TOD-Tree: Task-Overlapped Direct Send Tree Image Compositing for Hybrid MPI Parallelism and GPUs

Modern supercomputers have thousands of nodes, each with CPUs and/or GPUs capable of several teraflops. However, the network connecting these nodes is relatively slow, on the order of gigabits per second. For time-critical workloads such as interactive visualization, the bottleneck is no longer comp...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on visualization and computer graphics Jg. 23; H. 6; S. 1677 - 1690
Hauptverfasser: Pascal Grosset, A. V., Prasad, Manasa, Christensen, Cameron, Knoll, Aaron, Hansen, Charles
Format: Journal Article
Sprache:Englisch
Veröffentlicht: United States IEEE 01.06.2017
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
ISSN:1077-2626, 1941-0506
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Modern supercomputers have thousands of nodes, each with CPUs and/or GPUs capable of several teraflops. However, the network connecting these nodes is relatively slow, on the order of gigabits per second. For time-critical workloads such as interactive visualization, the bottleneck is no longer computation but communication. In this paper, we present an image compositing algorithm that works on both CPU-only and GPU-accelerated supercomputers and focuses on communication avoidance and overlapping communication with computation at the expense of evenly balancing the workload. The algorithm has three stages: a parallel direct send stage, followed by a tree compositing stage and a gather stage. We compare our algorithm with radix-k and binary-swap from the IceT library in a hybrid OpenMP/MPI setting on the Stampede and Edison supercomputers, show strong scaling results and explain how we generally achieve better performance than these two algorithms. We developed a GPU-based image compositing algorithm where we use CUDA kernels for computation and GPU Direct RDMA for inter-node GPU communication. We tested the algorithm on the Piz Daint GPU-accelerated supercomputer and show that we achieve performance on par with CPUs. Last, we introduce a workflow in which both rendering and compositing are done on the GPU.
AbstractList Modern supercomputers have thousands of nodes, each with CPUs and/or GPUs capable of several teraflops. However, the network connecting these nodes is relatively slow, on the order of gigabits per second. For time-critical workloads such as interactive visualization, the bottleneck is no longer computation but communication. In this paper, we present an image compositing algorithm that works on both CPU-only and GPU-accelerated supercomputers and focuses on communication avoidance and overlapping communication with computation at the expense of evenly balancing the workload. The algorithm has three stages: a parallel direct send stage, followed by a tree compositing stage and a gather stage. We compare our algorithm with radix-k and binary-swap from the IceT library in a hybrid OpenMP/MPI setting on the Stampede and Edison supercomputers, show strong scaling results and explain how we generally achieve better performance than these two algorithms. We developed a GPU-based image compositing algorithm where we use CUDA kernels for computation and GPU Direct RDMA for inter-node GPU communication. We tested the algorithm on the Piz Daint GPU-accelerated supercomputer and show that we achieve performance on par with CPUs. Last, we introduce a workflow in which both rendering and compositing are done on the GPU.
Not provided.
Author Hansen, Charles
Pascal Grosset, A. V.
Knoll, Aaron
Prasad, Manasa
Christensen, Cameron
Author_xml – sequence: 1
  givenname: A. V.
  surname: Pascal Grosset
  fullname: Pascal Grosset, A. V.
  email: pgrosset@sci.utah.edu
  organization: Sci. Comput. & Imaging Inst., Univ. of Utah, Salt Lake City, UT, USA
– sequence: 2
  givenname: Manasa
  surname: Prasad
  fullname: Prasad, Manasa
  email: pbmanasa@gmail.com
  organization: Google, Mountain View, CA, USA
– sequence: 3
  givenname: Cameron
  surname: Christensen
  fullname: Christensen, Cameron
  email: cam@sci.utah.edu
  organization: Sci. Comput. & Imaging Inst., Univ. of Utah, Salt Lake City, UT, USA
– sequence: 4
  givenname: Aaron
  surname: Knoll
  fullname: Knoll, Aaron
  email: knolla@sci.utah.edu
  organization: Sci. Comput. & Imaging Inst., Univ. of Utah, Salt Lake City, UT, USA
– sequence: 5
  givenname: Charles
  surname: Hansen
  fullname: Hansen, Charles
  email: hansen@cs.utah.edu
  organization: Sci. Comput. & Imaging Inst., Univ. of Utah, Salt Lake City, UT, USA
BackLink https://www.ncbi.nlm.nih.gov/pubmed/26992102$$D View this record in MEDLINE/PubMed
https://www.osti.gov/biblio/1536699$$D View this record in Osti.gov
BookMark eNp9kU9v2yAYh9HUqX_SfoBp0oS2yy7OAGNsepvSLo3UKZHq7sowft3R2eCBU6nffkRJd-hhF-DwPK_eH78zdOS8A4TeUTKnlMgv9Y_Fcs4IFXNWcEaEfINOqeQ0IwURR-lNyjJjgokTdBbjIyGU80oeoxMmpGSUsFP0s15fZXUAuMS1jr-z9ROEXo8jtPjKBjATvgPX4h2BV4N-ALzww-ijnax7wJ0P-Oa5CbbF3zcrvNFB9z30Ng5YJ2u5uY_n6G2n-wgXh3uG7r9d14ub7Ha9XC2-3maGUzllugJqDKuYrqQxpDFSaJmTpuRFwTratUWpoW1J2eS50GUuWw7ADGXMtIQDy2fo436uj5NV0dgJzC_jnUsZFC1ykSIn6PMeGoP_s4U4qcFGA32vHfhtVLRiQghSCZLQT6_QR78NLkVQjJY8Z4Snc4Y-HKhtM0CrxmAHHZ7VywcnoNwDJvgYA3QqbaYn690UtO0VJWpXpdpVqXZVqkOVyaSvzJfh_3Pe7x0LAP_4tGfORZX_BXiRpjY
CODEN ITVGEA
CitedBy_id crossref_primary_10_1111_cgf_13702
crossref_primary_10_1016_j_parco_2021_102809
crossref_primary_10_1109_TVCG_2024_3427335
crossref_primary_10_1145_3543861
crossref_primary_10_1002_cpe_6148
crossref_primary_10_1109_ACCESS_2019_2913280
crossref_primary_10_1007_s00371_022_02740_7
Cites_doi 10.1109/TVCG.2007.1026
10.1109/TVCG.2008.104
10.2172/1005031
10.1145/361002.361007
10.1109/LDAV.2014.7013198
10.1007/978-3-642-03770-2_24
10.1109/TVCG.2011.24
10.1109/99.660313
10.1002/cpe.1206
10.1109/38.291528
10.1109/ICPP.2013.17
10.1109/PRS.1993.586080
10.1145/566654.566639
10.1016/B978-012387582-2/50038-1
10.1145/1816038.1816021
10.1109/38.291531
ContentType Journal Article
Copyright Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2017
Copyright_xml – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2017
CorporateAuthor Univ. of Utah, Salt Lake City, UT (United States)
CorporateAuthor_xml – name: Univ. of Utah, Salt Lake City, UT (United States)
DBID 97E
RIA
RIE
AAYXX
CITATION
NPM
7SC
7SP
8FD
JQ2
L7M
L~C
L~D
7X8
OTOTI
DOI 10.1109/TVCG.2016.2542069
DatabaseName IEEE All-Society Periodicals Package (ASPP) 2005–Present
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE Electronic Library (IEL)
CrossRef
PubMed
Computer and Information Systems Abstracts
Electronics & Communications 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
MEDLINE - Academic
OSTI.GOV
DatabaseTitle CrossRef
PubMed
Technology Research Database
Computer and Information Systems Abstracts – Academic
Electronics & Communications Abstracts
ProQuest Computer Science Collection
Computer and Information Systems Abstracts
Advanced Technologies Database with Aerospace
Computer and Information Systems Abstracts Professional
MEDLINE - Academic
DatabaseTitleList
MEDLINE - Academic
PubMed

Technology Research Database
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
– sequence: 3
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
Computer Science
EISSN 1941-0506
EndPage 1690
ExternalDocumentID 1536699
26992102
10_1109_TVCG_2016_2542069
7433468
Genre orig-research
Research Support, U.S. Gov't, Non-P.H.S
Journal Article
GrantInformation_xml – fundername: NNSA
  grantid: DE-NA0002375
  funderid: 10.13039/100006168
– fundername: DOE
  funderid: 10.13039/100000015
– fundername: University of Texas at Austin
  funderid: 10.13039/100008562
– fundername: SciDAC Institute of Scalable Data Management Analysis and Visualization
– fundername: DOE
  grantid: DE-SC0007446; NSF ACI-1339881; NSF IIS-1162013
  funderid: 10.13039/100000015
– fundername: Texas Advanced Computing Center (TACC)
– fundername: National Energy Research Scientific Computing Center (NERSC)
– fundername: PSAAP
– fundername: Swiss National Supercomputing Centre (CSCS)
GroupedDBID ---
-~X
.DC
0R~
29I
4.4
53G
5GY
6IK
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABQJQ
ABVLG
ACGFO
ACIWK
AENEX
AGQYO
AHBIQ
AKJIK
AKQYR
ALMA_UNASSIGNED_HOLDINGS
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CS3
DU5
EBS
EJD
F5P
HZ~
IEDLZ
IFIPE
IPLJI
JAVBF
LAI
M43
O9-
OCL
P2P
PQQKQ
RIA
RIE
RNS
TN5
AAYXX
CITATION
5VS
AAYOK
AETIX
AGSQL
AI.
AIBXA
ALLEH
H~9
IFJZH
NPM
PKN
RIC
RIG
RNI
RZB
VH1
Z5M
7SC
7SP
8FD
JQ2
L7M
L~C
L~D
7X8
OTOTI
PQEST
ID FETCH-LOGICAL-c419t-a8e1cc282a89cc0bc96a930b74552f1fd57aedd07b336a739d4ee2c122cd04e23
IEDL.DBID RIE
ISICitedReferencesCount 10
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000400527500010&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1077-2626
IngestDate Fri May 19 02:16:37 EDT 2023
Sat Sep 27 16:40:34 EDT 2025
Sun Nov 09 08:34:29 EST 2025
Wed Feb 19 02:43:38 EST 2025
Tue Nov 18 20:52:05 EST 2025
Sat Nov 29 06:05:36 EST 2025
Wed Aug 27 02:47:56 EDT 2025
IsDoiOpenAccess false
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 6
Language English
License https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
https://doi.org/10.15223/policy-029
https://doi.org/10.15223/policy-037
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c419t-a8e1cc282a89cc0bc96a930b74552f1fd57aedd07b336a739d4ee2c122cd04e23
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
NA0002375; SC0007446
USDOE National Nuclear Security Administration (NNSA)
ORCID 0000-0003-2192-3843
0000000321923843
OpenAccessLink https://www.osti.gov/biblio/1756081
PMID 26992102
PQID 2174320474
PQPubID 75741
PageCount 14
ParticipantIDs proquest_miscellaneous_1826660860
crossref_citationtrail_10_1109_TVCG_2016_2542069
crossref_primary_10_1109_TVCG_2016_2542069
ieee_primary_7433468
pubmed_primary_26992102
osti_scitechconnect_1536699
proquest_journals_2174320474
PublicationCentury 2000
PublicationDate 2017-06-01
PublicationDateYYYYMMDD 2017-06-01
PublicationDate_xml – month: 06
  year: 2017
  text: 2017-06-01
  day: 01
PublicationDecade 2010
PublicationPlace United States
PublicationPlace_xml – name: United States
– name: New York
PublicationTitle IEEE transactions on visualization and computer graphics
PublicationTitleAbbrev TVCG
PublicationTitleAlternate IEEE Trans Vis Comput Graph
PublicationYear 2017
Publisher IEEE
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher_xml – name: IEEE
– name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
References ref13
ref15
ref36
grosset (ref14) 0
kruger (ref31) 0
marchesin (ref16) 0
james (ref37) 2015
ref1
(ref44) 2015
(ref35) 0
shaw (ref40) 1991
ref18
moreland (ref11) 0
cullip (ref30) 1994
rizzi (ref17) 0
(ref46) 2015
fogal (ref2) 0
wang (ref34) 2011
shalf (ref4) 0
hsu (ref21) 0
yu (ref3) 2010
fogal (ref33) 0
humphreys (ref28) 2002
messmer (ref41) 2016
(ref43) 2015
ref45
ref26
klein (ref39) 2014
ref25
ref20
ref42
ref22
müller (ref32) 0
(ref38) 2014
cavin (ref27) 0
peterka (ref10) 0
ref29
ref8
rabenseifner (ref6) 0
ref9
howison (ref7) 2010
yu (ref24) 0
ref5
(ref47) 2015
(ref12) 2015
childs (ref19) 0
stompel (ref23) 2003
References_xml – start-page: 59
  year: 0
  ident: ref32
  publication-title: Proc 6th Eurographics Conf Parallel Graphics Vis
– start-page: 1
  year: 0
  ident: ref4
  publication-title: Proc 9th Int Conf High Perform Comput Comput Sci
– ident: ref26
  doi: 10.1109/TVCG.2007.1026
– ident: ref29
  doi: 10.1109/TVCG.2008.104
– start-page: 45
  year: 2010
  ident: ref3
  publication-title: IEEE Comput Graph Appl
– ident: ref25
  doi: 10.2172/1005031
– start-page: 1
  year: 2010
  ident: ref7
  publication-title: Proc 10th Eurographics Conf Parallel Graphics Vis
– start-page: 25:1
  year: 0
  ident: ref11
  publication-title: Proc Int Conf High Perform Comput Netw Storage Anal
– year: 2015
  ident: ref44
– start-page: 1
  year: 0
  ident: ref16
  publication-title: Proc 8th Eurographics Conf Parallel Graphics Vis
– year: 2015
  ident: ref46
– start-page: 57
  year: 0
  ident: ref33
  publication-title: Proc High-Performance Graphics
– ident: ref1
  doi: 10.1145/361002.361007
– year: 2015
  ident: ref37
– year: 2015
  ident: ref12
– start-page: 257
  year: 2011
  ident: ref34
  publication-title: Comput Sci
– start-page: 67
  year: 0
  ident: ref14
  publication-title: Proc Eurographics Parallel Graphics and Vis Symp
– year: 2015
  ident: ref43
– ident: ref15
  doi: 10.1109/LDAV.2014.7013198
– ident: ref5
  doi: 10.1007/978-3-642-03770-2_24
– ident: ref8
  doi: 10.1109/TVCG.2011.24
– ident: ref45
  doi: 10.1109/99.660313
– ident: ref42
  doi: 10.1002/cpe.1206
– year: 2015
  ident: ref47
– start-page: 183
  year: 1991
  ident: ref40
  article-title: A VLSI architecture for image composition
  publication-title: Advances in Computer Graphics Hardware III
– ident: ref20
  doi: 10.1109/38.291528
– ident: ref36
  doi: 10.1109/ICPP.2013.17
– ident: ref9
  doi: 10.1109/PRS.1993.586080
– start-page: 38
  year: 0
  ident: ref31
  publication-title: Proc 14th IEEE Vis
– start-page: 129
  year: 0
  ident: ref27
  article-title: Shift-based parallel image compositing on infiniband TM fat-trees
  publication-title: Proc Eurographics Parallel Graphics and Vis Symp
– start-page: 7
  year: 0
  ident: ref21
  publication-title: Proc Symp Parallel Rendering
– start-page: 6
  year: 2003
  ident: ref23
  publication-title: Proc IEEE Symp Par and Large Data Vis and Graphics
– start-page: 190
  year: 0
  ident: ref19
  article-title: A contract-based system for large data visualization
  publication-title: Proc IEEE Vis
– year: 2014
  ident: ref39
  article-title: Unlocking the full potential of the cray XK7 accelerator mark
  publication-title: Proc Cray User Group Conf
– start-page: 48:1
  year: 0
  ident: ref24
  publication-title: Proc ACM/IEEE Conf Supercomput
– start-page: 4:1
  year: 0
  ident: ref10
  publication-title: Proc Conf High Perform Comput Netw Storage Anal
– start-page: 693
  year: 2002
  ident: ref28
  publication-title: ACM Trans Graph
  doi: 10.1145/566654.566639
– year: 1994
  ident: ref30
  article-title: Accelerating volume reconstruction with 3D texture hardware
– start-page: 81
  year: 0
  ident: ref2
  publication-title: Proc 11th Eurographics Conf Parallel Graph Vis
– year: 2014
  ident: ref38
  article-title: Remote visualization on server-class tesla GPUs
– year: 0
  ident: ref35
– ident: ref18
  doi: 10.1016/B978-012387582-2/50038-1
– year: 2016
  ident: ref41
– start-page: 427
  year: 0
  ident: ref6
  publication-title: Proc 2009 17th Euromicro Int Conf Parallel Distrib Netw -Based Process
– ident: ref13
  doi: 10.1145/1816038.1816021
– start-page: 65
  year: 0
  ident: ref17
  article-title: Performance modeling of vl3 volume rendering on GPU-based clusters
  publication-title: Proc Eurographics Parallel Graphics and Vis Symp
– ident: ref22
  doi: 10.1109/38.291531
SSID ssj0014489
Score 2.2966235
Snippet Modern supercomputers have thousands of nodes, each with CPUs and/or GPUs capable of several teraflops. However, the network connecting these nodes is...
Not provided.
SourceID osti
proquest
pubmed
crossref
ieee
SourceType Open Access Repository
Aggregation Database
Index Database
Enrichment Source
Publisher
StartPage 1677
SubjectTerms Algorithms
Communication
Computation
Computer Science
Data visualization
Distributed volume rendering
Graphics processing units
image compositing
Loading
Message systems
Nodes
Parallel processing
Rendering (computer graphics)
Supercomputers
Workflow
Title TOD-Tree: Task-Overlapped Direct Send Tree Image Compositing for Hybrid MPI Parallelism and GPUs
URI https://ieeexplore.ieee.org/document/7433468
https://www.ncbi.nlm.nih.gov/pubmed/26992102
https://www.proquest.com/docview/2174320474
https://www.proquest.com/docview/1826660860
https://www.osti.gov/biblio/1536699
Volume 23
WOSCitedRecordID wos000400527500010&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
journalDatabaseRights – providerCode: PRVIEE
  databaseName: IEEE Electronic Library (IEL)
  customDbUrl:
  eissn: 1941-0506
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0014489
  issn: 1077-2626
  databaseCode: RIE
  dateStart: 19950101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1Lb9QwEB61FQc48CqP0FIZiRPCreM4dswNFfqQoF2JtNqbsR0HVbTZ1Wa3Ev8eT5KNQAIkTokUW4nyjcfzeV4Ar3PMx8xyR52oNRUit7SIhjFN68q6wmoruvSxy0_q7KyYTvVkA96OuTAhhC74LOzjbefLr2Z-hUdlB3G3y4QsNmFTKdnnao0eg0gzdB9fqCiPVvrgwUyZPigvD48xiEvuRzbEmcRKoVxqjWznt-2o668SL7O4uv5ucXY7z9GD__vmh3B_sDDJ-14kHsFGaB7DvV_qDm7D1_L8Ay0XIbwjpW2_0_NbPNWbz0NFehVIvoSmIjiCnN5EjUNQbXThXc03Es1ccvIDM73I58kpmdgF9mO5vmpviI2zjicX7RO4OPpYHp7QodUC9SLVS2qLkHof6ZcttPfMeS2tzphTIs95HYHLlQ1VxZTLMmlVpisRAvcp575iIvDsKWw1syY8B5IzW4taSZcFJ4SLDLLCrn_CcewGyF0CbP3HjR_qkGM7jGvT8RGmDeJlEC8z4JXAm3HKvC_C8a_B24jCOHAAIIEdhNVEwwKr43oMI_JLExW-jKKQwO4abTMs4tZ0bI0zoUQCr8bHcfmhT8U2YbZqDdIzKSMvZAk866VkfPNaxl78-Yt24C5HG6E70tmFreViFV7CHX-7vGoXe1HGp8VeJ-M_AdOC8ks
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1bb9MwFD4aAwl44DYuYQOMxBPCm-M4F_OGBlsruq4S2bQ3YzvONG1Lq6adxL_Hx00jkACJp0SKrUT5jo_P53MDeJdiPmaSGmpELakQqaaFN4xpXFfaFFpqEdLHTkf5eFycncnJBnzoc2GccyH4zO3ibfDlV1O7xKOyPb_bJSIrbsFt7JzVZWv1PgNPNOQqwjCn3NvpnQ8zZnKvPN0_xDCubNfzIc4yrBXKMymR7_y2IYUOK_4y9evr7zZn2HsOHv7fVz-CB52NST6thOIxbLjmCdz_pfLgFnwvjz_Tcu7cR1Lq9pIe3-C53mzmKrJSguSbayqCI8jw2uscgoojBHg158QbumTwA3O9yNFkSCZ6jh1Zri7aa6L9rMPJSfsUTg6-lPsD2jVboFbEckF14WJrPQHThbSWGSszLRNmcpGmvPbQpbl2VcVykySZzhNZCee4jTm3FROOJ89gs5k27gWQlOla1HlmEmeEMJ5DVtj3TxiO_QC5iYCt_7iyXSVybIhxpQIjYVIhXgrxUh1eEbzvp8xWZTj-NXgLUegHdgBEsI2wKm9aYH1ci4FEdqG8ys-8KESws0Zbdcu4VYGvcSZyEcHb_rFfgOhV0Y2bLluFBC3LPDNkETxfSUn_5rWMvfzzF72Bu4PyaKRGw_HXbbjH0WIIBzw7sLmYL90ruGNvFhft_HWQ9J_kZ_Ss
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=TOD-Tree%3A+Task-Overlapped+Direct+Send+Tree+Image+Compositing+for+Hybrid+MPI+Parallelism+and+GPUs&rft.jtitle=IEEE+transactions+on+visualization+and+computer+graphics&rft.au=Grosset%2C+A.+V.+Pascal&rft.au=Prasad%2C+Manasa&rft.au=Christensen%2C+Cameron&rft.au=Knoll%2C+Aaron&rft.date=2017-06-01&rft.issn=1077-2626&rft.volume=23&rft.issue=6&rft.spage=1677&rft.epage=1690&rft_id=info:doi/10.1109%2FTVCG.2016.2542069&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TVCG_2016_2542069
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1077-2626&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1077-2626&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1077-2626&client=summon