Efficient shared memory and RDMA based collectives on multi-rail QsNet SMP clusters

Clusters of Symmetric Multiprocessors (SMP) are more commonplace than ever in achieving high-performance. Scientific applications running on clusters employ collective communications extensively. Shared memory communication and Remote Direct Memory Access (RDMA) over multi-rail networks are promisin...

Full description

Saved in:
Bibliographic Details
Published inCluster computing Vol. 11; no. 4; pp. 341 - 354
Main Authors Qian, Ying, Afsahi, Ahmad
Format Journal Article
LanguageEnglish
Published 01.12.2008
Online AccessGet full text
ISSN1386-7857
DOI10.1007/s10586-008-0065-8

Cover

Abstract Clusters of Symmetric Multiprocessors (SMP) are more commonplace than ever in achieving high-performance. Scientific applications running on clusters employ collective communications extensively. Shared memory communication and Remote Direct Memory Access (RDMA) over multi-rail networks are promising approaches in addressing the increasing demand on intra-node and inter-node communications, and thereby in boosting the performance of collectives in emerging multi-core SMP clusters. In this regard, this paper designs and evaluates two classes of collective communication algorithms directly at the Elan user-level over multi-rail Quadrics QsNet with message striping: 1) RDMA-based traditional multi-port algorithms for gather, all-gather, and all-to-all collectives for medium to large messages, and 2) RDMA-based and SMP-aware multi-port all-gather algorithms for small to medium size messages.
AbstractList Clusters of Symmetric Multiprocessors (SMP) are more commonplace than ever in achieving high-performance. Scientific applications running on clusters employ collective communications extensively. Shared memory communication and Remote Direct Memory Access (RDMA) over multi-rail networks are promising approaches in addressing the increasing demand on intra-node and inter-node communications, and thereby in boosting the performance of collectives in emerging multi-core SMP clusters. In this regard, this paper designs and evaluates two classes of collective communication algorithms directly at the Elan user-level over multi-rail Quadrics QsNet with message striping: 1) RDMA-based traditional multi-port algorithms for gather, all-gather, and all-to-all collectives for medium to large messages, and 2) RDMA-based and SMP-aware multi-port all-gather algorithms for small to medium size messages.
Author Qian, Ying
Afsahi, Ahmad
Author_xml – sequence: 1
  givenname: Ying
  surname: Qian
  fullname: Qian, Ying
– sequence: 2
  givenname: Ahmad
  surname: Afsahi
  fullname: Afsahi, Ahmad
BookMark eNotzEtLw0AUBeBZVLCt_gB3s3IXvZN5L0ttq9D6qq5LMr3ByCSpuRPBf29AF4cD34EzY5O2a5GxKwE3AsDekgDtTAbgxhiduQmbCjmKddqesxnRJwB4m_sp26-qqg41tonTR9HjkTfYdP0PL9ojf73bLXhZ0KihixFDqr-ReNfyZoipzvqijvyFHjHx_e6ZhzhQwp4u2FlVRMLL_56z9_XqbXmfbZ82D8vFNjsJZ1JmcqMDVMFaA1JVQXqV5ziqMk4578rgxzH4PPhSOGFBWqtE6XVALUpQcs6u_35Pffc1IKVDU1PAGIsWu4EOUlullHbyF_qbUgU
ContentType Journal Article
DBID 7SC
8FD
JQ2
L7M
L~C
L~D
DOI 10.1007/s10586-008-0065-8
DatabaseName Computer and Information Systems Abstracts
Technology Research Database
ProQuest Computer Science Collection
Advanced Technologies Database with Aerospace
Computer and Information Systems Abstracts – Academic
Computer and Information Systems Abstracts Professional
DatabaseTitle Computer and Information Systems Abstracts
Technology Research Database
Computer and Information Systems Abstracts – Academic
Advanced Technologies Database with Aerospace
ProQuest Computer Science Collection
Computer and Information Systems Abstracts Professional
DatabaseTitleList Computer and Information Systems Abstracts
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EndPage 354
GroupedDBID -Y2
-~C
.86
.DC
.VR
06D
0R~
0VY
1N0
203
29B
2J2
2JN
2JY
2KG
2LR
2VQ
2~H
30V
4.4
406
408
409
40D
40E
5GY
5VS
67Z
6NX
78A
7SC
8FD
8TC
8UJ
95-
95.
95~
96X
AAAVM
AABHQ
AACDK
AAHNG
AAIAL
AAJBT
AAJKR
AANZL
AAPKM
AARHV
AARTL
AASML
AATNV
AATVU
AAUYE
AAWCG
AAYIU
AAYQN
AAYTO
AAYZH
ABAKF
ABBBX
ABBRH
ABBXA
ABDBE
ABDZT
ABECU
ABFTD
ABFTV
ABHLI
ABHQN
ABJNI
ABJOX
ABKCH
ABKTR
ABMNI
ABMQK
ABNWP
ABQBU
ABQSL
ABRTQ
ABSXP
ABTEG
ABTHY
ABTKH
ABTMW
ABULA
ABWNU
ABXPI
ACAOD
ACBXY
ACDTI
ACGFS
ACHSB
ACHXU
ACKNC
ACMDZ
ACMLO
ACOKC
ACOMO
ACPIV
ACSNA
ACZOJ
ADHHG
ADHIR
ADHKG
ADKFA
ADKNI
ADKPE
ADRFC
ADTPH
ADURQ
ADYFF
ADZKW
AEBTG
AEFQL
AEGAL
AEGNC
AEJHL
AEJRE
AEKMD
AEMSY
AEOHA
AEPYU
AESKC
AETLH
AEVLU
AEXYK
AFDZB
AFGCZ
AFKRA
AFLOW
AFQWF
AFWTZ
AFZKB
AGAYW
AGDGC
AGJBK
AGMZJ
AGQEE
AGQMX
AGQPQ
AGRTI
AGWIL
AGWZB
AGYKE
AHBYD
AHKAY
AHPBZ
AHYZX
AIAKS
AIGIU
AIIXL
AILAN
AITGF
AJBLW
AJRNO
AJZVZ
ALMA_UNASSIGNED_HOLDINGS
ALWAN
AMKLP
AMXSW
AMYLF
AMYQR
AOCGG
ARAPS
ARMRJ
ASPBG
ATHPR
AVWKF
AXYYD
AYFIA
AYJHY
AZFZN
B-.
BA0
BDATZ
BENPR
BGLVJ
BGNMA
BSONS
CAG
CCPQU
COF
CS3
CSCUP
DDRTE
DL5
DNIVK
DPUIP
EBLON
EBS
EIOEI
EJD
ESBYG
FEDTE
FERAY
FFXSO
FIGPU
FINBP
FNLPD
FRRFC
FSGXE
FWDCC
GGCAI
GGRSB
GJIRD
GNWQR
GQ7
GQ8
GXS
HCIFZ
HF~
HG5
HG6
HMJXF
HQYDN
HRMNR
HVGLF
HZ~
I09
IJ-
IKXTQ
IWAJR
IXC
IXD
IXE
IZIGR
IZQ
I~X
I~Z
J-C
J0Z
JBSCW
JCJTX
JQ2
JZLTJ
K7-
KDC
KOV
L7M
LAK
LLZTM
L~C
L~D
M4Y
MA-
N2Q
NB0
NPVJJ
NQJWS
NU0
O9-
O93
O9J
OAM
P9O
PF0
PHGZM
PHGZT
PQGLB
PT4
PT5
PUEGO
QOS
R89
R9I
RNS
ROL
RPX
RSV
S16
S27
S3B
SAP
SCO
SDH
SHX
SISQX
SJYHP
SNE
SNPRN
SNX
SOHCF
SOJ
SPISZ
SRMVM
SSLCW
STPWE
SZN
T13
TSG
TSK
TSV
TUC
U2A
UG4
UOJIU
UTJUX
UZXMN
VC2
VFIZW
W23
W48
WK8
YLTOR
Z45
ZMTXR
~A9
ID FETCH-LOGICAL-p186t-6265c0fc776034fc39422e6264684898bc9fc7c92c9b1817037741b95ce51b043
ISSN 1386-7857
IngestDate Fri Sep 05 09:46:21 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 4
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-p186t-6265c0fc776034fc39422e6264684898bc9fc7c92c9b1817037741b95ce51b043
Notes ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
PQID 35744458
PQPubID 23500
PageCount 14
ParticipantIDs proquest_miscellaneous_35744458
PublicationCentury 2000
PublicationDate 2008-12-01
PublicationDateYYYYMMDD 2008-12-01
PublicationDate_xml – month: 12
  year: 2008
  text: 2008-12-01
  day: 01
PublicationDecade 2000
PublicationTitle Cluster computing
PublicationYear 2008
SSID ssj0009729
Score 1.7940525
Snippet Clusters of Symmetric Multiprocessors (SMP) are more commonplace than ever in achieving high-performance. Scientific applications running on clusters employ...
SourceID proquest
SourceType Aggregation Database
StartPage 341
Title Efficient shared memory and RDMA based collectives on multi-rail QsNet SMP clusters
URI https://www.proquest.com/docview/35744458
Volume 11
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVPQU
  databaseName: ProQuest Central
  issn: 1386-7857
  databaseCode: BENPR
  dateStart: 19980101
  customDbUrl: http://www.proquest.com/pqcentral?accountid=15518
  isFulltext: true
  dateEnd: 20241101
  titleUrlDefault: https://www.proquest.com/central
  omitProxy: true
  ssIdentifier: ssj0009729
  providerName: ProQuest
– providerCode: PRVAVX
  databaseName: SpringerLINK - Czech Republic Consortium
  issn: 1386-7857
  databaseCode: AGYKE
  dateStart: 19980101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: http://link.springer.com
  omitProxy: false
  ssIdentifier: ssj0009729
  providerName: Springer Nature
– providerCode: PRVAVX
  databaseName: SpringerLink Journals (ICM)
  issn: 1386-7857
  databaseCode: U2A
  dateStart: 19980101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: http://www.springerlink.com/journals/
  omitProxy: true
  ssIdentifier: ssj0009729
  providerName: Springer Nature
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnZ3NT9swFMCtrVx22WAf2sYAH6ZdIk9J7NjOsUARQrRj0ErdqYpdRyBB2jXphb9-z3bSBIG0sUtUvVZp1V_03vP7ROgrExk4-SYmqQkNYYbmJNPKZgpFnlMmZGhsaGA44qcTdjZNpu1GQdddUqnv-v7JvpL_oQoy4Gq7ZJ9BdnNTEMBr4AtXIAzXf2I8cPMfbDa_vHaF5He2btZPVLo8HvYDa6Ns25oLzvsBs4WvISSr7OY2-FmOTBVcDS8Cfbu2IxPKrrN65GWu7HxdNTbORklvfNj0V0fWz8vMLQgO-td32fxBNEF2KjO8AqSSEyH90OiNhow6TwLrqDvqh1bVlpP6cdCPlHLYNCkncG9XbgF-D5GtBWqy7qMfs5PJ-flsPJiOvy1_E7sbzObQ60UpL9FWDLo77KGtw8Ho4rIdrizcNrrNb2-y175F8uG3PrK5zpEYb6PX9QkA9z3OHfTCFG_Rm2a7Bq6V7Tt0taGLPV3s6WKgiy1d7OjiDl28KHBLFzu6GOjihu57NDkZjI9OSb0CgywjySsCx81Eh7kWgoeU5ZqmLI4NSBmXTKZS6RTe1GmsUxXZWYsU3PlIpYk2SaRCRj-gXrEozEeEFc-5oopmPIMjpLCLQ8H3NxGdczHXwnxCB80fMwMVY_NGWWEW63JGE8EYS-Tnv35iF71qH6ovqFet1mYPnLZK7dfM_gBB_0NA
linkProvider ProQuest
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=Efficient+shared+memory+and+RDMA+based+collectives+on+multi-rail+QsNet+SMP+clusters&rft.jtitle=Cluster+computing&rft.au=Qian%2C+Ying&rft.au=Afsahi%2C+Ahmad&rft.date=2008-12-01&rft.issn=1386-7857&rft.volume=11&rft.issue=4&rft.spage=341&rft.epage=354&rft_id=info:doi/10.1007%2Fs10586-008-0065-8&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1386-7857&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1386-7857&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1386-7857&client=summon