Sparse Matrix-Vector Multiplication on GPGPUs

The multiplication of a sparse matrix by a dense vector (SpMV) is a centerpiece of scientific computing applications: it is the essential kernel for the solution of sparse linear systems and sparse eigenvalue problems by iterative methods. The efficient implementation of the sparse matrix-vector mul...

Full description

Saved in:
Bibliographic Details
Published inACM transactions on mathematical software Vol. 43; no. 4; pp. 1 - 49
Main Authors Filippone, Salvatore, Cardellini, Valeria, Barbieri, Davide, Fanfarillo, Alessandro
Format Journal Article
LanguageEnglish
Published 31.12.2017
Online AccessGet full text
ISSN0098-3500
1557-7295
1557-7295
DOI10.1145/3017994

Cover

Abstract The multiplication of a sparse matrix by a dense vector (SpMV) is a centerpiece of scientific computing applications: it is the essential kernel for the solution of sparse linear systems and sparse eigenvalue problems by iterative methods. The efficient implementation of the sparse matrix-vector multiplication is therefore crucial and has been the subject of an immense amount of research, with interest renewed with every major new trend in high-performance computing architectures. The introduction of General-Purpose Graphics Processing Units (GPGPUs) is no exception, and many articles have been devoted to this problem. With this article, we provide a review of the techniques for implementing the SpMV kernel on GPGPUs that have appeared in the literature of the last few years. We discuss the issues and tradeoffs that have been encountered by the various researchers, and a list of solutions, organized in categories according to common features. We also provide a performance comparison across different GPGPU models and on a set of test matrices coming from various application domains.
AbstractList The multiplication of a sparse matrix by a dense vector (SpMV) is a centerpiece of scientific computing applications: it is the essential kernel for the solution of sparse linear systems and sparse eigenvalue problems by iterative methods. The efficient implementation of the sparse matrix-vector multiplication is therefore crucial and has been the subject of an immense amount of research, with interest renewed with every major new trend in high-performance computing architectures. The introduction of General-Purpose Graphics Processing Units (GPGPUs) is no exception, and many articles have been devoted to this problem. With this article, we provide a review of the techniques for implementing the SpMV kernel on GPGPUs that have appeared in the literature of the last few years. We discuss the issues and tradeoffs that have been encountered by the various researchers, and a list of solutions, organized in categories according to common features. We also provide a performance comparison across different GPGPU models and on a set of test matrices coming from various application domains.
Author Fanfarillo, Alessandro
Cardellini, Valeria
Filippone, Salvatore
Barbieri, Davide
Author_xml – sequence: 1
  givenname: Salvatore
  surname: Filippone
  fullname: Filippone, Salvatore
  organization: Cranfield University, Cranfield, United Kingdom
– sequence: 2
  givenname: Valeria
  surname: Cardellini
  fullname: Cardellini, Valeria
  organization: Università degli Studi di Roma “Tor Vergata”, Roma, Italy
– sequence: 3
  givenname: Davide
  surname: Barbieri
  fullname: Barbieri, Davide
  organization: Università degli Studi di Roma “Tor Vergata”, Roma, Italy
– sequence: 4
  givenname: Alessandro
  surname: Fanfarillo
  fullname: Fanfarillo, Alessandro
  organization: Università degli Studi di Roma “Tor Vergata”, Roma, Italy
BookMark eNp9z0FLwzAUwPEgE9ym-BV600s0SZOmOcrQKmw40HktL2kCkdqWJEP37VfdTgrCg3d4Px78Z2jS9Z1F6JKSG0q5uM0JlUrxEzSlQkgsmRITNCVElTgXhJyhWYzvhBBGJZ0i_DJAiDZbQQr-C79Zk_qQrbZt8kPrDSTfd9k41bpab-I5OnXQRntx3HO0ebh_XTzi5XP1tLhbYsNEmTAzTakI16QRotCi0cKAlNKAcLRodON0CeNNW2adLlyhtVUFNww4ZXkueT5H14e_226A3Se0bT0E_wFhV1NSf2fWx8yRXh2oCX2Mwbp_JP4ljU8_gSmAb__4PVNHYUw
CitedBy_id crossref_primary_10_1134_S1064230718010033
crossref_primary_10_1016_j_jocs_2022_101609
crossref_primary_10_1142_S0129626419500014
crossref_primary_10_3390_electronics12173687
crossref_primary_10_1016_j_compfluid_2024_106247
crossref_primary_10_1016_j_jpdc_2021_07_007
crossref_primary_10_1109_TPDS_2022_3223512
crossref_primary_10_1177_1094342019886628
crossref_primary_10_1007_s12046_018_0892_0
crossref_primary_10_1016_j_softx_2024_101775
crossref_primary_10_1089_3dp_2024_0165
crossref_primary_10_1109_TPDS_2023_3287238
crossref_primary_10_1002_cpe_5207
crossref_primary_10_1007_s00466_018_1623_4
crossref_primary_10_1016_j_jcp_2023_112133
crossref_primary_10_1016_j_cma_2022_115598
crossref_primary_10_1145_3226228
crossref_primary_10_1145_3466795
crossref_primary_10_1145_3414685_3417763
crossref_primary_10_1016_j_jpdc_2017_10_013
crossref_primary_10_1109_TPDS_2022_3177291
crossref_primary_10_1088_1742_6596_1828_1_012013
crossref_primary_10_1002_cpe_4701
crossref_primary_10_1145_3571157
crossref_primary_10_1109_TPDS_2019_2907537
crossref_primary_10_1109_ACCESS_2018_2871219
crossref_primary_10_21205_deufmd_2020226501
crossref_primary_10_1002_cpe_6230
crossref_primary_10_1016_j_jcde_2018_11_001
crossref_primary_10_3389_fninf_2022_877945
crossref_primary_10_1007_s11227_020_03489_3
crossref_primary_10_1145_3372153
crossref_primary_10_1145_3218823
crossref_primary_10_1109_ACCESS_2024_3425638
crossref_primary_10_1145_3572774
crossref_primary_10_1002_nme_7421
crossref_primary_10_1002_cpe_6515
crossref_primary_10_1016_j_jpdc_2020_05_020
crossref_primary_10_1016_j_matcom_2023_09_016
crossref_primary_10_1177_1094342019879985
crossref_primary_10_1109_ACCESS_2023_3317293
crossref_primary_10_1145_3232850
crossref_primary_10_1002_acm2_12835
crossref_primary_10_1002_cpe_7871
crossref_primary_10_1145_3722114
crossref_primary_10_1016_j_future_2020_02_076
crossref_primary_10_1007_s11227_024_06394_1
crossref_primary_10_1002_cpe_8366
crossref_primary_10_1007_s00500_023_08125_9
crossref_primary_10_3390_electronics13203981
crossref_primary_10_3390_electronics9101675
crossref_primary_10_1007_s11227_024_05949_6
crossref_primary_10_1109_ACCESS_2024_3449914
crossref_primary_10_1016_j_cirpj_2022_12_012
crossref_primary_10_1137_20M134914X
crossref_primary_10_1145_3128571
crossref_primary_10_1016_j_parco_2021_102831
crossref_primary_10_1145_3380930
crossref_primary_10_1093_jcde_qwae024
crossref_primary_10_1109_TPDS_2024_3477431
Cites_doi 10.1016/j.jpdc.2016.03.011
10.1109/HiPC.2015.55
10.1109/ASAP.2015.7245713
10.1137/130930352
10.1109/ICPP.2011.53
10.1007/s101070100263
10.1109/SC.2014.68
10.1145/2751205.2751244
10.1109/ICCASM.2010.5623237
10.1080/17445760802337010
10.1109/ICCSE.2012.28
10.2528/PIER11031607
10.1007/978-3-642-39640-3_15
10.1111/j.1467-8659.2012.03227.x
10.1145/1837853.1693471
10.1016/j.parco.2011.08.003
10.1016/j.parco.2012.07.002
10.1109/TC.2014.2366731
10.1137/120900216
10.1007/3-540-60902-4_12
10.1145/275323.275327
10.1137/1.9781611970944
10.1145/2491956.2462181
10.1016/j.jpdc.2014.11.001
10.5555/323215
10.1137/1.9780898717839
10.1145/2304576.2304624
10.1016/j.jpdc.2014.03.002
10.1145/2751205.2751209
10.1109/ICPADS.2011.91
10.1109/SNPD.2012.20
10.1109/HIPC.2009.5433184
10.1155/2014/469753
10.1007/978-3-642-38750-0_12
10.1109/IPDPSW.2012.211
10.1137/12088313X
10.1145/2049662.2049663
10.1145/1498765.1498785
10.1007/s11227-012-0825-3
10.1137/1.9781611970937
10.1145/882262.882364
10.1002/cpe.2978
10.1002/cpe.1732
10.1007/978-3-642-11515-8_10
10.1145/992200.992206
10.1109/ICCIS.2010.285
10.1109/ICPP.2011.82
10.1109/SC.2014.69
10.1145/2331130.2331131
10.1109/TPDS.2015.2401575
10.2172/7093021
10.1145/2159430.2159436
10.1109/HPCC.2010.85
10.1109/HPCC.2010.67
10.1109/HPCC.2012.193
10.14778/1938545.1938548
10.1109/TPDS.2014.2308221
10.1002/cpe.1658
10.1109/IPEMC.2012.6259153
10.1177/1094342016646844
10.1002/cpe.3217
10.1145/1188455.1188672
10.1109/HPCC.2012.192
10.1007/978-3-642-03138-0_32
10.1109/ICPP.2013.10
10.5555/2606265.2606948
10.1109/HPCC.2012.68
10.1002/nme.4865
10.1137/1.9780898718003
10.1016/j.parco.2008.12.006
10.1016/j.micpro.2011.05.005
10.1109/TPDS.2014.2357437
10.1007/978-3-642-01970-8_90
10.1201/b10376-8
10.1109/TPDS.2013.123
10.1007/978-3-540-85268-1
10.1177/1094342013501126
10.1016/j.cam.2015.04.035
10.1145/1964179.1964196
10.1145/2503210.2503234
10.1109/IPDPSW.2013.271
10.1016/j.parco.2013.09.005
10.1109/IPDPSW.2014.106
10.1145/1365490.1365500
10.1109/ICPP.2009.21
10.5555/2029256.2029301
10.1109/TMAG.2010.2043511
10.1145/2555243.2555255
10.5555/1413370.1413402
10.1145/1654059.1654078
ContentType Journal Article
DBID AAYXX
CITATION
ADTOC
UNPAY
DOI 10.1145/3017994
DatabaseName CrossRef
Unpaywall for CDI: Periodical Content
Unpaywall
DatabaseTitle CrossRef
DatabaseTitleList CrossRef
Database_xml – sequence: 1
  dbid: UNPAY
  name: Unpaywall
  url: https://proxy.k.utb.cz/login?url=https://unpaywall.org/
  sourceTypes: Open Access Repository
DeliveryMethod fulltext_linktorsrc
Discipline Sciences (General)
Computer Science
EISSN 1557-7295
EndPage 49
ExternalDocumentID oai:dspace.lib.cranfield.ac.uk:1826/11636
10_1145_3017994
GroupedDBID --Z
-DZ
-~X
.DC
23M
2FS
4.4
5GY
5VS
6J9
6OB
85S
8US
AAIKC
AAKMM
AALFJ
AAMNW
AAYFX
AAYXX
ABFSI
ABPPZ
ACGFO
ACGOD
ACIWK
ACM
ACNCT
ADBCU
ADL
AEBYY
AEFXT
AEJOY
AENEX
AENSD
AETEA
AFWIH
AFWXC
AGHSJ
AIKLT
AKRVB
ALMA_UNASSIGNED_HOLDINGS
AMVHM
ASPBG
AVWKF
BDXCO
CCLIF
CITATION
CS3
D0L
EBS
EJD
FEDTE
GUFHI
HGAVV
H~9
I07
IAO
ICD
LHSKQ
MS~
P1C
P2P
PQQKQ
RNS
ROL
RXW
TAE
TWZ
U5U
UHB
UPT
X6Y
ZCA
9M8
ADTOC
AFFNX
AI.
E.L
HF~
IEA
IGS
IOF
ITC
MVM
NHB
OHT
UNPAY
VH1
XJT
XOL
ZY4
ID FETCH-LOGICAL-c258t-2cd8904b0d556b5db5ca777ca5f16dbdfb8a0d5be2efb6f6bbe964c2a41233743
IEDL.DBID UNPAY
ISSN 0098-3500
1557-7295
IngestDate Sun Oct 26 03:36:37 EDT 2025
Thu Apr 24 23:07:14 EDT 2025
Wed Oct 01 05:56:19 EDT 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 4
Language English
License cc-by
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c258t-2cd8904b0d556b5db5ca777ca5f16dbdfb8a0d5be2efb6f6bbe964c2a41233743
OpenAccessLink https://proxy.k.utb.cz/login?url=https://dspace.lib.cranfield.ac.uk/handle/1826/11636
PageCount 49
ParticipantIDs unpaywall_primary_10_1145_3017994
crossref_primary_10_1145_3017994
crossref_citationtrail_10_1145_3017994
ProviderPackageCode CITATION
AAYXX
PublicationCentury 2000
PublicationDate 2017-12-31
PublicationDateYYYYMMDD 2017-12-31
PublicationDate_xml – month: 12
  year: 2017
  text: 2017-12-31
  day: 31
PublicationDecade 2010
PublicationTitle ACM transactions on mathematical software
PublicationYear 2017
References Heller M. (e_1_2_1_48_1) 2012
e_1_2_1_111_1
e_1_2_1_81_1
Oberhuber T. (e_1_2_1_83_1) 2011; 56
e_1_2_1_20_1
e_1_2_1_62_1
e_1_2_1_43_1
e_1_2_1_85_1
e_1_2_1_47_1
Patankar S. V. (e_1_2_1_84_1)
e_1_2_1_92_1
e_1_2_1_103_1
e_1_2_1_107_1
e_1_2_1_31_1
e_1_2_1_54_1
e_1_2_1_77_1
e_1_2_1_8_1
Neelima B. (e_1_2_1_80_1) 2012; 13
e_1_2_1_12_1
e_1_2_1_73_1
e_1_2_1_96_1
e_1_2_1_4_1
e_1_2_1_110_1
e_1_2_1_16_1
e_1_2_1_39_1
e_1_2_1_58_1
e_1_2_1_82_1
e_1_2_1_114_1
e_1_2_1_40_1
e_1_2_1_67_1
e_1_2_1_21_1
e_1_2_1_63_1
e_1_2_1_86_1
e_1_2_1_25_1
e_1_2_1_29_1
Cardellini V. (e_1_2_1_18_1); 25
Davis T. (e_1_2_1_28_1) 2007; 7
e_1_2_1_93_1
e_1_2_1_70_1
e_1_2_1_102_1
e_1_2_1_106_1
e_1_2_1_7_1
e_1_2_1_55_1
e_1_2_1_78_1
e_1_2_1_3_1
e_1_2_1_13_1
e_1_2_1_51_1
e_1_2_1_97_1
e_1_2_1_32_1
Hwu W. W. (e_1_2_1_50_1)
e_1_2_1_74_1
Anzt H. (e_1_2_1_6_1) 2014
e_1_2_1_17_1
e_1_2_1_36_1
e_1_2_1_59_1
Wang Z. (e_1_2_1_100_1); 4
e_1_2_1_60_1
e_1_2_1_113_1
e_1_2_1_41_1
e_1_2_1_68_1
Baskaran M. M. (e_1_2_1_11_1) 2009
CUSP. (e_1_2_1_24_1) 2016
e_1_2_1_45_1
e_1_2_1_22_1
e_1_2_1_64_1
e_1_2_1_49_1
e_1_2_1_26_1
e_1_2_1_109_1
e_1_2_1_101_1
e_1_2_1_71_1
e_1_2_1_90_1
Reguly I. (e_1_2_1_87_1)
e_1_2_1_105_1
e_1_2_1_56_1
e_1_2_1_79_1
e_1_2_1_98_1
e_1_2_1_10_1
e_1_2_1_33_1
e_1_2_1_52_1
e_1_2_1_75_1
e_1_2_1_94_1
e_1_2_1_2_1
e_1_2_1_14_1
e_1_2_1_37_1
e_1_2_1_112_1
e_1_2_1_42_1
e_1_2_1_65_1
e_1_2_1_88_1
e_1_2_1_23_1
e_1_2_1_46_1
e_1_2_1_27_1
Guo D. (e_1_2_1_44_1)
e_1_2_1_69_1
e_1_2_1_108_1
e_1_2_1_91_1
e_1_2_1_104_1
e_1_2_1_30_1
e_1_2_1_76_1
e_1_2_1_5_1
e_1_2_1_57_1
e_1_2_1_99_1
Sanders J. (e_1_2_1_89_1) 2010
e_1_2_1_34_1
e_1_2_1_72_1
e_1_2_1_1_1
e_1_2_1_53_1
e_1_2_1_95_1
e_1_2_1_38_1
Langr D. (e_1_2_1_61_1)
e_1_2_1_15_1
e_1_2_1_9_1
e_1_2_1_19_1
Liu H. (e_1_2_1_66_1) 2012; 3
References_xml – ident: e_1_2_1_72_1
  doi: 10.1016/j.jpdc.2016.03.011
– ident: e_1_2_1_25_1
  doi: 10.1109/HiPC.2015.55
– ident: e_1_2_1_68_1
  doi: 10.1109/ASAP.2015.7245713
– ident: e_1_2_1_57_1
  doi: 10.1137/130930352
– ident: e_1_2_1_82_1
– ident: e_1_2_1_93_1
  doi: 10.1109/ICPP.2011.53
– ident: e_1_2_1_32_1
  doi: 10.1007/s101070100263
– ident: e_1_2_1_41_1
  doi: 10.1109/SC.2014.68
– volume: 13
  start-page: 159
  year: 2012
  ident: e_1_2_1_80_1
  article-title: New sparse matrix storage format to improve the performance of total SpMV time
  publication-title: Scalable Comput.: Pract. Exper.
– ident: e_1_2_1_90_1
  doi: 10.1145/2751205.2751244
– ident: e_1_2_1_17_1
  doi: 10.1109/ICCASM.2010.5623237
– ident: e_1_2_1_16_1
  doi: 10.1080/17445760802337010
– ident: e_1_2_1_112_1
  doi: 10.1109/ICCSE.2012.28
– ident: e_1_2_1_34_1
  doi: 10.2528/PIER11031607
– ident: e_1_2_1_79_1
  doi: 10.1007/978-3-642-39640-3_15
– volume-title: Proc. of Extreme Scaling Workshop (BW-XSEDE’12)
  ident: e_1_2_1_44_1
– ident: e_1_2_1_101_1
  doi: 10.1111/j.1467-8659.2012.03227.x
– ident: e_1_2_1_22_1
  doi: 10.1145/1837853.1693471
– volume-title: Example: An Introduction to General-purpose GPU Programming
  year: 2010
  ident: e_1_2_1_89_1
– ident: e_1_2_1_96_1
  doi: 10.1016/j.parco.2011.08.003
– ident: e_1_2_1_98_1
  doi: 10.1016/j.parco.2012.07.002
– ident: e_1_2_1_110_1
  doi: 10.1109/TC.2014.2366731
– ident: e_1_2_1_55_1
  doi: 10.1137/120900216
– volume: 3
  start-page: 185
  year: 2012
  ident: e_1_2_1_66_1
  article-title: Sparse matrix-vector multiplication on NVIDIA GPU
  publication-title: Int. J. Numer. Analysis Model. Series B
– ident: e_1_2_1_21_1
  doi: 10.1007/3-540-60902-4_12
– ident: e_1_2_1_33_1
  doi: 10.1145/275323.275327
– ident: e_1_2_1_53_1
  doi: 10.1137/1.9781611970944
– ident: e_1_2_1_63_1
  doi: 10.1145/2491956.2462181
– ident: e_1_2_1_8_1
  doi: 10.1016/j.jpdc.2014.11.001
– ident: e_1_2_1_4_1
  doi: 10.5555/323215
– volume-title: Proc. of Innovative Parallel Computing (InPar’12)
  ident: e_1_2_1_87_1
– ident: e_1_2_1_62_1
  doi: 10.1137/1.9780898717839
– ident: e_1_2_1_92_1
  doi: 10.1145/2304576.2304624
– ident: e_1_2_1_114_1
  doi: 10.1016/j.jpdc.2014.03.002
– ident: e_1_2_1_67_1
  doi: 10.1145/2751205.2751209
– volume: 4
  volume-title: Proc. of 2nd Int’l Conf. on Education Technology and Computer (ICETC’10)
  ident: e_1_2_1_100_1
– ident: e_1_2_1_37_1
  doi: 10.1109/ICPADS.2011.91
– ident: e_1_2_1_106_1
  doi: 10.1109/SNPD.2012.20
– ident: e_1_2_1_75_1
  doi: 10.1109/HIPC.2009.5433184
– ident: e_1_2_1_19_1
  doi: 10.1155/2014/469753
– ident: e_1_2_1_2_1
  doi: 10.1007/978-3-642-38750-0_12
– ident: e_1_2_1_56_1
  doi: 10.1109/IPDPSW.2012.211
– ident: e_1_2_1_59_1
  doi: 10.1137/12088313X
– ident: e_1_2_1_30_1
  doi: 10.1145/2049662.2049663
– volume: 7
  start-page: 12
  year: 2007
  ident: e_1_2_1_28_1
  article-title: Wilkinson’s sparse matrix definition
  publication-title: NA Digest
– ident: e_1_2_1_104_1
  doi: 10.1145/1498765.1498785
– volume-title: CUSP : A C++ Templated Sparse Matrix Library
  year: 2016
  ident: e_1_2_1_24_1
– volume-title: Technical Report EECS-14-727. University of Tennessee.
  year: 2014
  ident: e_1_2_1_6_1
– ident: e_1_2_1_65_1
  doi: 10.1007/s11227-012-0825-3
– ident: e_1_2_1_42_1
  doi: 10.1137/1.9781611970937
– ident: e_1_2_1_15_1
  doi: 10.1145/882262.882364
– ident: e_1_2_1_38_1
  doi: 10.1002/cpe.2978
– ident: e_1_2_1_74_1
– ident: e_1_2_1_36_1
  doi: 10.1002/cpe.1732
– ident: e_1_2_1_78_1
  doi: 10.1007/978-3-642-11515-8_10
– ident: e_1_2_1_29_1
  doi: 10.1145/992200.992206
– ident: e_1_2_1_45_1
  doi: 10.1109/ICCIS.2010.285
– ident: e_1_2_1_76_1
  doi: 10.1109/ICPP.2011.82
– ident: e_1_2_1_7_1
  doi: 10.1109/SC.2014.69
– ident: e_1_2_1_39_1
  doi: 10.1145/2331130.2331131
– ident: e_1_2_1_60_1
  doi: 10.1109/TPDS.2015.2401575
– ident: e_1_2_1_54_1
  doi: 10.2172/7093021
– ident: e_1_2_1_12_1
– ident: e_1_2_1_40_1
  doi: 10.1145/2159430.2159436
– ident: e_1_2_1_49_1
  doi: 10.1109/HPCC.2010.85
– ident: e_1_2_1_113_1
  doi: 10.1109/HPCC.2010.67
– ident: e_1_2_1_3_1
  doi: 10.1109/HPCC.2012.193
– ident: e_1_2_1_111_1
  doi: 10.14778/1938545.1938548
– ident: e_1_2_1_64_1
  doi: 10.1109/TPDS.2014.2308221
– ident: e_1_2_1_97_1
  doi: 10.1002/cpe.1658
– volume-title: Proc. Algoritmy
  year: 2012
  ident: e_1_2_1_48_1
– ident: e_1_2_1_108_1
  doi: 10.1109/IPEMC.2012.6259153
– ident: e_1_2_1_5_1
  doi: 10.1177/1094342016646844
– volume-title: Proc. of 2013 Federated Conference on Computer Science and Information Systems (FedCSIS’13)
  ident: e_1_2_1_61_1
– ident: e_1_2_1_46_1
  doi: 10.1002/cpe.3217
– ident: e_1_2_1_13_1
– ident: e_1_2_1_69_1
  doi: 10.1145/1188455.1188672
– ident: e_1_2_1_91_1
  doi: 10.1109/HPCC.2012.192
– ident: e_1_2_1_77_1
  doi: 10.1007/978-3-642-03138-0_32
– ident: e_1_2_1_10_1
– volume: 25
  volume-title: Advances in Parallel Computing
  ident: e_1_2_1_18_1
– ident: e_1_2_1_70_1
  doi: 10.1109/ICPP.2013.10
– ident: e_1_2_1_51_1
  doi: 10.5555/2606265.2606948
– ident: e_1_2_1_1_1
  doi: 10.1109/HPCC.2012.68
– volume: 56
  start-page: 447
  year: 2011
  ident: e_1_2_1_83_1
  article-title: New row-grouped CSR format for storing the sparse matrices on GPU with implementation in CUDA
  publication-title: Acta Technica
– ident: e_1_2_1_23_1
– ident: e_1_2_1_105_1
  doi: 10.1002/nme.4865
– ident: e_1_2_1_88_1
  doi: 10.1137/1.9780898718003
– ident: e_1_2_1_103_1
  doi: 10.1016/j.parco.2008.12.006
– ident: e_1_2_1_85_1
  doi: 10.1016/j.micpro.2011.05.005
– ident: e_1_2_1_94_1
  doi: 10.1109/TPDS.2014.2357437
– ident: e_1_2_1_20_1
  doi: 10.1007/978-3-642-01970-8_90
– ident: e_1_2_1_102_1
  doi: 10.1201/b10376-8
– volume-title: GPU Computing Gems Emerald Edition
  ident: e_1_2_1_50_1
– ident: e_1_2_1_47_1
  doi: 10.1109/TPDS.2013.123
– ident: e_1_2_1_86_1
  doi: 10.1007/978-3-540-85268-1
– ident: e_1_2_1_109_1
  doi: 10.1177/1094342013501126
– ident: e_1_2_1_26_1
  doi: 10.1016/j.cam.2015.04.035
– ident: e_1_2_1_43_1
  doi: 10.1145/1964179.1964196
– ident: e_1_2_1_95_1
  doi: 10.1145/2503210.2503234
– ident: e_1_2_1_73_1
  doi: 10.1109/IPDPSW.2013.271
– ident: e_1_2_1_27_1
  doi: 10.1016/j.parco.2013.09.005
– ident: e_1_2_1_71_1
  doi: 10.1109/IPDPSW.2014.106
– ident: e_1_2_1_81_1
  doi: 10.1145/1365490.1365500
– ident: e_1_2_1_52_1
  doi: 10.1109/ICPP.2009.21
– ident: e_1_2_1_58_1
  doi: 10.5555/2029256.2029301
– ident: e_1_2_1_31_1
  doi: 10.1109/TMAG.2010.2043511
– volume-title: Numerical Heat Transfer and Fluid Flow
  ident: e_1_2_1_84_1
– ident: e_1_2_1_107_1
  doi: 10.1145/2555243.2555255
– ident: e_1_2_1_99_1
  doi: 10.5555/1413370.1413402
– ident: e_1_2_1_9_1
– volume-title: Technical Report RC24704. IBM Research.
  year: 2009
  ident: e_1_2_1_11_1
– ident: e_1_2_1_14_1
  doi: 10.1145/1654059.1654078
SSID ssj0002171
Score 2.5350487
Snippet The multiplication of a sparse matrix by a dense vector (SpMV) is a centerpiece of scientific computing applications: it is the essential kernel for the...
SourceID unpaywall
crossref
SourceType Open Access Repository
Enrichment Source
Index Database
StartPage 1
Title Sparse Matrix-Vector Multiplication on GPGPUs
URI https://dspace.lib.cranfield.ac.uk/handle/1826/11636
UnpaywallVersion submittedVersion
Volume 43
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVEBS
  databaseName: Mathematics Source
  customDbUrl:
  eissn: 1557-7295
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0002171
  issn: 1557-7295
  databaseCode: AMVHM
  dateStart: 20110301
  isFulltext: true
  titleUrlDefault: https://www.ebsco.com/products/research-databases/mathematics-source
  providerName: EBSCOhost
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3dS8MwED90e9AX56bi_CKCiD7EdVmTto9D3IbQMdCO-TSSpnlx1OE6_PjrTdp0-IGg0IeWHvRoLtzvwv1-B3DmOIIFnlI4ULpWdTmXmHPRwYoI6ahAsHbePB4O2SBybyd0Yklhhgsj9T6Kc1qVrjt5mrdxWVZVITvQMoC41dYwgq1DlVGNwCtQjYaj7kMpj9mhOfNEZ0rPTGqlBVdWf5S2OiYAA_dLEtpYpnP-9sJns0-ZpVeDqPSpaCh5vFpm2qn3b3KN_3V6G7Ys1ETdIjbqsJakDaiVYxyQ3dUNqNu7BbqwItSXO4Dv5rrkTVBoJPxf8Tg_3Edh0X5oz_mQvvqj_iha7ELUu7m_HmA7WgHHhPoZJrH0A8cVjqSUCSoFjbnneTGnqs2kkEr4XL8TCUmUYIoJkQTMjQl3dabraNSxB5X0KU32AUmPUcmEKXWI0T31pVLENw8ej3XN24Tz8l9PY6s7bsZfzKYFJ5pO7aI0Aa0M54XUxk-T09Vi_WZz8AebQ9gkJj3ngo1HUMmel8mxBheZOIFqNxwPwhMbUx9wpswh
linkProvider Unpaywall
linkToUnpaywall http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LS8NAEB60HvRibVWsL1YQ0cPadJvdJMcitkVoKWiknso-shdLLDbFx693N9kUHwgKOSRkIEN2lvlmme8bgFPPEywKtMaRNrWqz7nCnIs21kQoT0eCtfLm8cGQ9WP_ZkzHjhRmuTDK7COZ06pM3cnTvI3LsaoK2YGmBcTNloERbBXWGDUIvAJr8XDUeSjlMds0Z56YTBnYSa204Mqaj9Jm2wZg5H9JQuuLdMbfXvh0-imzdKsQlz4VDSWPl4vMOPX-Ta7xv05vwaaDmqhTxEYNVpK0DtVyjANyu7oONXc3R-dOhPpiG_DtzJS8CRpYCf9XfJ8f7qNB0X7ozvmQuXqj3iie70Dcvb676mM3WgFLQsMME6nCyPOFpyhlgipBJQ-CQHKqW0wJpUXIzTuRkEQLppkQScR8SbhvMl3boI5dqKRPabIHSAWMKiZsqUOs7mmotCahfQi4NDVvA87Kfz2RTnfcjr-YTgpONJ24RWkAWhrOCqmNnyYny8X6zWb_DzYHsEFses4FGw-hkj0vkiMDLjJx7GLpAxdeyoU
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%3Ajournal&rft.genre=article&rft.atitle=Sparse+Matrix-Vector+Multiplication+on+GPGPUs&rft.jtitle=ACM+transactions+on+mathematical+software&rft.au=Filippone%2C+Salvatore&rft.au=Cardellini%2C+Valeria&rft.au=Barbieri%2C+Davide&rft.au=Fanfarillo%2C+Alessandro&rft.date=2017-12-31&rft.issn=0098-3500&rft.eissn=1557-7295&rft.volume=43&rft.issue=4&rft.spage=1&rft.epage=49&rft_id=info:doi/10.1145%2F3017994&rft.externalDBID=n%2Fa&rft.externalDocID=10_1145_3017994
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0098-3500&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0098-3500&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0098-3500&client=summon