A multilevel algorithm for scalable independent task assignment

Assigning a large number of independent tasks to heterogeneous processors is a fundamental problem in modern computing, with applications in many domains such as cloud services, web crawling, and AI training. Exact and matheuristic approaches deliver high-quality assignments but incur superlinear or...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Future generation computer systems Jg. 176; S. 108183
Hauptverfasser: Burhan Tabak, H., Kartal Tabak, E., Aykanat, Cevdet
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.03.2026
Schlagworte:
ISSN:0167-739X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Assigning a large number of independent tasks to heterogeneous processors is a fundamental problem in modern computing, with applications in many domains such as cloud services, web crawling, and AI training. Exact and matheuristic approaches deliver high-quality assignments but incur superlinear or even exponential runtime costs, making them impractical, especially on large problem instances. Conversely, lightweight heuristics run efficiently at scale but often produce assignments with much lower quality. To address this issue, we present the first multilevel framework for the independent task assignment problem that maintains an end-to-end linear runtime bound of O(KN), where K×N is the size of the expected-time-to-compute matrix, with K and N respectively representing the number of processors and tasks. We propose (i) novel high-quality coarsening metrics that numerically define task characteristics and similarity; (ii) an efficient and effective matching algorithm that incorporates these metrics while maintaining linear time complexity with respect to the input size; (iii) an initial solution scheme that generates base solutions using complementary heuristics, which are disjointly projected back through the uncoarsening levels; (iv) an effective and efficient uncoarsening algorithm that iteratively improves assignment quality with different refinement algorithms. Extensive experimental evaluations involving hundreds of millions of tasks demonstrate that our algorithm achieves significantly higher quality and runs faster than known high-quality heuristics, making it a practical choice for the problem instances at high scale.
AbstractList Assigning a large number of independent tasks to heterogeneous processors is a fundamental problem in modern computing, with applications in many domains such as cloud services, web crawling, and AI training. Exact and matheuristic approaches deliver high-quality assignments but incur superlinear or even exponential runtime costs, making them impractical, especially on large problem instances. Conversely, lightweight heuristics run efficiently at scale but often produce assignments with much lower quality. To address this issue, we present the first multilevel framework for the independent task assignment problem that maintains an end-to-end linear runtime bound of O(KN), where K×N is the size of the expected-time-to-compute matrix, with K and N respectively representing the number of processors and tasks. We propose (i) novel high-quality coarsening metrics that numerically define task characteristics and similarity; (ii) an efficient and effective matching algorithm that incorporates these metrics while maintaining linear time complexity with respect to the input size; (iii) an initial solution scheme that generates base solutions using complementary heuristics, which are disjointly projected back through the uncoarsening levels; (iv) an effective and efficient uncoarsening algorithm that iteratively improves assignment quality with different refinement algorithms. Extensive experimental evaluations involving hundreds of millions of tasks demonstrate that our algorithm achieves significantly higher quality and runs faster than known high-quality heuristics, making it a practical choice for the problem instances at high scale.
ArticleNumber 108183
Author Burhan Tabak, H.
Aykanat, Cevdet
Kartal Tabak, E.
Author_xml – sequence: 1
  givenname: H.
  orcidid: 0009-0002-5043-6921
  surname: Burhan Tabak
  fullname: Burhan Tabak, H.
  email: burhantabak@alumni.bilkent.edu.tr
– sequence: 2
  givenname: E.
  orcidid: 0009-0002-3859-9269
  surname: Kartal Tabak
  fullname: Kartal Tabak, E.
  email: tabak@cs.bilkent.edu.tr
– sequence: 3
  givenname: Cevdet
  orcidid: 0000-0002-4559-1321
  surname: Aykanat
  fullname: Aykanat, Cevdet
  email: aykanat@cs.bilkent.edu.tr
BookMark eNp9j01LAzEQhnOoYFv9Bx7yB7bmo91sL0opaoWCFwVvIR-TmprdLUm24L83ZT17mWHe4X1nnhmadH0HCN1RsqCE1vfHhRvyEGHBCFsVqaENn6BpWYlK8PXnNZqldCSEUMHpFD1ucDuE7AOcIWAVDn30-avFro84GRWUDoB9Z-EEpXQZZ5W-sUrJH7q2zDfoyqmQ4Pavz9HH89P7dlft315et5t9ZdhK5IqKco41zmmyttrYelUzqhvKmbICVE20Ykta1842GngjnDJLzszaEE2KrvkcLcdcE_uUIjh5ir5V8UdSIi_g8ihHcHkBlyN4sT2MNii_nT1EmYyHzoD1EUyWtvf_B_wCXQRoRg
Cites_doi 10.1007/s10462-022-10247-9
10.1016/0305-0548(91)90034-O
10.1145/322003.322011
10.1006/jpdc.1997.1392
10.1006/jpdc.2000.1714
10.1016/j.jpdc.2007.03.003
10.1145/361011.361064
10.1109/92.748202
10.1145/1362622.1362674
10.1109/TPDS.2013.107
10.1006/jpdc.1999.1581
10.1145/321941.321951
10.1111/j.1540-5915.1980.tb01163.x
10.1016/j.cor.2023.106504
10.1007/s12532-009-0002-8
10.1145/2049662.2049663
10.1109/71.780863
10.1016/j.parco.2008.12.006
10.1145/3340531.3412684
ContentType Journal Article
Copyright 2025 The Author(s)
Copyright_xml – notice: 2025 The Author(s)
DBID 6I.
AAFTH
AAYXX
CITATION
DOI 10.1016/j.future.2025.108183
DatabaseName ScienceDirect Open Access Titles
Elsevier:ScienceDirect:Open Access
CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
ExternalDocumentID 10_1016_j_future_2025_108183
S0167739X25004777
GroupedDBID --K
--M
-~X
.DC
.~1
0R~
1B1
1~.
1~5
29H
4.4
457
4G.
5GY
5VS
6I.
7-5
71M
8P~
9JN
AAEDT
AAEDW
AAFTH
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAQXK
AATTM
AAXKI
AAXUO
AAYFN
AAYWO
ABBOA
ABDPE
ABFNM
ABJNI
ABMAC
ABWVN
ABXDB
ACDAQ
ACGFS
ACLOT
ACNNM
ACRLP
ACRPL
ACZNC
ADBBV
ADEZE
ADJOM
ADMUD
ADNMO
AEBSH
AEIPS
AEKER
AFJKZ
AFTJW
AGHFR
AGQPQ
AGUBO
AGYEJ
AHHHB
AHZHX
AIALX
AIEXJ
AIIUN
AIKHN
AITUG
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
ANKPU
AOUOD
APXCP
ASPBG
AVWKF
AXJTR
AZFZN
BKOJK
BLXMC
CS3
EBS
EFJIC
EFKBS
EFLBG
EJD
EO8
EO9
EP2
EP3
F5P
FDB
FEDTE
FGOYB
FIRID
FNPLU
FYGXN
G-Q
GBLVA
GBOLZ
HLZ
HVGLF
HZ~
IHE
J1W
KOM
LG9
M41
MO0
MS~
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
PC.
Q38
R2-
ROL
RPZ
SBC
SDF
SDG
SES
SEW
SPC
SPCBC
SSV
SSZ
T5K
UHS
WUQ
XPP
ZMT
~G-
~HD
9DU
AAYXX
CITATION
ID FETCH-LOGICAL-c257t-1717328ffb09dbcd65621b8132ad7ea60ba24166fd8be387fac432c9c0b0416b3
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001600346200001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0167-739X
IngestDate Sat Nov 29 07:09:42 EST 2025
Sat Nov 15 16:53:36 EST 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Keywords Heterogeneous systems
Multilevel framework
Parallel and distributed computing
Independent task assignment
Geographically distributed web crawling
Distributed and parallelized LLM training
Load balancing
Language English
License This is an open access article under the CC BY-NC license.
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c257t-1717328ffb09dbcd65621b8132ad7ea60ba24166fd8be387fac432c9c0b0416b3
ORCID 0009-0002-5043-6921
0009-0002-3859-9269
0000-0002-4559-1321
OpenAccessLink https://dx.doi.org/10.1016/j.future.2025.108183
ParticipantIDs crossref_primary_10_1016_j_future_2025_108183
elsevier_sciencedirect_doi_10_1016_j_future_2025_108183
PublicationCentury 2000
PublicationDate March 2026
2026-03-00
PublicationDateYYYYMMDD 2026-03-01
PublicationDate_xml – month: 03
  year: 2026
  text: March 2026
PublicationDecade 2020
PublicationTitle Future generation computer systems
PublicationYear 2026
Publisher Elsevier B.V
Publisher_xml – name: Elsevier B.V
References Fonseca, Figueiroa, Toffolo (bib0020) 2024; 163
Davis, Hu (bib0026) 2011; 38
De, Morton (bib0015) 1980; 11
Bruno, Coffman, Sethi (bib0001) 1974; 17
Fey, Lenssen (bib0027) 2019; abs/1903.02428
MAWI Working Group, mawi_201512020330: Backbone traffic graph, 2015, SuiteSparse Matrix Collection, matrix #2804
Kolmogorov (bib0021) 2009; 1
Leskovec, Kleinberg, Faloutsos (bib0034) 2005
Uludoğan, Balal, Akkurt, Turker, Gungor, Üsküdarlı (bib0024) 2024
H. Zeng, H. Zhou, A. Srivastava, R. Kannan, V.K. Prasanna, Flickr dataset, 2024
Luo, Lü, Shi (bib0019) 2007; 67
S. Beamer, K. Asanović, D. Patterson, GAP-kron: Kronecker graph, 2017a, SuiteSparse Matrix Collection, matrix #2855
Ritchie, Levine (bib0018) 2004
CMU-LTI, The ClueWeb09 Dataset, 2009
Horowitz, Sahni (bib0014) 1976; 23
Wang, Siegel, Roychowdhury, Maciejewski (bib0008) 1997; 47
S. Beamer, K. Asanović, D. Patterson, The GAP Benchmark Suite, 2017b.
Ibarra, Kim (bib0002) 1977; 24
Bui, Jones (bib0012) 1993
Armstrong, Hensgen, Kidd (bib0003) 1998
Karypis, Aggarwal, Kumar, Shekhar (bib0011) 1999; 7
H. Kwak, C. Lee, H. Park, S. Moon, twitter7: Twitter follower network, 2010a, SuiteSparse Matrix Collection, matrix #2796
Wu, Shu (bib0006) 2001
Tabak, Cambazoglu, Aykanat (bib0009) 2014; 25
C.C. Foundation, Common Crawl February/March 2024 Crawl Archive (CC-MAIN-2024-10), 2024
J. Sybrandt, I. Tyagin, M. Shtutman, I. Safro, AGATHA_2015: Automatic Graph-mining And Transformer based Hypothesis Generation Approach graph, 2020a, (SuiteSparse Matrix Collection, matrix #2893
.
Kwak, Lee, Park, Moon (bib0036) 2010
Zeng, Zhou, Srivastava, Kannan, Prasanna (bib0029) 2020
Durasević, Jakobović (bib0013) 2023; 56
Ali, Siegel, Maheswaran, Hensgen, Ali (bib0025) 2000
J. Leskovec, J. Kleinberg, C. Faloutsos, as-Skitter: Internet topology graph, 2005a, SuiteSparse Matrix Collection, matrix #2321
Catalyurek, Aykanat (bib0010) 1999; 10
Hariri, Potts (bib0016) 1991; 18
R. Tromble, language: finite-state machine for natural-language processing, 2003, SuiteSparse Matrix Collection, matrix #982
S. Williams, L. Oliker, R. Vuduc, J. Shalf, K. Yelick, J. Demmel, webbase-1M: Web connectivity matrix, 2008, SuiteSparse Matrix Collection, matrix #2379
Freund, Gherrity, Ambrosius, Campbell, Halderman, Hensgen, Keith, Kidd, Kussow, Lima, Mirabile, Moore, Rust, Siegel (bib0004) 1998
Williams, Oliker, Vuduc, Shalf, Yelick, Demmel (bib0032) 2009; 35
Braun, Siegel, Beck, Bölöni, Maheswaran, Reuther, Robertson, Theys, Yao, Hensgen, Freund (bib0005) 2001; 61
Maheswaran, Ali, Siegel, Hensgen, Freund (bib0007) 1999; 59
Parsa, Entezari-Maleki (bib0017) 2009; 3
Sybrandt, Tyagin, Shtutman, Safro (bib0040) 2020
Ibarra (10.1016/j.future.2025.108183_bib0002) 1977; 24
10.1016/j.future.2025.108183_bib0037
Durasević (10.1016/j.future.2025.108183_bib0013) 2023; 56
Braun (10.1016/j.future.2025.108183_bib0005) 2001; 61
10.1016/j.future.2025.108183_bib0039
Catalyurek (10.1016/j.future.2025.108183_bib0010) 1999; 10
Zeng (10.1016/j.future.2025.108183_bib0029) 2020
10.1016/j.future.2025.108183_bib0038
10.1016/j.future.2025.108183_bib0033
10.1016/j.future.2025.108183_bib0035
Williams (10.1016/j.future.2025.108183_bib0032) 2009; 35
Leskovec (10.1016/j.future.2025.108183_bib0034) 2005
Luo (10.1016/j.future.2025.108183_bib0019) 2007; 67
Fonseca (10.1016/j.future.2025.108183_bib0020) 2024; 163
Wang (10.1016/j.future.2025.108183_bib0008) 1997; 47
10.1016/j.future.2025.108183_bib0031
Tabak (10.1016/j.future.2025.108183_bib0009) 2014; 25
De (10.1016/j.future.2025.108183_bib0015) 1980; 11
10.1016/j.future.2025.108183_bib0030
Freund (10.1016/j.future.2025.108183_bib0004) 1998
Sybrandt (10.1016/j.future.2025.108183_bib0040) 2020
Hariri (10.1016/j.future.2025.108183_bib0016) 1991; 18
Horowitz (10.1016/j.future.2025.108183_bib0014) 1976; 23
10.1016/j.future.2025.108183_bib0028
10.1016/j.future.2025.108183_bib0022
Kolmogorov (10.1016/j.future.2025.108183_bib0021) 2009; 1
Bruno (10.1016/j.future.2025.108183_bib0001) 1974; 17
10.1016/j.future.2025.108183_bib0023
Parsa (10.1016/j.future.2025.108183_bib0017) 2009; 3
Fey (10.1016/j.future.2025.108183_bib0027) 2019; abs/1903.02428
Ali (10.1016/j.future.2025.108183_bib0025) 2000
Wu (10.1016/j.future.2025.108183_bib0006) 2001
Davis (10.1016/j.future.2025.108183_bib0026) 2011; 38
Kwak (10.1016/j.future.2025.108183_bib0036) 2010
Karypis (10.1016/j.future.2025.108183_bib0011) 1999; 7
Bui (10.1016/j.future.2025.108183_bib0012) 1993
Armstrong (10.1016/j.future.2025.108183_bib0003) 1998
Ritchie (10.1016/j.future.2025.108183_bib0018) 2004
Maheswaran (10.1016/j.future.2025.108183_bib0007) 1999; 59
Uludoğan (10.1016/j.future.2025.108183_bib0024) 2024
10.1016/j.future.2025.108183_bib0041
References_xml – reference: J. Leskovec, J. Kleinberg, C. Faloutsos, as-Skitter: Internet topology graph, 2005a, SuiteSparse Matrix Collection, matrix #2321,
– year: 2020
  ident: bib0040
  article-title: AGATHA: Automatic graph-mining and transformer based hypothesis generation approach
  publication-title: Proceedings of the 29th ACM International Conference on Information & Knowledge Management (CIKM)
– start-page: 184
  year: 1998
  end-page: 199
  ident: bib0004
  article-title: Scheduling resources in multi-user, heterogeneous, computing environments with SmartNet
  publication-title: Proceedings of the Seventh IEEE Heterogeneous Computing Workshop
– volume: 25
  start-page: 1244
  year: 2014
  end-page: 1256
  ident: bib0009
  article-title: Improving the performance of independent task assignment heuristics MinMin, MaxMin and Sufferage
  publication-title: IEEE Trans. Parallel Distrib. Syst.
– volume: 11
  start-page: 586
  year: 1980
  end-page: 602
  ident: bib0015
  article-title: Scheduling to minimize makespan on unequal parallel processors
  publication-title: Decision Sci.
– reference: J. Sybrandt, I. Tyagin, M. Shtutman, I. Safro, AGATHA_2015: Automatic Graph-mining And Transformer based Hypothesis Generation Approach graph, 2020a, (SuiteSparse Matrix Collection, matrix #2893,
– volume: 47
  start-page: 8
  year: 1997
  end-page: 22
  ident: bib0008
  article-title: Task matching and scheduling in heterogeneous computing environments using a genetic-algorithm-based approach
  publication-title: J. Parallel Distrib. Comput.
– volume: 38
  year: 2011
  ident: bib0026
  article-title: The University of Florida Sparse Matrix Collection
  publication-title: ACM Trans. Math. Softw.
– start-page: 185
  year: 2000
  end-page: 199
  ident: bib0025
  article-title: Task execution time modeling for heterogeneous computing systems
  publication-title: Proc. 9th Heterogeneous Computing Workshop
– start-page: 79
  year: 1998
  end-page: 87
  ident: bib0003
  article-title: The relative performance of various mapping algorithms is independent of sizable variances in run-time predictions
  publication-title: Proceedings of the 7th IEEE Heterogeneous Computing Workshop
– reference: S. Williams, L. Oliker, R. Vuduc, J. Shalf, K. Yelick, J. Demmel, webbase-1M: Web connectivity matrix, 2008, SuiteSparse Matrix Collection, matrix #2379,
– volume: 23
  start-page: 317
  year: 1976
  end-page: 327
  ident: bib0014
  article-title: Exact and approximate algorithms for scheduling nonidentical processors
  publication-title: J. ACM
– volume: 24
  start-page: 280
  year: 1977
  end-page: 289
  ident: bib0002
  article-title: Heuristic algorithms for scheduling independent tasks on nonidentical processors
  publication-title: J. ACM
– volume: 163
  year: 2024
  ident: bib0020
  article-title: A fix-and-optimize heuristic for the unrelated parallel machine scheduling problem
  publication-title: Comput. Oper. Res.
– volume: 56
  start-page: 3181
  year: 2023
  end-page: 3289
  ident: bib0013
  article-title: Heuristic and metaheuristic methods for the parallel unrelated machines scheduling problem: a survey
  publication-title: Artif. Intell. Rev.
– volume: 10
  start-page: 673
  year: 1999
  end-page: 693
  ident: bib0010
  article-title: Hypergraph-partitioning-based decomposition for parallel sparse-matrix vector multiplication
  publication-title: IEEE Trans. Parallel Distrib. Syst.
– start-page: 74
  year: 2001
  end-page: 79
  ident: bib0006
  article-title: A high-performance mapping algorithm for heterogeneous computing systems
  publication-title: Proceedings of the 15th International Parallel and Distributed Processing Symposium
– year: 2004
  ident: bib0018
  article-title: A hybrid ant algorithm for scheduling independent jobs in heterogeneous computing environments
  publication-title: PlanSIG 2004
– volume: 17
  start-page: 382
  year: 1974
  end-page: 387
  ident: bib0001
  article-title: Scheduling independent tasks to reduce mean finishing time
  publication-title: Commun. ACM
– volume: 7
  start-page: 69
  year: 1999
  end-page: 79
  ident: bib0011
  article-title: Multilevel hypergraph partitioning: applications in VLSI domain
  publication-title: IEEE Trans. Very Large Scale Integr. VLSI Syst.
– start-page: 591
  year: 2010
  end-page: 600
  ident: bib0036
  article-title: What is Twitter, a social network or a news media?
  publication-title: WWW ’10: Proc. the 19Th Intl. Conf. on World Wide Web
– volume: 61
  start-page: 810
  year: 2001
  end-page: 837
  ident: bib0005
  article-title: A comparison of eleven static heuristics for mapping a class of independent tasks onto heterogeneous distributed computing systems
  publication-title: J. Parallel Distrib. Comput.
– volume: 3
  start-page: 91
  year: 2009
  end-page: 99
  ident: bib0017
  article-title: RASA: A new grid task scheduling algorithm
  publication-title: Intern. J. Digital Content Technol. Appl.
– start-page: 445
  year: 1993
  end-page: 452
  ident: bib0012
  article-title: A heuristic for reducing fill-in in sparse matrix factorization
  publication-title: Proceedings of the 6th SIAM Conference on Parallel Processing for Scientific Computing
– volume: 18
  start-page: 323
  year: 1991
  end-page: 331
  ident: bib0016
  article-title: Heuristics for scheduling unrelated parallel machines
  publication-title: Comput. Oper. Res.
– volume: abs/1903.02428
  year: 2019
  ident: bib0027
  article-title: Fast graph representation learning with PyTorch geometric
  publication-title: CoRR
– reference: H. Kwak, C. Lee, H. Park, S. Moon, twitter7: Twitter follower network, 2010a, SuiteSparse Matrix Collection, matrix #2796,
– volume: 59
  start-page: 107
  year: 1999
  end-page: 131
  ident: bib0007
  article-title: Dynamic mapping of a class of independent tasks onto heterogeneous computing systems
  publication-title: J. Parallel Distrib. Comput.
– start-page: 10103
  year: 2024
  end-page: 10117
  ident: bib0024
  article-title: TURNA: a Turkish encoder-decoder language model for enhanced understanding and generation
  publication-title: Findings of the Association for Computational Linguistics: ACL 2024
– reference: S. Beamer, K. Asanović, D. Patterson, The GAP Benchmark Suite, 2017b.
– volume: 35
  start-page: 178
  year: 2009
  end-page: 194
  ident: bib0032
  article-title: Optimization of sparse matrix–vector multiplication on emerging multicore platforms
  publication-title: Parallel Comput.
– reference: MAWI Working Group, mawi_201512020330: Backbone traffic graph, 2015, SuiteSparse Matrix Collection, matrix #2804,
– reference: S. Beamer, K. Asanović, D. Patterson, GAP-kron: Kronecker graph, 2017a, SuiteSparse Matrix Collection, matrix #2855,
– volume: 1
  start-page: 43
  year: 2009
  end-page: 67
  ident: bib0021
  article-title: Blossom V: A new implementation of a minimum cost perfect matching algorithm
  publication-title: Math. Program. Comput.
– reference: .
– volume: 67
  start-page: 695
  year: 2007
  end-page: 714
  ident: bib0019
  article-title: A revisit of fast greedy heuristics for mapping a class of independent tasks onto heterogeneous computing systems
  publication-title: J. Parallel Distrib. Comput.
– reference: CMU-LTI, The ClueWeb09 Dataset, 2009,
– year: 2020
  ident: bib0029
  article-title: GraphSAINT: graph sampling based inductive learning method
  publication-title: International Conference on Learning Representations (ICLR)
– reference: R. Tromble, language: finite-state machine for natural-language processing, 2003, SuiteSparse Matrix Collection, matrix #982,
– start-page: 177
  year: 2005
  end-page: 187
  ident: bib0034
  article-title: Graphs over time: densification laws, shrinking diameters and possible explanations
  publication-title: Proceedings of the Eleventh ACM SIGKDD International Conference on Knowledge Discovery in Data Mining
– reference: C.C. Foundation, Common Crawl February/March 2024 Crawl Archive (CC-MAIN-2024-10), 2024,
– reference: H. Zeng, H. Zhou, A. Srivastava, R. Kannan, V.K. Prasanna, Flickr dataset, 2024,
– volume: 56
  start-page: 3181
  issue: 4
  year: 2023
  ident: 10.1016/j.future.2025.108183_bib0013
  article-title: Heuristic and metaheuristic methods for the parallel unrelated machines scheduling problem: a survey
  publication-title: Artif. Intell. Rev.
  doi: 10.1007/s10462-022-10247-9
– volume: 18
  start-page: 323
  issue: 3
  year: 1991
  ident: 10.1016/j.future.2025.108183_bib0016
  article-title: Heuristics for scheduling unrelated parallel machines
  publication-title: Comput. Oper. Res.
  doi: 10.1016/0305-0548(91)90034-O
– ident: 10.1016/j.future.2025.108183_bib0022
– start-page: 591
  year: 2010
  ident: 10.1016/j.future.2025.108183_bib0036
  article-title: What is Twitter, a social network or a news media?
– ident: 10.1016/j.future.2025.108183_bib0028
– ident: 10.1016/j.future.2025.108183_bib0041
– start-page: 184
  year: 1998
  ident: 10.1016/j.future.2025.108183_bib0004
  article-title: Scheduling resources in multi-user, heterogeneous, computing environments with SmartNet
– volume: 24
  start-page: 280
  issue: 2
  year: 1977
  ident: 10.1016/j.future.2025.108183_bib0002
  article-title: Heuristic algorithms for scheduling independent tasks on nonidentical processors
  publication-title: J. ACM
  doi: 10.1145/322003.322011
– ident: 10.1016/j.future.2025.108183_bib0037
– volume: 47
  start-page: 8
  issue: 1
  year: 1997
  ident: 10.1016/j.future.2025.108183_bib0008
  article-title: Task matching and scheduling in heterogeneous computing environments using a genetic-algorithm-based approach
  publication-title: J. Parallel Distrib. Comput.
  doi: 10.1006/jpdc.1997.1392
– ident: 10.1016/j.future.2025.108183_bib0035
– year: 2004
  ident: 10.1016/j.future.2025.108183_bib0018
  article-title: A hybrid ant algorithm for scheduling independent jobs in heterogeneous computing environments
– ident: 10.1016/j.future.2025.108183_bib0033
– volume: 61
  start-page: 810
  issue: 6
  year: 2001
  ident: 10.1016/j.future.2025.108183_bib0005
  article-title: A comparison of eleven static heuristics for mapping a class of independent tasks onto heterogeneous distributed computing systems
  publication-title: J. Parallel Distrib. Comput.
  doi: 10.1006/jpdc.2000.1714
– volume: 67
  start-page: 695
  year: 2007
  ident: 10.1016/j.future.2025.108183_bib0019
  article-title: A revisit of fast greedy heuristics for mapping a class of independent tasks onto heterogeneous computing systems
  publication-title: J. Parallel Distrib. Comput.
  doi: 10.1016/j.jpdc.2007.03.003
– volume: 3
  start-page: 91
  issue: 4
  year: 2009
  ident: 10.1016/j.future.2025.108183_bib0017
  article-title: RASA: A new grid task scheduling algorithm
  publication-title: Intern. J. Digital Content Technol. Appl.
– volume: 17
  start-page: 382
  issue: 7
  year: 1974
  ident: 10.1016/j.future.2025.108183_bib0001
  article-title: Scheduling independent tasks to reduce mean finishing time
  publication-title: Commun. ACM
  doi: 10.1145/361011.361064
– volume: 7
  start-page: 69
  issue: 1
  year: 1999
  ident: 10.1016/j.future.2025.108183_bib0011
  article-title: Multilevel hypergraph partitioning: applications in VLSI domain
  publication-title: IEEE Trans. Very Large Scale Integr. VLSI Syst.
  doi: 10.1109/92.748202
– ident: 10.1016/j.future.2025.108183_bib0031
  doi: 10.1145/1362622.1362674
– ident: 10.1016/j.future.2025.108183_bib0023
– volume: 25
  start-page: 1244
  issue: 5
  year: 2014
  ident: 10.1016/j.future.2025.108183_bib0009
  article-title: Improving the performance of independent task assignment heuristics MinMin, MaxMin and Sufferage
  publication-title: IEEE Trans. Parallel Distrib. Syst.
  doi: 10.1109/TPDS.2013.107
– volume: 59
  start-page: 107
  issue: 2
  year: 1999
  ident: 10.1016/j.future.2025.108183_bib0007
  article-title: Dynamic mapping of a class of independent tasks onto heterogeneous computing systems
  publication-title: J. Parallel Distrib. Comput.
  doi: 10.1006/jpdc.1999.1581
– volume: 23
  start-page: 317
  issue: 2
  year: 1976
  ident: 10.1016/j.future.2025.108183_bib0014
  article-title: Exact and approximate algorithms for scheduling nonidentical processors
  publication-title: J. ACM
  doi: 10.1145/321941.321951
– start-page: 74
  year: 2001
  ident: 10.1016/j.future.2025.108183_bib0006
  article-title: A high-performance mapping algorithm for heterogeneous computing systems
– start-page: 185
  year: 2000
  ident: 10.1016/j.future.2025.108183_bib0025
  article-title: Task execution time modeling for heterogeneous computing systems
– year: 2020
  ident: 10.1016/j.future.2025.108183_bib0040
  article-title: AGATHA: Automatic graph-mining and transformer based hypothesis generation approach
– volume: 11
  start-page: 586
  issue: 4
  year: 1980
  ident: 10.1016/j.future.2025.108183_bib0015
  article-title: Scheduling to minimize makespan on unequal parallel processors
  publication-title: Decision Sci.
  doi: 10.1111/j.1540-5915.1980.tb01163.x
– volume: 163
  year: 2024
  ident: 10.1016/j.future.2025.108183_bib0020
  article-title: A fix-and-optimize heuristic for the unrelated parallel machine scheduling problem
  publication-title: Comput. Oper. Res.
  doi: 10.1016/j.cor.2023.106504
– volume: abs/1903.02428
  year: 2019
  ident: 10.1016/j.future.2025.108183_bib0027
  article-title: Fast graph representation learning with PyTorch geometric
  publication-title: CoRR
– year: 2020
  ident: 10.1016/j.future.2025.108183_bib0029
  article-title: GraphSAINT: graph sampling based inductive learning method
– start-page: 445
  year: 1993
  ident: 10.1016/j.future.2025.108183_bib0012
  article-title: A heuristic for reducing fill-in in sparse matrix factorization
– volume: 1
  start-page: 43
  issue: 1
  year: 2009
  ident: 10.1016/j.future.2025.108183_bib0021
  article-title: Blossom V: A new implementation of a minimum cost perfect matching algorithm
  publication-title: Math. Program. Comput.
  doi: 10.1007/s12532-009-0002-8
– start-page: 177
  year: 2005
  ident: 10.1016/j.future.2025.108183_bib0034
  article-title: Graphs over time: densification laws, shrinking diameters and possible explanations
– start-page: 79
  year: 1998
  ident: 10.1016/j.future.2025.108183_bib0003
  article-title: The relative performance of various mapping algorithms is independent of sizable variances in run-time predictions
– volume: 38
  issue: 1
  year: 2011
  ident: 10.1016/j.future.2025.108183_bib0026
  article-title: The University of Florida Sparse Matrix Collection
  publication-title: ACM Trans. Math. Softw.
  doi: 10.1145/2049662.2049663
– volume: 10
  start-page: 673
  issue: 7
  year: 1999
  ident: 10.1016/j.future.2025.108183_bib0010
  article-title: Hypergraph-partitioning-based decomposition for parallel sparse-matrix vector multiplication
  publication-title: IEEE Trans. Parallel Distrib. Syst.
  doi: 10.1109/71.780863
– volume: 35
  start-page: 178
  issue: 3
  year: 2009
  ident: 10.1016/j.future.2025.108183_bib0032
  article-title: Optimization of sparse matrix–vector multiplication on emerging multicore platforms
  publication-title: Parallel Comput.
  doi: 10.1016/j.parco.2008.12.006
– start-page: 10103
  year: 2024
  ident: 10.1016/j.future.2025.108183_bib0024
  article-title: TURNA: a Turkish encoder-decoder language model for enhanced understanding and generation
– ident: 10.1016/j.future.2025.108183_bib0030
– ident: 10.1016/j.future.2025.108183_bib0038
– ident: 10.1016/j.future.2025.108183_bib0039
  doi: 10.1145/3340531.3412684
SSID ssj0001731
Score 2.4538808
Snippet Assigning a large number of independent tasks to heterogeneous processors is a fundamental problem in modern computing, with applications in many domains such...
SourceID crossref
elsevier
SourceType Index Database
Publisher
StartPage 108183
SubjectTerms Distributed and parallelized LLM training
Geographically distributed web crawling
Heterogeneous systems
Independent task assignment
Load balancing
Multilevel framework
Parallel and distributed computing
Title A multilevel algorithm for scalable independent task assignment
URI https://dx.doi.org/10.1016/j.future.2025.108183
Volume 176
WOSCitedRecordID wos001600346200001&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: 0167-739X
  databaseCode: AIEXJ
  dateStart: 19950201
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.sciencedirect.com
  omitProxy: false
  ssIdentifier: ssj0001731
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1BT9swGLW2wmGXjcGmFRjygVuVKonj2D6haCoqHCoOReotsp2EFlhAbYrYv-dz7CQdQwgOu0SV0zhRvpfnz9bz9xA6VkIyTjPmBbQwExTNPZkx3wtIBqNJzCJJrNkEm0z4bCYu3BaCVW0nwMqSPz6K-_8aamiDYJuts-8Id9spNMBvCDocIexwfFPgEysSvDVqoIG8vbqD6f_8dy0nXEFA6q1Si9b8thpUcnUzgBR6cVW2KpjGt7MuOGJclnMHFO1MIFwF6G6Nfb2cA1NMpZI1vY6HLZNLk993Z0btmeTPjSyllZ3kD5nbie1WIMK4k2A1i5JAtozUlrgdq7JNXgxM5TzyImXb1YProa2hAjP2kA67v_9dIfvZyNXqCRup2nVqe0lNL6nt5SPaChkVwHhbydlodt6O0wFzbpXu6ZuNlbX679-neTlx2UhGpjvos5tF4MRG_yv6kJe76Evj0IEdYe-hkwR3YMAtGDCAATdgwBtgwAYMuAPDN3R5Opr-GnvOMsPTwL2VFxhRRciLQvkiUzqDbD0MFA9ICF9gLmNfSUjZ4rjIuMoJZ4XUEQm10L7yoV2R76hX3pX5D4SJIFGhuRIFLSLmCx4zYrJvqiNKNaN95DXvI723lVHS1-LQR6x5aanL7mzWlgISXr1y_513OkCfOpgeol61XOc_0bZ-qBar5ZGDwRMzYXLG
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=A+multilevel+algorithm+for+scalable+independent+task+assignment&rft.jtitle=Future+generation+computer+systems&rft.au=Burhan+Tabak%2C+H.&rft.au=Kartal+Tabak%2C+E.&rft.au=Aykanat%2C+Cevdet&rft.date=2026-03-01&rft.issn=0167-739X&rft.volume=176&rft.spage=108183&rft_id=info:doi/10.1016%2Fj.future.2025.108183&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_future_2025_108183
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0167-739X&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0167-739X&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0167-739X&client=summon