a sequence database is very uneven; 2) multi-threaded code partition which schedules computing on CPU and GPU in a cooperative way. Using this pattern, we develop CuHMMer based on HMMER which is one of the most important algorithms in bioinformatics. The experimental result demonstrates that CuHMMer get 13x to 45x speed up over available CPU implementations and could also outperform the traditional CUDA implementations which use CPU-control-GPU-compute pattern.
Author Wenting Han
Ping Yao
Hong An
Mu Xu
Yaobin Wang
Gu Liu
Xiaoqiang Li
Author_xml – sequence: 1
  surname: Ping Yao
  fullname: Ping Yao
  email: yaop@mail.ustc.edu.cn
  organization: Sch. of Comput. Sci. & Technol., Univ. of Sci. & Technol. of China, Hefei, China
– sequence: 2
  surname: Hong An
  fullname: Hong An
  organization: Sch. of Comput. Sci. & Technol., Univ. of Sci. & Technol. of China, Hefei, China
– sequence: 3
  surname: Mu Xu
  fullname: Mu Xu
  organization: Sch. of Comput. Sci. & Technol., Univ. of Sci. & Technol. of China, Hefei, China
– sequence: 4
  surname: Gu Liu
  fullname: Gu Liu
  organization: Sch. of Comput. Sci. & Technol., Univ. of Sci. & Technol. of China, Hefei, China
– sequence: 5
  surname: Xiaoqiang Li
  fullname: Xiaoqiang Li
  organization: Sch. of Comput. Sci. & Technol., Univ. of Sci. & Technol. of China, Hefei, China
– sequence: 6
  surname: Yaobin Wang
  fullname: Yaobin Wang
  organization: Sch. of Comput. Sci. & Technol., Univ. of Sci. & Technol. of China, Hefei, China
– sequence: 7
  surname: Wenting Han
  fullname: Wenting Han
  email: han@ustc.edu.cn
  organization: Sch. of Comput. Sci. & Technol., Univ. of Sci. & Technol. of China, Hefei, China
BookMark eNo1j99KwzAYxSMq6GYfQLzJC3R--dMl8W4UtwobFtyux5c2gUjXlHYKvr0B57k553cuDpwZuelj7wh5ZLBgDMxzVZcfCw4Ji0IqVpgrMmOSS7nU3Jhrkhml_1nBHcmm6ROSZME5Y_dkU35Vu50bX-iKdhHb3GKHfeNaWtaHfFMfaBPj4EY8h29HbYih93E8JWwmisPQhSbl2D-QW4_d5LKLz8l-_bovq3z7vnkrV9s8GDjnxuJSe-m14bLlVngteGqAC2Zk44UCJTygYLZFr52VTnCrQIBSKaMUc_L0Nxucc8dhDCccf46X6-IX-OhNWg
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/HPCS.2010.5547159
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Xplore
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
EISBN 1424468299
9781424468300
9781424468294
1424468302
EndPage 30
ExternalDocumentID 5547159
Genre orig-research
GroupedDBID 6IE
6IF
6IK
6IL
6IN
AAJGR
AAWTH
ADFMO
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
IEGSK
IERZE
OCL
RIE
RIL
ID FETCH-LOGICAL-i90t-9ba68f4f8924d2b3f832a68023194cf37073f0a31bdaf8eb4e32b703077b4ea43
IEDL.DBID RIE
ISBN 9781424468270
1424468272
IngestDate Wed Aug 27 02:54:30 EDT 2025
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i90t-9ba68f4f8924d2b3f832a68023194cf37073f0a31bdaf8eb4e32b703077b4ea43
PageCount 7
ParticipantIDs ieee_primary_5547159
PublicationCentury 2000
PublicationDate 2010-June
PublicationDateYYYYMMDD 2010-06-01
PublicationDate_xml – month: 06
  year: 2010
  text: 2010-June
PublicationDecade 2010
PublicationTitle 2010 International Conference on High Performance Computing and Simulation
PublicationTitleAbbrev HPCSIM
PublicationYear 2010
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0000452211
Score 1.493633
Snippet GPUs have recently been used to accelerate data-parallel applications for they provide easier programmability and increased generality while maintaining the...
SourceID ieee
SourceType Publisher
StartPage 24
SubjectTerms Computer architecture
CUDA
data-parallel computation
Graphics processing unit
Hidden Markov models
HMMER
Kernel
load-balanced
multi-threaded programming
Title CuHMMer: A load-balanced CPU-GPU cooperative bioinformatics application
URI https://ieeexplore.ieee.org/document/5547159
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3LS8MwGA_b8OBJZRPf5ODRuDZpm8abFLcd3Ci4wW4jjy8wkHXMzb_fJO2mghdvSSAhD5JfvtfvQ-heO6Ehs0wRDkoRBwGSSEkFkamhsUmFkTac9CufTPL5XJQt9HCIhQGA4HwGj74YbPmm0juvKus76OMOftuozXlWx2od9CmBGjyO97FbWU453VM6NfW9VTOORH9UFm-1Y1cz6K_sKgFcBif_m9Yp6n1H6eHygD9nqAWrLhoWu9F4DJsn_IzfK2mI8r6LGgwuyhkZljOsq2oNNd83VsuqYU71bM34hzW7h6aDl2kxIk2yBLIU0ZYIJbPcJjZ38pShill3U12LZ3cTibaMu6tsI8li5XY_B5UAoyqQgXFXlgk7R51VtYILhA2FBNzPT6eZTHQaKald1xQkcy8AlfQSdf0eLNY1HcaiWf7V383X6Lg2uHvFxQ3qbDc7uEVH-nO7_NjchTP8AsEpmLA
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1JSwMxFH7UKuhJpRV3c_Bo7EyS2bzJYDtiWwZsobeS5Q0USqfU1t9vZmlV8OItCSRkIfnytu8B3GsrNPgZVzRApaiFAEmlZBGVnmGu8SIjs_Kk-8FwGE4mUdqAh10sDCKWzmf4WBRLW77J9aZQlXUs9AUWfvdg3xOCOVW01k6jUpKDu-42essPWcC2pE51fWvXdJ2ok6Txe-XaVQ_7K79KCS_d4_9N7ATa33F6JN0h0Ck0cNGCXrxJBgNcPZFnMs-loarwXtRoSJyOaS8dE53nS6wYv4ma5TV3asHXTH7Ys9sw6r6M4oTW6RLoLHLWNFLSDzORhVaiMkzxzN5V21Lwu0VCZzywlzlzJHeV3f8QlUDOVEkHFtiyFPwMmot8gedADEOB9u-nPV8K7TlKatvVQ8ntG8Aku4BWsQfTZUWIMa2Xf_l38x0cJqNBf9p_Hb5dwVFlfi_UGNfQXK82eAMH-nM9-1jdluf5BemTm_c
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=2010+International+Conference+on+High+Performance+Computing+and+Simulation&rft.atitle=CuHMMer%3A+A+load-balanced+CPU-GPU+cooperative+bioinformatics+application&rft.au=Ping+Yao&rft.au=Hong+An&rft.au=Mu+Xu&rft.au=Gu+Liu&rft.date=2010-06-01&rft.pub=IEEE&rft.isbn=9781424468270&rft.spage=24&rft.epage=30&rft_id=info:doi/10.1109%2FHPCS.2010.5547159&rft.externalDocID=5547159
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=9781424468270/lc.gif&client=summon&freeimage=true
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=9781424468270/mc.gif&client=summon&freeimage=true
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=9781424468270/sc.gif&client=summon&freeimage=true