Lifeline-based load balancing schemes for Asynchronous Many-Task runtimes in clusters
A popular approach to program scalable irregular applications is Asynchronous Many-Task (AMT) Programming. Here, programs define tasks according to task models such as dynamic independent tasks (DIT) or nested fork-join (NFJ). We consider cluster AMTs, in which a runtime system maps the tasks to wor...
Uloženo v:
| Vydáno v: | Parallel computing Ročník 116; s. 103020 |
|---|---|
| Hlavní autoři: | , , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
Elsevier B.V
01.07.2023
|
| Témata: | |
| ISSN: | 0167-8191 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | A popular approach to program scalable irregular applications is Asynchronous Many-Task (AMT) Programming. Here, programs define tasks according to task models such as dynamic independent tasks (DIT) or nested fork-join (NFJ). We consider cluster AMTs, in which a runtime system maps the tasks to worker threads in multiple processes.
Thereby, dynamic load balancing can be achieved via cooperative work stealing, coordinated work stealing, or work sharing. A well-performing cooperative work stealing variant is the lifeline scheme. While previous implementations of this scheme are restricted to single-worker processes, a recent hybrid extension combines it with intra-process work sharing between multiple workers. The hybrid scheme, which was proposed for both DIT and NFJ, comes at the price of a higher complexity.
This paper investigates whether this complexity is indispensable for multi-worker processes by contrasting the hybrid scheme with a novel pure work stealing extension of the lifeline scheme to multiple workers. We independently implemented the extension for DIT and NFJ. In experiments based on four benchmarks, we observed the pure scheme to be on a par or even outperform the hybrid one by up to 18% for DIT and up to 5% for NFJ.
Building on this main result, we studied a modification of the pure scheme, which prefers local over global victims, and more heavily loaded over less loaded ones. The modification improves the performance of the pure scheme by up to 15%. Finally, we explored whether the lifeline scheme can profit from a change to coordinated work stealing. We developed a coordinated multi-worker implementation for DIT and observed a performance improvement over the cooperative scheme by up to 17%.
•Pure work stealing may outperform hybrid work stealing/sharing schemes in clusters.•Lifeline-based work stealing may profit from locality and load-aware victim selection.•Work stealing in clusters may profit from the usage of shared vs. private task queues. |
|---|---|
| AbstractList | A popular approach to program scalable irregular applications is Asynchronous Many-Task (AMT) Programming. Here, programs define tasks according to task models such as dynamic independent tasks (DIT) or nested fork-join (NFJ). We consider cluster AMTs, in which a runtime system maps the tasks to worker threads in multiple processes.
Thereby, dynamic load balancing can be achieved via cooperative work stealing, coordinated work stealing, or work sharing. A well-performing cooperative work stealing variant is the lifeline scheme. While previous implementations of this scheme are restricted to single-worker processes, a recent hybrid extension combines it with intra-process work sharing between multiple workers. The hybrid scheme, which was proposed for both DIT and NFJ, comes at the price of a higher complexity.
This paper investigates whether this complexity is indispensable for multi-worker processes by contrasting the hybrid scheme with a novel pure work stealing extension of the lifeline scheme to multiple workers. We independently implemented the extension for DIT and NFJ. In experiments based on four benchmarks, we observed the pure scheme to be on a par or even outperform the hybrid one by up to 18% for DIT and up to 5% for NFJ.
Building on this main result, we studied a modification of the pure scheme, which prefers local over global victims, and more heavily loaded over less loaded ones. The modification improves the performance of the pure scheme by up to 15%. Finally, we explored whether the lifeline scheme can profit from a change to coordinated work stealing. We developed a coordinated multi-worker implementation for DIT and observed a performance improvement over the cooperative scheme by up to 17%.
•Pure work stealing may outperform hybrid work stealing/sharing schemes in clusters.•Lifeline-based work stealing may profit from locality and load-aware victim selection.•Work stealing in clusters may profit from the usage of shared vs. private task queues. |
| ArticleNumber | 103020 |
| Author | Reitz, Mia Hardenbicker, Kai Fohry, Claudia Werner, Tobias |
| Author_xml | – sequence: 1 givenname: Mia surname: Reitz fullname: Reitz, Mia email: mia.reitz@uni-kassel.de – sequence: 2 givenname: Kai surname: Hardenbicker fullname: Hardenbicker, Kai – sequence: 3 givenname: Tobias surname: Werner fullname: Werner, Tobias – sequence: 4 givenname: Claudia surname: Fohry fullname: Fohry, Claudia |
| BookMark | eNqFkLtOAzEQRV0EiQT4Ahr_wIbxPuOCIop4SUE0SW3NesfEYWNHtoOUv2dDqCigGunOPSPNmbCR844YuxUwFSDqu-10j0H7aQ55MSQF5DBi42HTZDMhxSWbxLgFgLqcwZitl9ZQbx1lLUbqeO-x4y326LR17zzqDe0ocuMDn8ej05vgnT9E_orumK0wfvBwcMmeOtZx3R9iohCv2YXBPtLNz7xi68eH1eI5W749vSzmy0wXUKSMcjACKmNQoqhNhXVpoMauRa2FFrJsDIApEKFsZI7V0GgAZ63MK4laUnHF5PmuDj7GQEZpmzBZ71JA2ysB6uREbdW3E3Vyos5OBrb4xe6D3WE4_kPdnyka3vq0FFTUlpymzgbSSXXe_sl_AbjLgfg |
| CitedBy_id | crossref_primary_10_1007_s42979_024_02624_8 |
| Cites_doi | 10.1002/cpe.1631 10.1145/2517327.2442538 10.1145/1103845.1094852 10.1007/s11227-018-2238-4 10.1007/s11227-018-2234-8 10.1145/324133.324234 10.2307/3033543 |
| ContentType | Journal Article |
| Copyright | 2023 Elsevier B.V. |
| Copyright_xml | – notice: 2023 Elsevier B.V. |
| DBID | AAYXX CITATION |
| DOI | 10.1016/j.parco.2023.103020 |
| DatabaseName | CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Computer Science |
| ExternalDocumentID | 10_1016_j_parco_2023_103020 S0167819123000261 |
| GroupedDBID | --K --M -~X .DC .~1 0R~ 123 1B1 1~. 1~5 29O 4.4 457 4G. 5VS 6OB 7-5 71M 8P~ 9JN AAEDT AAEDW AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AATTM AAXKI AAXUO AAYFN AAYWO ABBOA ABDPE ABEFU ABFNM ABJNI ABMAC ABWVN ABXDB ACDAQ ACGFS ACNNM ACRLP ACRPL ACVFH ACZNC ADBBV ADCNI ADEZE ADJOM ADMUD ADNMO ADTZH AEBSH AECPX AEIPS AEKER AENEX AEUPX AFJKZ AFPUW AFTJW AGCQF AGHFR AGQPQ AGUBO AGYEJ AHHHB AHJVU AHZHX AIALX AIEXJ AIGII AIIUN AIKHN AITUG AKBMS AKRWK AKYEP ALMA_UNASSIGNED_HOLDINGS AMRAJ ANKPU AOUOD APXCP ASPBG AVWKF AXJTR AZFZN BJAXD BKOJK BLXMC CS3 DU5 EBS EFJIC EFKBS EFLBG EJD EO8 EO9 EP2 EP3 F5P FDB FEDTE FGOYB FIRID FNPLU FYGXN G-Q GBLVA GBOLZ HLZ HVGLF HZ~ H~9 IHE J1W JJJVA KOM LG9 M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 R2- ROL RPZ SBC SCC SDF SDG SDP SES SEW SPC SPCBC SST SSV SSZ T5K WH7 WUQ XPP ZMT ~G- 9DU AAYXX ACLOT CITATION ~HD |
| ID | FETCH-LOGICAL-c303t-e20f105ffa9a16f5a64f06adbacc1c1947f00f3aa04792a5f5a70a8b9259ac9e3 |
| ISICitedReferencesCount | 2 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000980984700001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0167-8191 |
| IngestDate | Sat Nov 29 07:37:03 EST 2025 Tue Nov 18 21:53:52 EST 2025 Sat Sep 06 17:19:39 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Work stealing Asynchronous Many-Tasking Work sharing Task-based parallel programming Runtime systems |
| Language | English |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c303t-e20f105ffa9a16f5a64f06adbacc1c1947f00f3aa04792a5f5a70a8b9259ac9e3 |
| ParticipantIDs | crossref_citationtrail_10_1016_j_parco_2023_103020 crossref_primary_10_1016_j_parco_2023_103020 elsevier_sciencedirect_doi_10_1016_j_parco_2023_103020 |
| PublicationCentury | 2000 |
| PublicationDate | July 2023 2023-07-00 |
| PublicationDateYYYYMMDD | 2023-07-01 |
| PublicationDate_xml | – month: 07 year: 2023 text: July 2023 |
| PublicationDecade | 2020 |
| PublicationTitle | Parallel computing |
| PublicationYear | 2023 |
| Publisher | Elsevier B.V |
| Publisher_xml | – name: Elsevier B.V |
| References | Freeman (b22) 1977; 40 Charles, Grothoff, Saraswat, Donawa, Kielstra, Ebcioglu, von Praun, Sarkar (b11) 2005; 40 Posner, Fohry (b12) 2016 Dinan, Krishnamoorthy, Larkins, Nieplocha, Sadayappan (b26) 2008 Acar, Charguéraud, Rainey (b28) 2013; 48 Reitz (b16) 2021 J. Posner, PLM-APGAS URL Olivier, Huan, Liu, Prins, Dinan, Sadayappan, Tseng (b21) 2006 Kaiser, Heller, Adelstein-Lelbach, Serio, Fey (b5) 2014 Saraswat, Almasi, Bikshandi, Cascaval, Cunningham, Grove, Kodali, Peshansky, Tardieu (b14) 2010 Lea (b20) 2000 Werner (b29) 2022 C. Fohry, An overview of task-based parallel programming models, in: Tutorial at European Network on High-performance Embedded Architecture and Compilation Conf. (HiPEAC), 2019. Dinan, Larkins, Sadayappan, Krishnamoorthy, Nieplocha (b27) 2009 TOP500.org, Goethe-HLR URL Ravicandran, Lee, Pande (b34) 2011 Pirkelbauer, Wilson, Peterson, Dechev (b8) 2019; 15 Saraswat, Kambadur, Kodali, Grove, Krishnamoorthy (b10) 2011 Augonnet, Thibault, Namyst, Wacrenier (b4) 2011; 23 Blumofe, Leiserson (b1) 1999; 46 Hoque, Herault, Bosilca, Dongarra (b3) 2017 Min, Iancu, Yelick (b36) 2011 Yamashita, Kamada (b37) 2016; 24 Kestor, Krishnamoorthy, Ma (b19) 2017 Schmaus, Pfeiffer, Schroder-Preikschat, Honig, Nolte (b9) 2021 Finnerty, Kamada, Ohta (b15) 2020 Posner, Fohry (b17) 2021 Dinan, Olivier, Sabin, Prins, Sadayappan, Tseng (b32) 2007 Zhang, Tardieu, Grove, Herta, Kamada, Saraswat, Takeuchi (b7) 2014 . Posner, Reitz, Fohry (b23) 2022; 12 Klinkenberg, Samfass, Bader, Terboven, Müller (b33) 2019; 138 Tardieu (b13) 2015 Thoman, Dichev, Heller (b30) 2018; 74 Posner, Fohry (b35) 2018 OpenMP Architecture Review Board (b2) 2021 Kulkarni, Lumsdaine (b31) 2019 Cartier, Dinan, Larkins (b18) 2021 Min (10.1016/j.parco.2023.103020_b36) 2011 Pirkelbauer (10.1016/j.parco.2023.103020_b8) 2019; 15 OpenMP Architecture Review Board (10.1016/j.parco.2023.103020_b2) 2021 Augonnet (10.1016/j.parco.2023.103020_b4) 2011; 23 Posner (10.1016/j.parco.2023.103020_b17) 2021 Kaiser (10.1016/j.parco.2023.103020_b5) 2014 Freeman (10.1016/j.parco.2023.103020_b22) 1977; 40 Blumofe (10.1016/j.parco.2023.103020_b1) 1999; 46 Acar (10.1016/j.parco.2023.103020_b28) 2013; 48 Thoman (10.1016/j.parco.2023.103020_b30) 2018; 74 Hoque (10.1016/j.parco.2023.103020_b3) 2017 Dinan (10.1016/j.parco.2023.103020_b27) 2009 Reitz (10.1016/j.parco.2023.103020_b16) 2021 Dinan (10.1016/j.parco.2023.103020_b32) 2007 Charles (10.1016/j.parco.2023.103020_b11) 2005; 40 Lea (10.1016/j.parco.2023.103020_b20) 2000 Olivier (10.1016/j.parco.2023.103020_b21) 2006 Klinkenberg (10.1016/j.parco.2023.103020_b33) 2019; 138 Werner (10.1016/j.parco.2023.103020_b29) 2022 Posner (10.1016/j.parco.2023.103020_b12) 2016 Kulkarni (10.1016/j.parco.2023.103020_b31) 2019 Saraswat (10.1016/j.parco.2023.103020_b10) 2011 Yamashita (10.1016/j.parco.2023.103020_b37) 2016; 24 Kestor (10.1016/j.parco.2023.103020_b19) 2017 10.1016/j.parco.2023.103020_b6 Finnerty (10.1016/j.parco.2023.103020_b15) 2020 Ravicandran (10.1016/j.parco.2023.103020_b34) 2011 Posner (10.1016/j.parco.2023.103020_b35) 2018 Zhang (10.1016/j.parco.2023.103020_b7) 2014 Tardieu (10.1016/j.parco.2023.103020_b13) 2015 Cartier (10.1016/j.parco.2023.103020_b18) 2021 Posner (10.1016/j.parco.2023.103020_b23) 2022; 12 10.1016/j.parco.2023.103020_b24 10.1016/j.parco.2023.103020_b25 Dinan (10.1016/j.parco.2023.103020_b26) 2008 Schmaus (10.1016/j.parco.2023.103020_b9) 2021 Saraswat (10.1016/j.parco.2023.103020_b14) 2010 |
| References_xml | – volume: 46 start-page: 720 year: 1999 end-page: 748 ident: b1 article-title: Scheduling multithreaded computations by work stealing publication-title: J. ACM – start-page: 25 year: 2015 end-page: 26 ident: b13 article-title: The APGAS library: Resilient parallel and distributed programming in java 8 publication-title: Proc. SIGPLAN Workshop on X10 – start-page: 201 year: 2011 end-page: 212 ident: b10 article-title: Lifeline-based global load balancing publication-title: Proc. SIGPLAN Symp. on Principles and Practice of Parallel Programming (PPoPP) – year: 2021 ident: b18 article-title: Optimizing work stealing communication with structured atomic operations publication-title: Proc. Int. Conf. on Parallel Processing – volume: 15 year: 2019 ident: b8 article-title: Blaze-tasks: A framework for computing parallel reductions over tasks publication-title: Trans. Archit. Code Optim. (TACO) – year: 2021 ident: b17 article-title: Transparent resource elasticity for task-based cluster environments with work stealing publication-title: Proc. Int. Conf. on Parallel Processing (ICPP) Workshops (P2S2) – volume: 48 start-page: 219 year: 2013 end-page: 228 ident: b28 article-title: Scheduling parallel programs by work stealing with private deques publication-title: SIGPLAN Not. – volume: 40 start-page: 35 year: 1977 ident: b22 article-title: A set of measures of centrality based on betweenness publication-title: Sociometry – start-page: 397 year: 2017 end-page: 408 ident: b19 article-title: Localized fault recovery for nested fork-join programs publication-title: Proc. Int. Symp. on Parallel and Distributed Processing – volume: 74 start-page: 1422 year: 2018 end-page: 1434 ident: b30 article-title: A taxonomy of task-based parallel programming technologies for high-performance computing publication-title: J. Supercomput. – start-page: 1 year: 2017 end-page: 8 ident: b3 article-title: Dynamic task discovery in PaRSEC: A data-flow task-based runtime publication-title: Proc. Int. Conf. on High Performance Computing, Networking, Storage and Analysis (SC) Workshops (ScalA) – volume: 138 year: 2019 ident: b33 article-title: CHAMELEON: Reactive load balancing for hybrid MPI+openMP task-parallel applications publication-title: J. Parallel Distrib. Comput. – volume: 23 start-page: 187 year: 2011 end-page: 198 ident: b4 article-title: StarPU: A unified platform for task scheduling on heterogeneous multicore architectures publication-title: Concurr. Comput.: Pract. Exp. (CCPE) – start-page: 1 year: 2014 end-page: 11 ident: b5 article-title: HPX: A task based programming model in a global address space publication-title: Proc. Int. Conf. on Partitioned Global Address Space Programming Models – year: 2022 ident: b29 article-title: Anwendung einer SplitQueue Datenstruktur auf Work Stealing in Laufzeitsystemen taskbasierter paralleler Programmiersysthemen – start-page: 1435 year: 2018 end-page: 1448 ident: b35 article-title: Hybrid work stealing of locality-flexible and cancelable tasks for the APGAS library publication-title: J. Supercomput. – start-page: 586 year: 2008 end-page: 593 ident: b26 article-title: Scioto: A framework for global-view task parallelism publication-title: Int. Conf. on Parallel Processing – start-page: 1 year: 2007 end-page: 8 ident: b32 article-title: Dynamic load balancing of unbalanced computations using message passing publication-title: Proc. Int. Parallel and Distributed Processing Symp. – start-page: 205 year: 2011 end-page: 217 ident: b34 article-title: Work stealing for multi-core HPC clusters publication-title: Proc. Euro-Par Parallel Processing – reference: . TOP500.org, Goethe-HLR URL – start-page: 31 year: 2014 end-page: 40 ident: b7 article-title: GLB: Lifeline-based global load balancing library in X10 publication-title: Proc. Workshop on Parallel Programming for Analytics Applications (PPAA) – volume: 24 start-page: 416 year: 2016 end-page: 424 ident: b37 article-title: Introducing a multithread and multistage mechanism for the global load balancing library of X10 publication-title: J. Inf. Process. – year: 2019 ident: b31 article-title: A comparative study of asynchronous many-tasking runtimes: Cilk, Charm++, ParalleX and AM++ – year: 2010 ident: b14 article-title: The asynchronous partitioned global address space model publication-title: Proc. SIGPLAN Workshop on Advances in Message Passing – reference: J. Posner, PLM-APGAS URL – year: 2021 ident: b2 article-title: OpenMP Application Programming Interface (Version 5.2) – year: 2020 ident: b15 article-title: Self-adjusting task granularity for global load balancer library on clusters of many-core processors publication-title: Proc. Int. Workshop on Programming Models and Applications for Multicores and Manycores – start-page: 817 year: 2021 end-page: 818 ident: b16 article-title: Load balancing policies for nested fork-join publication-title: Proc. Int. Conf. on Cluster Computing (CLUSTER), Extended Abstract – year: 2011 ident: b36 article-title: Hierarchical work stealing on manycore clusters publication-title: Proc. Int. Conf. on Partitioned Global Address Space Programming Models – start-page: 360 year: 2021 end-page: 371 ident: b9 article-title: Nowa: A wait-free continuation-stealing concurrency platform – reference: . – start-page: 235 year: 2006 end-page: 250 ident: b21 article-title: UTS: An unbalanced tree search benchmark publication-title: Languages and Compilers for Parallel Computing – volume: 12 start-page: 47 year: 2022 end-page: 72 ident: b23 article-title: Task-level resilience: Checkpointing vs. Supervision publication-title: Spec. Issue Int. J. Netw. Comput. (IJNC) – reference: C. Fohry, An overview of task-based parallel programming models, in: Tutorial at European Network on High-performance Embedded Architecture and Compilation Conf. (HiPEAC), 2019. – start-page: 13 year: 2016 end-page: 17 ident: b12 article-title: Cooperation vs. Coordination for lifeline-based global load balancing in APGAS publication-title: Proc. SIGPLAN Workshop on X10 – start-page: 36 year: 2000 end-page: 43 ident: b20 article-title: A java fork/join framework publication-title: Proc. of the Conf. on Java Grande – volume: 40 start-page: 519 year: 2005 end-page: 538 ident: b11 article-title: X10: An object-oriented approach to non-uniform cluster computing publication-title: SIGPLAN Not. – year: 2009 ident: b27 article-title: Scalable work stealing publication-title: Proc. Int. Conf. on High Performance Computing, Networking, Storage and Analysis – year: 2021 ident: 10.1016/j.parco.2023.103020_b2 – volume: 23 start-page: 187 year: 2011 ident: 10.1016/j.parco.2023.103020_b4 article-title: StarPU: A unified platform for task scheduling on heterogeneous multicore architectures publication-title: Concurr. Comput.: Pract. Exp. (CCPE) doi: 10.1002/cpe.1631 – start-page: 1 year: 2017 ident: 10.1016/j.parco.2023.103020_b3 article-title: Dynamic task discovery in PaRSEC: A data-flow task-based runtime – start-page: 36 year: 2000 ident: 10.1016/j.parco.2023.103020_b20 article-title: A java fork/join framework – volume: 48 start-page: 219 issue: 8 year: 2013 ident: 10.1016/j.parco.2023.103020_b28 article-title: Scheduling parallel programs by work stealing with private deques publication-title: SIGPLAN Not. doi: 10.1145/2517327.2442538 – year: 2019 ident: 10.1016/j.parco.2023.103020_b31 – volume: 40 start-page: 519 issue: 10 year: 2005 ident: 10.1016/j.parco.2023.103020_b11 article-title: X10: An object-oriented approach to non-uniform cluster computing publication-title: SIGPLAN Not. doi: 10.1145/1103845.1094852 – start-page: 1 year: 2007 ident: 10.1016/j.parco.2023.103020_b32 article-title: Dynamic load balancing of unbalanced computations using message passing – ident: 10.1016/j.parco.2023.103020_b25 – start-page: 1 year: 2014 ident: 10.1016/j.parco.2023.103020_b5 article-title: HPX: A task based programming model in a global address space – start-page: 235 year: 2006 ident: 10.1016/j.parco.2023.103020_b21 article-title: UTS: An unbalanced tree search benchmark – volume: 74 start-page: 1422 issue: 4 year: 2018 ident: 10.1016/j.parco.2023.103020_b30 article-title: A taxonomy of task-based parallel programming technologies for high-performance computing publication-title: J. Supercomput. doi: 10.1007/s11227-018-2238-4 – start-page: 13 year: 2016 ident: 10.1016/j.parco.2023.103020_b12 article-title: Cooperation vs. Coordination for lifeline-based global load balancing in APGAS – year: 2011 ident: 10.1016/j.parco.2023.103020_b36 article-title: Hierarchical work stealing on manycore clusters – start-page: 360 year: 2021 ident: 10.1016/j.parco.2023.103020_b9 – start-page: 205 year: 2011 ident: 10.1016/j.parco.2023.103020_b34 article-title: Work stealing for multi-core HPC clusters – volume: 15 issue: 4 year: 2019 ident: 10.1016/j.parco.2023.103020_b8 article-title: Blaze-tasks: A framework for computing parallel reductions over tasks publication-title: Trans. Archit. Code Optim. (TACO) – volume: 12 start-page: 47 issue: 1 year: 2022 ident: 10.1016/j.parco.2023.103020_b23 article-title: Task-level resilience: Checkpointing vs. Supervision publication-title: Spec. Issue Int. J. Netw. Comput. (IJNC) – start-page: 1435 year: 2018 ident: 10.1016/j.parco.2023.103020_b35 article-title: Hybrid work stealing of locality-flexible and cancelable tasks for the APGAS library publication-title: J. Supercomput. doi: 10.1007/s11227-018-2234-8 – volume: 46 start-page: 720 issue: 5 year: 1999 ident: 10.1016/j.parco.2023.103020_b1 article-title: Scheduling multithreaded computations by work stealing publication-title: J. ACM doi: 10.1145/324133.324234 – year: 2010 ident: 10.1016/j.parco.2023.103020_b14 article-title: The asynchronous partitioned global address space model – year: 2020 ident: 10.1016/j.parco.2023.103020_b15 article-title: Self-adjusting task granularity for global load balancer library on clusters of many-core processors – start-page: 397 year: 2017 ident: 10.1016/j.parco.2023.103020_b19 article-title: Localized fault recovery for nested fork-join programs – start-page: 201 year: 2011 ident: 10.1016/j.parco.2023.103020_b10 article-title: Lifeline-based global load balancing – start-page: 25 year: 2015 ident: 10.1016/j.parco.2023.103020_b13 article-title: The APGAS library: Resilient parallel and distributed programming in java 8 – year: 2021 ident: 10.1016/j.parco.2023.103020_b17 article-title: Transparent resource elasticity for task-based cluster environments with work stealing – volume: 24 start-page: 416 issue: 2 year: 2016 ident: 10.1016/j.parco.2023.103020_b37 article-title: Introducing a multithread and multistage mechanism for the global load balancing library of X10 publication-title: J. Inf. Process. – ident: 10.1016/j.parco.2023.103020_b24 – start-page: 817 year: 2021 ident: 10.1016/j.parco.2023.103020_b16 article-title: Load balancing policies for nested fork-join – start-page: 586 year: 2008 ident: 10.1016/j.parco.2023.103020_b26 article-title: Scioto: A framework for global-view task parallelism – year: 2009 ident: 10.1016/j.parco.2023.103020_b27 article-title: Scalable work stealing – year: 2022 ident: 10.1016/j.parco.2023.103020_b29 – year: 2021 ident: 10.1016/j.parco.2023.103020_b18 article-title: Optimizing work stealing communication with structured atomic operations – ident: 10.1016/j.parco.2023.103020_b6 – start-page: 31 year: 2014 ident: 10.1016/j.parco.2023.103020_b7 article-title: GLB: Lifeline-based global load balancing library in X10 – volume: 138 year: 2019 ident: 10.1016/j.parco.2023.103020_b33 article-title: CHAMELEON: Reactive load balancing for hybrid MPI+openMP task-parallel applications publication-title: J. Parallel Distrib. Comput. – volume: 40 start-page: 35 issue: 1 year: 1977 ident: 10.1016/j.parco.2023.103020_b22 article-title: A set of measures of centrality based on betweenness publication-title: Sociometry doi: 10.2307/3033543 |
| SSID | ssj0006480 |
| Score | 2.3580477 |
| Snippet | A popular approach to program scalable irregular applications is Asynchronous Many-Task (AMT) Programming. Here, programs define tasks according to task models... |
| SourceID | crossref elsevier |
| SourceType | Enrichment Source Index Database Publisher |
| StartPage | 103020 |
| SubjectTerms | Asynchronous Many-Tasking Runtime systems Task-based parallel programming Work sharing Work stealing |
| Title | Lifeline-based load balancing schemes for Asynchronous Many-Task runtimes in clusters |
| URI | https://dx.doi.org/10.1016/j.parco.2023.103020 |
| Volume | 116 |
| WOSCitedRecordID | wos000980984700001&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: 0167-8191 databaseCode: AIEXJ dateStart: 19950101 customDbUrl: isFulltext: true dateEnd: 99991231 titleUrlDefault: https://www.sciencedirect.com omitProxy: false ssIdentifier: ssj0006480 providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1La9tAEF5cp4dekvRFnmUPvbkq8kqWtEcTEpLShEAc8E2M1lqqRMjGj5D8-8ysdiUHl9AWehFGaFdm5tPsvIexr2RVZT6BN9TCC7WWXobWjyeTgKI-CDJQZthEfHWVjMfyutO5cbUwD2VcVcnjo5z9V1bjPWQ2lc7-BbubTfEG_kam4xXZjtc_YvzPQlONee7RATXplVNAHZMSGJVxHSCTqD8TZRcOF0-Voua4lAZ7iVLBG8Hivke9CEwPJ0pRL1fUSWGxrsNew5wGsJQmHX21dGcfRW3ywoyH7V0W0Io2qpPIijZ9o2ijQXNbbjOaZgW0lSjTX3Vw_6SE1cRuZT0TImiyWK27bKNkpvZgomQmK_GFCK7rLTfEee1ZuPs-QyRQpaYIqEmAL_z29GpyCm9oZ9oYjSpjWb5hWyIeyKTLtoYXp-MfzQEdhWagXvNPXDMqk_a38arfKyxrSshol21b64EPa66_Z528-sB23GQObgX1R3b7EgScQMAbEHALAo4g4Osg4A0IuAMBLyruQPCJ3Z6djk7OPTtAw1OomSy9XPga9WetQUI_0gOIQu1HMMlAqb7qyzDWvq8DAJozIGCAT-C3mWQSbWJQMg8-s241rfI9xvtayVAJqUO0wGNAOzuIBwLiKKY4rC_3mXBUSpXtLk9DTsrUpRHepYa0KZE2rUm7z741i2Z1c5XXH48c-VOrH9Z6X4p4eW3hwb8uPGTvWmAfse5yvsqP2Vv1sCwW8y8WV897-ItG |
| 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=Lifeline-based+load+balancing+schemes+for+Asynchronous+Many-Task+runtimes+in+clusters&rft.jtitle=Parallel+computing&rft.au=Reitz%2C+Mia&rft.au=Hardenbicker%2C+Kai&rft.au=Werner%2C+Tobias&rft.au=Fohry%2C+Claudia&rft.date=2023-07-01&rft.pub=Elsevier+B.V&rft.issn=0167-8191&rft.volume=116&rft_id=info:doi/10.1016%2Fj.parco.2023.103020&rft.externalDocID=S0167819123000261 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0167-8191&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0167-8191&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0167-8191&client=summon |