Scheduling moldable tasks on homogeneous multi-cluster platforms with GPUs

This paper examines task scheduling in homogeneous multi-cluster platforms, equipped with Graphics Processing Units (GPUs), with the aim of minimizing the makespan. We assume that tasks can be parallelized across these platforms under the moldable model. Recognizing the NP-hard nature of the problem...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Computers & operations research Ročník 179; s. 107041
Hlavní autoři: Wu, Fangfang, Zhang, Run, Zhang, Xiandong
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier Ltd 01.07.2025
Témata:
ISSN:0305-0548
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract This paper examines task scheduling in homogeneous multi-cluster platforms, equipped with Graphics Processing Units (GPUs), with the aim of minimizing the makespan. We assume that tasks can be parallelized across these platforms under the moldable model. Recognizing the NP-hard nature of the problem, our goal is to develop algorithms that provide approximation ratios. While existing research has established algorithms for single-cluster GPU environments, scaling these to multi-cluster platforms introduces new challenges, especially due to the restriction that tasks cannot use processors from different clusters. We propose an integer programming-based algorithm that achieves an approximation ratio of 32+ϵ, trading off runtime for an improved approximation ratio. Additionally, leveraging recent theoretical advancements, we have created a polynomial-time algorithm with an approximation ratio of 2+ϵ. Empirical computational experiments show that our algorithms surpass their counterparts in empirical approximation ratios.
AbstractList This paper examines task scheduling in homogeneous multi-cluster platforms, equipped with Graphics Processing Units (GPUs), with the aim of minimizing the makespan. We assume that tasks can be parallelized across these platforms under the moldable model. Recognizing the NP-hard nature of the problem, our goal is to develop algorithms that provide approximation ratios. While existing research has established algorithms for single-cluster GPU environments, scaling these to multi-cluster platforms introduces new challenges, especially due to the restriction that tasks cannot use processors from different clusters. We propose an integer programming-based algorithm that achieves an approximation ratio of 32+ϵ, trading off runtime for an improved approximation ratio. Additionally, leveraging recent theoretical advancements, we have created a polynomial-time algorithm with an approximation ratio of 2+ϵ. Empirical computational experiments show that our algorithms surpass their counterparts in empirical approximation ratios.
ArticleNumber 107041
Author Zhang, Xiandong
Zhang, Run
Wu, Fangfang
Author_xml – sequence: 1
  givenname: Fangfang
  surname: Wu
  fullname: Wu, Fangfang
  email: ffwu@shmtu.edu.cn
  organization: School of Economics and Management, Shanghai Maritime University, Shanghai, 201306, China
– sequence: 2
  givenname: Run
  surname: Zhang
  fullname: Zhang, Run
  email: zhangr20@fudan.edu.cn
  organization: School of Management, Fudan University, Shanghai, 200433, China
– sequence: 3
  givenname: Xiandong
  orcidid: 0000-0003-1442-3859
  surname: Zhang
  fullname: Zhang, Xiandong
  email: xiandongzhang@fudan.edu.cn
  organization: School of Management, Fudan University, Shanghai, 200433, China
BookMark eNp9kE1LAzEQhnOoYFv9Ad7yB7ZOdrMfxZMUrUpBQXsO-WxTs5uSZBX_vSn17BxmeAeeYXhmaDL4QSN0Q2BBgDS3h4X0YVFCWefcAiUTNIUK6gJq2l2iWYwHyNWWZIpe3uVeq9HZYYd77xQXTuPE42fEfsB73_udHrQfI-5Hl2wh3RiTDvjoeDI-9BF_27TH67dtvEIXhruor__mHG0fHz5WT8Xmdf28ut8UsgSachdqKerGGAFUt9xQkM2ygdLISnZGUa6ogLzuBAHBQSlBBe942VEliWqrOSLnuzL4GIM27Bhsz8MPI8BOAtiBZQHsJICdBWTm7szo_NiX1YFFafUgtbJBy8SUt__Qv9pIaYY
Cites_doi 10.1145/378580.378640
10.1142/S1793830911001413
10.1145/2788396
10.1016/j.endm.2016.10.003
10.1109/71.993206
10.1137/0402042
10.1016/j.ijpe.2018.04.013
10.1016/j.ejor.2017.06.017
10.1016/j.ejor.2023.02.044
10.1145/1815961.1816021
10.1007/s00453-001-0085-8
10.1007/s10951-006-8497-6
10.1287/opre.2022.0168
10.1137/080736491
10.1109/TC.2013.115
10.1016/j.tcs.2015.07.003
10.1007/s10951-022-00733-x
10.1109/TPDS.2017.2675891
10.1142/S0129054105002954
10.1145/140901.141909
10.1145/3387110
10.1016/j.ejor.2022.08.034
10.1145/7531.7535
10.1057/jors.2010.54
10.1080/24725854.2019.1680907
10.1109/TPDS.2009.11
10.1002/cpe.1378
10.1016/j.ijpe.2015.05.038
10.1137/S0097539701385995
ContentType Journal Article
Copyright 2025 Elsevier Ltd
Copyright_xml – notice: 2025 Elsevier Ltd
DBID AAYXX
CITATION
DOI 10.1016/j.cor.2025.107041
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
Computer Science
Business
ExternalDocumentID 10_1016_j_cor_2025_107041
S0305054825000693
GroupedDBID --K
--M
-~X
.DC
.~1
0R~
186
1B1
1OL
1RT
1~.
1~5
29F
4.4
457
4G.
5GY
5VS
6J9
7-5
71M
8P~
9JN
9JO
AAAKF
AAAKG
AABNK
AACTN
AAEDT
AAEDW
AAFJI
AAIKJ
AAKOC
AALRI
AAOAW
AAQXK
AARIN
AATTM
AAXKI
AAXUO
AAYFN
AAYOK
ABAOU
ABBOA
ABDPE
ABEFU
ABFNM
ABFRF
ABJNI
ABMAC
ABMMH
ABUCO
ABWVN
ABXDB
ACDAQ
ACGFO
ACGFS
ACNCT
ACNNM
ACRLP
ACRPL
ACZNC
ADBBV
ADEZE
ADGUI
ADJOM
ADMUD
ADNMO
AEBSH
AEFWE
AEHXG
AEIPS
AEKER
AENEX
AFFNX
AFJKZ
AFTJW
AFXIZ
AGCQF
AGHFR
AGQPQ
AGUBO
AGYEJ
AHHHB
AHZHX
AI.
AIALX
AIEXJ
AIGVJ
AIKHN
AITUG
AKRWK
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
ANKPU
AOMHK
AOUOD
APLSM
APXCP
ARUGR
ASPBG
AVARZ
AVWKF
AXJTR
AZFZN
BKOJK
BKOMP
BLXMC
BNPGV
CS3
DU5
EBS
EFJIC
EJD
EO8
EO9
EP2
EP3
FDB
FEDTE
FGOYB
FIRID
FNPLU
FYGXN
G-Q
GBLVA
GBOLZ
HAMUX
HVGLF
HZ~
H~9
IHE
J1W
KOM
LY1
M41
MHUIS
MO0
MS~
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
PQQKQ
PRBVW
Q38
R2-
RIG
ROL
RPZ
RXW
SDF
SDG
SDP
SDS
SES
SEW
SPC
SPCBC
SSB
SSD
SSH
SSO
SSV
SSW
SSZ
T5K
TAE
TN5
U5U
UPT
VH1
WUQ
XPP
ZMT
~02
~G-
9DU
AAYWO
AAYXX
ACLOT
ACVFH
ADCNI
AEUPX
AFPUW
AIGII
AIIUN
AKBMS
AKYEP
CITATION
EFKBS
EFLBG
~HD
ID FETCH-LOGICAL-c204t-c2bd9b56ffb04e7af40c69602fc3c8fd4ad4b0af48b10ba0ddb4ba8a284dc1d73
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001446714600001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0305-0548
IngestDate Sat Nov 29 07:58:53 EST 2025
Sat May 03 15:40:47 EDT 2025
IsPeerReviewed true
IsScholarly true
Keywords Dual approximation algorithm
Multi-cluster platform
Moldable task scheduling
Approximation algorithm
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c204t-c2bd9b56ffb04e7af40c69602fc3c8fd4ad4b0af48b10ba0ddb4ba8a284dc1d73
ORCID 0000-0003-1442-3859
ParticipantIDs crossref_primary_10_1016_j_cor_2025_107041
elsevier_sciencedirect_doi_10_1016_j_cor_2025_107041
PublicationCentury 2000
PublicationDate 2025-07-01
PublicationDateYYYYMMDD 2025-07-01
PublicationDate_xml – month: 07
  year: 2025
  text: 2025-07-01
  day: 01
PublicationDecade 2020
PublicationTitle Computers & operations research
PublicationYear 2025
Publisher Elsevier Ltd
Publisher_xml – name: Elsevier Ltd
References Wu, Zhang, Chen (b54) 2023; 306
Ye, Han, Zhang (b55) 2009
Dolgui, Kovalev, Kovalyov, Malyutin, Soukhal (b13) 2018; 264
Feitelson, Rudolph, Schwiegelshohn, Sevcik, Wong (b23) 1997
Jansen, Porkolab (b33) 2002; 32
Amert, Otterness, Yang, Anderson, Smith (b1) 2017
Dutot, P.-F (b20) 2005
Bleuse, Hunold, Kedad-Sidhoum, Monna, Mounié, Trystram (b6) 2017; 28
Pascual, Rzadca, Trystram (b45) 2009; 21
Johannes (b38) 2006; 9
Blazewicz, Cheng, Machowiak, Oguz (b5) 2011; 62
Germann, Jansen, Ohnesorge, Tutas (b28) 2024
Bougeret, Dutot, Jansen, Otte, Trystram (b8) 2010
Battaïa, Delorme, Dolgui, Hagemann, Horlemann, Kovalev, Malyutin (b2) 2015; 170
Pricopi, Mitra (b46) 2013; 63
Tchernykh, Ramírez, Avetisyan, Kuzjurin, Grushin, Zhuk (b48) 2005
Du, Leung (b15) 1989; 2
Fotakis, Matuschke, Papadigenopoulos (b25) 2024
Diabat, Dolgui, Janiak, Kovalyov (b12) 2020; 52
Schwiegelshohn, Tchernykh, Yahyapour (b47) 2008
Dutot, Jansen, Robenek, Trystram (b17) 2013
Fotakis, Matuschke, Papadigenopoulos (b24) 2023; 26
Jansen, Land (b32) 2018
Fujiwara, Tanaka, Taura, Torisawa (b27) 2018
Dutot, Mounié, Trystram (b18) 2004
Mounié, Rapine, Trystram (b44) 2007; 37
Turek, J., Wolf, J.L., Yu, P.S., 1992. Approximate algorithms for scheduling parallelizable tasks. In: Proceedings of the Fourth Annual ACM Symposium on Parallel Algorithms and Architectures. pp. 323–332.
Fotakis, Matuschke, Papadigenopoulos (b26) 2024
Lee, V.W., Kim, C., Chhugani, J., Deisher, M., Kim, D., Nguyen, A.D., Satish, N., Smelyanskiy, M., Chennupaty, S., Hammarlund, P., et al., 2010. Debunking the 100x gpu vs. cpu myth: an evaluation of throughput computing on cpu and gpu. In: Proceedings of the 37th Annual International Symposium on Computer Architecture. pp. 451–460.
Jansen (b31) 2012
Wu, Wang (b53) 2018; 201
Beaumont, Bonichon, Courtès, Dolstra, Hanin (b3) 2012
Hochbaum, Shmoys (b29) 1987; 34
Wu, Jiang, Zhang, Zhang (b51) 2023
Jansen, Trystram (b37) 2016; 55
Drozdowski (b14) 2009
Dutot, Eyraud, Mounié, Trystram (b16) 2005; 16
Jansen, Rau (b35) 2019
Bougeret, Dutot, Jansen, Robenek, Trystram (b10) 2012
Topcuoglu, Hariri, Wu (b49) 2002; 13
Feitelson (b22) 2005
Dutot, N’takpé, Suter, Casanova (b19) 2009; 20
Jansen (b30) 2004; PP
Ludwig, Tiwari (b41) 1994
Mittal, Vetter (b42) 2015; 47
Wu, Loiseau (b52) 2023; 310
Jansen, Thöle (b36) 2010; 39
Bougeret, Dutot, Jansen, Robenek, Trystram (b9) 2011; 3
Lee, Samadi, Park, Mahlke (b40) 2013
Bougeret, Dutot, Trystram, Jansen, Robenek (b11) 2015; 600
Dutot, P.-F., Trystram, D., 2001. Scheduling on hierarchical clusters using malleable tasks. In: Proceedings of the Thirteenth Annual ACM Symposium on Parallel Algorithms and Architectures. pp. 199–208.
Beaumont, Canon, Eyraud-Dubois, Lucarelli, Marchal, Mommessin, Simon, Trystram (b4) 2020; 53
Mounié, Rapine, Trystram (b43) 1999
Jansen, Rau (b34) 2019
Bougeret, Dutot, Jansen, Otte, Trystram (b7) 2010
Drozdowski (10.1016/j.cor.2025.107041_b14) 2009
Fujiwara (10.1016/j.cor.2025.107041_b27) 2018
Jansen (10.1016/j.cor.2025.107041_b36) 2010; 39
Pascual (10.1016/j.cor.2025.107041_b45) 2009; 21
Fotakis (10.1016/j.cor.2025.107041_b24) 2023; 26
Blazewicz (10.1016/j.cor.2025.107041_b5) 2011; 62
Mounié (10.1016/j.cor.2025.107041_b43) 1999
10.1016/j.cor.2025.107041_b39
Pricopi (10.1016/j.cor.2025.107041_b46) 2013; 63
Mittal (10.1016/j.cor.2025.107041_b42) 2015; 47
Dutot (10.1016/j.cor.2025.107041_b16) 2005; 16
10.1016/j.cor.2025.107041_b50
Dolgui (10.1016/j.cor.2025.107041_b13) 2018; 264
Bougeret (10.1016/j.cor.2025.107041_b9) 2011; 3
Lee (10.1016/j.cor.2025.107041_b40) 2013
Bougeret (10.1016/j.cor.2025.107041_b7) 2010
Dutot (10.1016/j.cor.2025.107041_b18) 2004
Dutot (10.1016/j.cor.2025.107041_b17) 2013
Hochbaum (10.1016/j.cor.2025.107041_b29) 1987; 34
Beaumont (10.1016/j.cor.2025.107041_b3) 2012
Johannes (10.1016/j.cor.2025.107041_b38) 2006; 9
Dutot (10.1016/j.cor.2025.107041_b20) 2005
Battaïa (10.1016/j.cor.2025.107041_b2) 2015; 170
Feitelson (10.1016/j.cor.2025.107041_b22) 2005
Jansen (10.1016/j.cor.2025.107041_b34) 2019
Fotakis (10.1016/j.cor.2025.107041_b25) 2024
Du (10.1016/j.cor.2025.107041_b15) 1989; 2
Jansen (10.1016/j.cor.2025.107041_b33) 2002; 32
Wu (10.1016/j.cor.2025.107041_b51) 2023
Fotakis (10.1016/j.cor.2025.107041_b26) 2024
Ye (10.1016/j.cor.2025.107041_b55) 2009
Amert (10.1016/j.cor.2025.107041_b1) 2017
10.1016/j.cor.2025.107041_b21
Jansen (10.1016/j.cor.2025.107041_b31) 2012
Jansen (10.1016/j.cor.2025.107041_b37) 2016; 55
Beaumont (10.1016/j.cor.2025.107041_b4) 2020; 53
Germann (10.1016/j.cor.2025.107041_b28) 2024
Mounié (10.1016/j.cor.2025.107041_b44) 2007; 37
Diabat (10.1016/j.cor.2025.107041_b12) 2020; 52
Jansen (10.1016/j.cor.2025.107041_b30) 2004; PP
Ludwig (10.1016/j.cor.2025.107041_b41) 1994
Bougeret (10.1016/j.cor.2025.107041_b11) 2015; 600
Wu (10.1016/j.cor.2025.107041_b54) 2023; 306
Schwiegelshohn (10.1016/j.cor.2025.107041_b47) 2008
Wu (10.1016/j.cor.2025.107041_b52) 2023; 310
Bleuse (10.1016/j.cor.2025.107041_b6) 2017; 28
Jansen (10.1016/j.cor.2025.107041_b32) 2018
Wu (10.1016/j.cor.2025.107041_b53) 2018; 201
Tchernykh (10.1016/j.cor.2025.107041_b48) 2005
Dutot (10.1016/j.cor.2025.107041_b19) 2009; 20
Jansen (10.1016/j.cor.2025.107041_b35) 2019
Bougeret (10.1016/j.cor.2025.107041_b10) 2012
Topcuoglu (10.1016/j.cor.2025.107041_b49) 2002; 13
Bougeret (10.1016/j.cor.2025.107041_b8) 2010
Feitelson (10.1016/j.cor.2025.107041_b23) 1997
References_xml – volume: 28
  start-page: 2689
  year: 2017
  end-page: 2702
  ident: b6
  article-title: Scheduling independent moldable tasks on multi-cores with gpus
  publication-title: IEEE Trans. Parallel Distrib. Syst.
– volume: 306
  start-page: 567
  year: 2023
  end-page: 578
  ident: b54
  article-title: An improved approximation algorithm for scheduling monotonic moldable tasks
  publication-title: European J. Oper. Res.
– volume: 16
  start-page: 217
  year: 2005
  end-page: 237
  ident: b16
  article-title: Scheduling on large scale distributed platforms: from models to implementations
  publication-title: Internat. J. Found. Comput. Sci.
– start-page: 103
  year: 2019
  end-page: 116
  ident: b35
  article-title: Linear time algorithms for multiple cluster scheduling and multiple strip packing
  publication-title: European Conference on Parallel Processing
– start-page: 878
  year: 2012
  end-page: 885
  ident: b10
  article-title: Tight approximation for scheduling parallel jobs on identical clusters
  publication-title: 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops & PhD Forum
– start-page: 91
  year: 2012
  end-page: 98
  ident: b3
  article-title: Mixed data-parallel scheduling for distributed continuous integration
  publication-title: 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops & PhD Forum
– start-page: 1
  year: 2024
  end-page: 25
  ident: b26
  article-title: A constant-factor approximation for generalized malleable scheduling under
  publication-title: Math. Program.
– volume: 34
  start-page: 144
  year: 1987
  end-page: 162
  ident: b29
  article-title: Using dual approximation algorithms for scheduling problems theoretical and practical results
  publication-title: J. ACM
– start-page: 302
  year: 2005
  end-page: 311
  ident: b20
  article-title: Hierarchical scheduling for moldable tasks
  publication-title: Euro-Par 2005 Parallel Processing: 11th International Euro-Par Conference, Lisbon, Portugal, August 30-September 2 2005. Proceedings 11
– year: 2024
  ident: b28
  article-title: 3/2-dual approximation for cpu/gpu scheduling
  publication-title: 22nd International Symposium on Experimental Algorithms
– volume: 170
  start-page: 489
  year: 2015
  end-page: 500
  ident: b2
  article-title: Workforce minimization for a mixed-model assembly line in the automotive industry
  publication-title: Int. J. Prod. Econ.
– reference: Dutot, P.-F., Trystram, D., 2001. Scheduling on hierarchical clusters using malleable tasks. In: Proceedings of the Thirteenth Annual ACM Symposium on Parallel Algorithms and Architectures. pp. 199–208.
– start-page: 1
  year: 1997
  end-page: 34
  ident: b23
  article-title: Theory and practice in parallel job scheduling
  publication-title: Job Scheduling Strategies for Parallel Processing: IPPS’97 Processing Workshop Geneva, Switzerland, April 5 1997 Proceedings 3
– year: 2019
  ident: b34
  article-title: Closing the gap for pseudo-polynomial strip packing
  publication-title: 27th Annual European Symposium on Algorithms
– volume: 55
  start-page: 9
  year: 2016
  end-page: 12
  ident: b37
  article-title: Scheduling parallel jobs on heterogeneous platforms
  publication-title: Electron. Notes Discrete Math.
– start-page: 157
  year: 2010
  end-page: 167
  ident: b8
  article-title: A fast 5/2-approximation algorithm for hierarchical scheduling
  publication-title: European Conference on Parallel Processing
– volume: 32
  start-page: 507
  year: 2002
  end-page: 520
  ident: b33
  article-title: Linear-time approximation schemes for scheduling malleable parallel tasks
  publication-title: Algorithmica
– volume: 21
  start-page: 905
  year: 2009
  end-page: 921
  ident: b45
  article-title: Cooperation in multi-organization scheduling
  publication-title: Concurr. Comput.: Pr. Exp.
– start-page: 245
  year: 2013
  end-page: 255
  ident: b40
  article-title: Transparent cpu-gpu collaboration for data-parallel kernels on heterogeneous systems
  publication-title: Proceedings of the 22nd International Conference on Parallel Architectures and Compilation Techniques
– start-page: 78
  year: 2013
  end-page: 89
  ident: b17
  article-title: A (2+
  publication-title: European Conference on Parallel Processing
– start-page: 389
  year: 2018
  end-page: 398
  ident: b27
  article-title: Effectiveness of moldable and malleable scheduling in deep learning tasks
  publication-title: 2018 Ieee 24th International Conference on Parallel and Distributed Systems
– volume: 3
  start-page: 553
  year: 2011
  end-page: 586
  ident: b9
  article-title: Approximation algorithms for multiple strip packing and scheduling parallel jobs in platforms
  publication-title: Discrete Math. Algorithms Appl.
– start-page: 172
  year: 2018
  end-page: 181
  ident: b32
  article-title: Scheduling monotone moldable jobs in linear time
  publication-title: 2018 IEEE International Parallel and Distributed Processing Symposium
– volume: 47
  year: 2015
  ident: b42
  article-title: A survey of cpu-gpu heterogeneous computing techniques
  publication-title: ACM Comput. Surv.
– start-page: 316
  year: 2010
  end-page: 327
  ident: b7
  article-title: Approximating the non-contiguous multiple organization packing problem
  publication-title: IFIP International Conference on Theoretical Computer Science
– start-page: 1
  year: 2023
  end-page: 16
  ident: b51
  article-title: Approximation algorithms for scheduling monotonic moldable tasks on multiple platforms
  publication-title: J. Sched.
– start-page: 5
  year: 2009
  end-page: 165
  ident: b55
  article-title: On-line multiple-strip packing
  publication-title: International Conference on Combinatorial Optimization and Applications, Vol. 15
– volume: 52
  start-page: 478
  year: 2020
  end-page: 485
  ident: b12
  article-title: Three parallel task assignment problems with shared resources
  publication-title: IISE Trans.
– volume: 2
  start-page: 473
  year: 1989
  end-page: 487
  ident: b15
  article-title: Complexity of scheduling parallel task systems
  publication-title: SIAM J. Discrete Math.
– volume: 9
  start-page: 433
  year: 2006
  end-page: 452
  ident: b38
  article-title: Scheduling parallel jobs to minimize the makespan
  publication-title: J. Sched.
– start-page: 104
  year: 2017
  end-page: 115
  ident: b1
  article-title: Gpu scheduling on the nvidia tx2: Hidden details revealed
  publication-title: 2017 IEEE Real-Time Systems Symposium
– volume: 53
  start-page: 1
  year: 2020
  end-page: 36
  ident: b4
  article-title: Scheduling on two types of resources: a survey
  publication-title: ACM Comput. Surv.
– volume: 20
  start-page: 940
  year: 2009
  end-page: 952
  ident: b19
  article-title: Scheduling parallel task graphs on (almost) homogeneous multicluster platforms
  publication-title: IEEE Trans. Parallel Distrib. Syst.
– reference: Lee, V.W., Kim, C., Chhugani, J., Deisher, M., Kim, D., Nguyen, A.D., Satish, N., Smelyanskiy, M., Chennupaty, S., Hammarlund, P., et al., 2010. Debunking the 100x gpu vs. cpu myth: an evaluation of throughput computing on cpu and gpu. In: Proceedings of the 37th Annual International Symposium on Computer Architecture. pp. 451–460.
– volume: 37
  start-page: 401
  year: 2007
  end-page: 412
  ident: b44
  article-title: A
  publication-title: SIAM J. Comput.
– start-page: 774
  year: 2005
  end-page: 781
  ident: b48
  article-title: Two level job-scheduling strategies for a computational grid
  publication-title: International Conference on Parallel Processing and Applied Mathematics
– volume: 600
  start-page: 70
  year: 2015
  end-page: 85
  ident: b11
  article-title: Improved approximation algorithms for scheduling parallel jobs on identical clusters
  publication-title: Theoret. Comput. Sci.
– volume: 26
  start-page: 425
  year: 2023
  end-page: 442
  ident: b24
  article-title: Malleable scheduling beyond identical machines
  publication-title: J. Sched.
– reference: Turek, J., Wolf, J.L., Yu, P.S., 1992. Approximate algorithms for scheduling parallelizable tasks. In: Proceedings of the Fourth Annual ACM Symposium on Parallel Algorithms and Architectures. pp. 323–332.
– start-page: 167
  year: 1994
  end-page: 176
  ident: b41
  article-title: Scheduling malleable and nonmalleable parallel tasks
  publication-title: Proceedings of the Fifth Annual ACM-SIAM Symposium on Discrete Algorithms
– year: 2024
  ident: b25
  article-title: Assigning and scheduling generalized malleable jobs under subadditive or submodular processing speeds
  publication-title: Oper. Res.
– volume: 63
  start-page: 2590
  year: 2013
  end-page: 2603
  ident: b46
  article-title: Task scheduling on adaptive multi-core
  publication-title: IEEE Trans. Comput.
– start-page: 1
  year: 2008
  end-page: 10
  ident: b47
  article-title: Online scheduling in grids
  publication-title: IEEE International Symposium on Parallel and Distributed Processing
– year: 2009
  ident: b14
  article-title: Scheduling for Parallel Processing, Vol. 18
– volume: 39
  start-page: 3571
  year: 2010
  end-page: 3615
  ident: b36
  article-title: Approximation algorithms for scheduling parallel jobs
  publication-title: SIAM J. Comput.
– volume: 201
  start-page: 26
  year: 2018
  end-page: 40
  ident: b53
  article-title: Exact and heuristic methods to solve the parallel machine scheduling problem with multi-processor tasks
  publication-title: Int. J. Prod. Econ.
– volume: PP
  start-page: 187
  year: 2004
  end-page: 200
  ident: b30
  article-title: Scheduling malleable parallel tasks: An asymptotic fully polynomial time approximation scheme
  publication-title: Algorithmica
– start-page: 23
  year: 1999
  end-page: 32
  ident: b43
  article-title: Efficient approximation algorithms for scheduling malleable tasks
  publication-title: Proceedings of the Eleventh Annual ACM Symposium on Parallel Algorithms and Architectures
– year: 2004
  ident: b18
  article-title: Scheduling parallel tasks: Approximation algorithms
  publication-title: Handbook of Scheduling: Algorithms, Models, and Performance Analysis
– start-page: 224
  year: 2012
  end-page: 235
  ident: b31
  article-title: A (3/2+
  publication-title: Proceedings of the Twenty-Fourth Annual ACM Symposium on Parallelism in Algorithms and Architectures
– volume: 13
  start-page: 260
  year: 2002
  end-page: 274
  ident: b49
  article-title: Performance-effective and low-complexity task scheduling for heterogeneous computing
  publication-title: IEEE Trans. Parallel Distrib. Syst.
– volume: 264
  start-page: 200
  year: 2018
  end-page: 211
  ident: b13
  article-title: Optimal workforce assignment to operations of a paced assembly line
  publication-title: European J. Oper. Res.
– volume: 62
  start-page: 1189
  year: 2011
  end-page: 1197
  ident: b5
  article-title: Berth and quay crane allocation: a moldable task scheduling model
  publication-title: J. Oper. Res. Soc.
– volume: 310
  start-page: 71
  year: 2023
  end-page: 83
  ident: b52
  article-title: Efficient approximation algorithms for scheduling moldable tasks
  publication-title: European J. Oper. Res.
– year: 2005
  ident: b22
  article-title: Parallel workloads archive
– ident: 10.1016/j.cor.2025.107041_b21
  doi: 10.1145/378580.378640
– start-page: 5
  year: 2009
  ident: 10.1016/j.cor.2025.107041_b55
  article-title: On-line multiple-strip packing
– start-page: 23
  year: 1999
  ident: 10.1016/j.cor.2025.107041_b43
  article-title: Efficient approximation algorithms for scheduling malleable tasks
– volume: 3
  start-page: 553
  issue: 04
  year: 2011
  ident: 10.1016/j.cor.2025.107041_b9
  article-title: Approximation algorithms for multiple strip packing and scheduling parallel jobs in platforms
  publication-title: Discrete Math. Algorithms Appl.
  doi: 10.1142/S1793830911001413
– start-page: 167
  year: 1994
  ident: 10.1016/j.cor.2025.107041_b41
  article-title: Scheduling malleable and nonmalleable parallel tasks
– volume: 47
  issue: 4
  year: 2015
  ident: 10.1016/j.cor.2025.107041_b42
  article-title: A survey of cpu-gpu heterogeneous computing techniques
  publication-title: ACM Comput. Surv.
  doi: 10.1145/2788396
– start-page: 157
  year: 2010
  ident: 10.1016/j.cor.2025.107041_b8
  article-title: A fast 5/2-approximation algorithm for hierarchical scheduling
– start-page: 245
  year: 2013
  ident: 10.1016/j.cor.2025.107041_b40
  article-title: Transparent cpu-gpu collaboration for data-parallel kernels on heterogeneous systems
– volume: 55
  start-page: 9
  year: 2016
  ident: 10.1016/j.cor.2025.107041_b37
  article-title: Scheduling parallel jobs on heterogeneous platforms
  publication-title: Electron. Notes Discrete Math.
  doi: 10.1016/j.endm.2016.10.003
– start-page: 224
  year: 2012
  ident: 10.1016/j.cor.2025.107041_b31
  article-title: A (3/2+ϵ) approximation algorithm for scheduling moldable and non-moldable parallel tasks
– start-page: 1
  year: 1997
  ident: 10.1016/j.cor.2025.107041_b23
  article-title: Theory and practice in parallel job scheduling
– volume: 13
  start-page: 260
  issue: 3
  year: 2002
  ident: 10.1016/j.cor.2025.107041_b49
  article-title: Performance-effective and low-complexity task scheduling for heterogeneous computing
  publication-title: IEEE Trans. Parallel Distrib. Syst.
  doi: 10.1109/71.993206
– start-page: 316
  year: 2010
  ident: 10.1016/j.cor.2025.107041_b7
  article-title: Approximating the non-contiguous multiple organization packing problem
– year: 2005
  ident: 10.1016/j.cor.2025.107041_b22
– start-page: 1
  year: 2008
  ident: 10.1016/j.cor.2025.107041_b47
  article-title: Online scheduling in grids
– volume: 2
  start-page: 473
  issue: 4
  year: 1989
  ident: 10.1016/j.cor.2025.107041_b15
  article-title: Complexity of scheduling parallel task systems
  publication-title: SIAM J. Discrete Math.
  doi: 10.1137/0402042
– volume: 201
  start-page: 26
  year: 2018
  ident: 10.1016/j.cor.2025.107041_b53
  article-title: Exact and heuristic methods to solve the parallel machine scheduling problem with multi-processor tasks
  publication-title: Int. J. Prod. Econ.
  doi: 10.1016/j.ijpe.2018.04.013
– volume: 264
  start-page: 200
  issue: 1
  year: 2018
  ident: 10.1016/j.cor.2025.107041_b13
  article-title: Optimal workforce assignment to operations of a paced assembly line
  publication-title: European J. Oper. Res.
  doi: 10.1016/j.ejor.2017.06.017
– year: 2004
  ident: 10.1016/j.cor.2025.107041_b18
  article-title: Scheduling parallel tasks: Approximation algorithms
– volume: 310
  start-page: 71
  issue: 1
  year: 2023
  ident: 10.1016/j.cor.2025.107041_b52
  article-title: Efficient approximation algorithms for scheduling moldable tasks
  publication-title: European J. Oper. Res.
  doi: 10.1016/j.ejor.2023.02.044
– volume: PP
  start-page: 187
  issue: 39
  year: 2004
  ident: 10.1016/j.cor.2025.107041_b30
  article-title: Scheduling malleable parallel tasks: An asymptotic fully polynomial time approximation scheme
  publication-title: Algorithmica
– ident: 10.1016/j.cor.2025.107041_b39
  doi: 10.1145/1815961.1816021
– volume: 32
  start-page: 507
  issue: 3
  year: 2002
  ident: 10.1016/j.cor.2025.107041_b33
  article-title: Linear-time approximation schemes for scheduling malleable parallel tasks
  publication-title: Algorithmica
  doi: 10.1007/s00453-001-0085-8
– year: 2019
  ident: 10.1016/j.cor.2025.107041_b34
  article-title: Closing the gap for pseudo-polynomial strip packing
– volume: 9
  start-page: 433
  issue: 5
  year: 2006
  ident: 10.1016/j.cor.2025.107041_b38
  article-title: Scheduling parallel jobs to minimize the makespan
  publication-title: J. Sched.
  doi: 10.1007/s10951-006-8497-6
– year: 2024
  ident: 10.1016/j.cor.2025.107041_b25
  article-title: Assigning and scheduling generalized malleable jobs under subadditive or submodular processing speeds
  publication-title: Oper. Res.
  doi: 10.1287/opre.2022.0168
– volume: 39
  start-page: 3571
  issue: 8
  year: 2010
  ident: 10.1016/j.cor.2025.107041_b36
  article-title: Approximation algorithms for scheduling parallel jobs
  publication-title: SIAM J. Comput.
  doi: 10.1137/080736491
– volume: 63
  start-page: 2590
  issue: 10
  year: 2013
  ident: 10.1016/j.cor.2025.107041_b46
  article-title: Task scheduling on adaptive multi-core
  publication-title: IEEE Trans. Comput.
  doi: 10.1109/TC.2013.115
– volume: 600
  start-page: 70
  year: 2015
  ident: 10.1016/j.cor.2025.107041_b11
  article-title: Improved approximation algorithms for scheduling parallel jobs on identical clusters
  publication-title: Theoret. Comput. Sci.
  doi: 10.1016/j.tcs.2015.07.003
– start-page: 104
  year: 2017
  ident: 10.1016/j.cor.2025.107041_b1
  article-title: Gpu scheduling on the nvidia tx2: Hidden details revealed
– start-page: 91
  year: 2012
  ident: 10.1016/j.cor.2025.107041_b3
  article-title: Mixed data-parallel scheduling for distributed continuous integration
– volume: 26
  start-page: 425
  issue: 5
  year: 2023
  ident: 10.1016/j.cor.2025.107041_b24
  article-title: Malleable scheduling beyond identical machines
  publication-title: J. Sched.
  doi: 10.1007/s10951-022-00733-x
– volume: 28
  start-page: 2689
  issue: 9
  year: 2017
  ident: 10.1016/j.cor.2025.107041_b6
  article-title: Scheduling independent moldable tasks on multi-cores with gpus
  publication-title: IEEE Trans. Parallel Distrib. Syst.
  doi: 10.1109/TPDS.2017.2675891
– start-page: 1
  year: 2023
  ident: 10.1016/j.cor.2025.107041_b51
  article-title: Approximation algorithms for scheduling monotonic moldable tasks on multiple platforms
  publication-title: J. Sched.
– start-page: 172
  year: 2018
  ident: 10.1016/j.cor.2025.107041_b32
  article-title: Scheduling monotone moldable jobs in linear time
– year: 2024
  ident: 10.1016/j.cor.2025.107041_b28
  article-title: 3/2-dual approximation for cpu/gpu scheduling
– volume: 16
  start-page: 217
  issue: 02
  year: 2005
  ident: 10.1016/j.cor.2025.107041_b16
  article-title: Scheduling on large scale distributed platforms: from models to implementations
  publication-title: Internat. J. Found. Comput. Sci.
  doi: 10.1142/S0129054105002954
– start-page: 103
  year: 2019
  ident: 10.1016/j.cor.2025.107041_b35
  article-title: Linear time algorithms for multiple cluster scheduling and multiple strip packing
– ident: 10.1016/j.cor.2025.107041_b50
  doi: 10.1145/140901.141909
– volume: 53
  start-page: 1
  issue: 3
  year: 2020
  ident: 10.1016/j.cor.2025.107041_b4
  article-title: Scheduling on two types of resources: a survey
  publication-title: ACM Comput. Surv.
  doi: 10.1145/3387110
– start-page: 878
  year: 2012
  ident: 10.1016/j.cor.2025.107041_b10
  article-title: Tight approximation for scheduling parallel jobs on identical clusters
– volume: 306
  start-page: 567
  issue: 2
  year: 2023
  ident: 10.1016/j.cor.2025.107041_b54
  article-title: An improved approximation algorithm for scheduling monotonic moldable tasks
  publication-title: European J. Oper. Res.
  doi: 10.1016/j.ejor.2022.08.034
– volume: 34
  start-page: 144
  issue: 1
  year: 1987
  ident: 10.1016/j.cor.2025.107041_b29
  article-title: Using dual approximation algorithms for scheduling problems theoretical and practical results
  publication-title: J. ACM
  doi: 10.1145/7531.7535
– start-page: 389
  year: 2018
  ident: 10.1016/j.cor.2025.107041_b27
  article-title: Effectiveness of moldable and malleable scheduling in deep learning tasks
– volume: 62
  start-page: 1189
  issue: 7
  year: 2011
  ident: 10.1016/j.cor.2025.107041_b5
  article-title: Berth and quay crane allocation: a moldable task scheduling model
  publication-title: J. Oper. Res. Soc.
  doi: 10.1057/jors.2010.54
– start-page: 1
  year: 2024
  ident: 10.1016/j.cor.2025.107041_b26
  article-title: A constant-factor approximation for generalized malleable scheduling under m♮-concave processing speeds
  publication-title: Math. Program.
– volume: 52
  start-page: 478
  issue: 4
  year: 2020
  ident: 10.1016/j.cor.2025.107041_b12
  article-title: Three parallel task assignment problems with shared resources
  publication-title: IISE Trans.
  doi: 10.1080/24725854.2019.1680907
– volume: 20
  start-page: 940
  issue: 7
  year: 2009
  ident: 10.1016/j.cor.2025.107041_b19
  article-title: Scheduling parallel task graphs on (almost) homogeneous multicluster platforms
  publication-title: IEEE Trans. Parallel Distrib. Syst.
  doi: 10.1109/TPDS.2009.11
– volume: 21
  start-page: 905
  issue: 7
  year: 2009
  ident: 10.1016/j.cor.2025.107041_b45
  article-title: Cooperation in multi-organization scheduling
  publication-title: Concurr. Comput.: Pr. Exp.
  doi: 10.1002/cpe.1378
– start-page: 78
  year: 2013
  ident: 10.1016/j.cor.2025.107041_b17
  article-title: A (2+ ϵ)-approximation for scheduling parallel jobs in platforms
– start-page: 774
  year: 2005
  ident: 10.1016/j.cor.2025.107041_b48
  article-title: Two level job-scheduling strategies for a computational grid
– volume: 170
  start-page: 489
  year: 2015
  ident: 10.1016/j.cor.2025.107041_b2
  article-title: Workforce minimization for a mixed-model assembly line in the automotive industry
  publication-title: Int. J. Prod. Econ.
  doi: 10.1016/j.ijpe.2015.05.038
– year: 2009
  ident: 10.1016/j.cor.2025.107041_b14
– volume: 37
  start-page: 401
  issue: 2
  year: 2007
  ident: 10.1016/j.cor.2025.107041_b44
  article-title: A 32-approximation algorithm for scheduling independent monotonic malleable tasks
  publication-title: SIAM J. Comput.
  doi: 10.1137/S0097539701385995
– start-page: 302
  year: 2005
  ident: 10.1016/j.cor.2025.107041_b20
  article-title: Hierarchical scheduling for moldable tasks
SSID ssj0000721
Score 2.4613106
Snippet This paper examines task scheduling in homogeneous multi-cluster platforms, equipped with Graphics Processing Units (GPUs), with the aim of minimizing the...
SourceID crossref
elsevier
SourceType Index Database
Publisher
StartPage 107041
SubjectTerms Approximation algorithm
Dual approximation algorithm
Moldable task scheduling
Multi-cluster platform
Title Scheduling moldable tasks on homogeneous multi-cluster platforms with GPUs
URI https://dx.doi.org/10.1016/j.cor.2025.107041
Volume 179
WOSCitedRecordID wos001446714600001&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: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  issn: 0305-0548
  databaseCode: AIEXJ
  dateStart: 19950101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.sciencedirect.com
  omitProxy: false
  ssIdentifier: ssj0000721
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1Lb9QwELZgixAceCyglpd84IQUyZvYsXOsUHlUqKpoK_YW-REDpU1Wm6Tqz2fs2NnQggRIXKxotHZW832yx5N5IPQKLH6easYSkVOdUMVIIpgySSY5HNhcqsx6pD_ygwOxXBaHodFl69sJ8LoWl5fF6r9CDTIA26XO_gXc46IggGcAHUaAHcY_Av4IYDC9zzI_b86MT43qZPvdfxf42pw3MLNyga8-ljDRZ72rleDaSXfOgA3pbu8OT9qp4Rq7P7SeK82qWocYulAtaPQqf-69PSzrL1aGU3Hql_7U19dkS6CoacKPgwciZWO0anCLxdSYTRyST8dyUYFsKKM5brVD45hr2_bgQTgFrbsSrSkDCSdDQawr1bCP3Lpu2dR1csiL7CbaSjkrxAxt7X7YW-5vjmHuk-7G_xE_afvgvisv-rVRMjE0jh-ge-GGgHcHZB-iG1U9R7djgsIc3Y9Q4LAvz9HdSVXJR2h_wwAcGYA9A3BT4wkD8E8MwCMDsGMAdgx4jE7e7h2_eZ-ElhmJTgntYFSmUCy3VhFacWkp0TlcUlOrMy2sodJQRUAs1IIoSYxRVEkhwUgxemF49gTN6qauthEWCytywuUiJxRu0VxqUlnl5xQyq_gOeh11Vq6GyihlDBk8LUHBpVNwOSh4B9Go1TKYdoPJVgIFfj_t6b9Ne4bubHj6HM26dV-9QLf0RfetXb8MRPkBjjt0bA
linkProvider Elsevier
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=Scheduling+moldable+tasks+on+homogeneous+multi-cluster+platforms+with+GPUs&rft.jtitle=Computers+%26+operations+research&rft.au=Wu%2C+Fangfang&rft.au=Zhang%2C+Run&rft.au=Zhang%2C+Xiandong&rft.date=2025-07-01&rft.pub=Elsevier+Ltd&rft.issn=0305-0548&rft.volume=179&rft_id=info:doi/10.1016%2Fj.cor.2025.107041&rft.externalDocID=S0305054825000693
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0305-0548&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0305-0548&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0305-0548&client=summon