LogSC: Model-based one-sided communication performance estimation
One-sided communication (also known as remote memory access, or RMA) in the Message Passing Interface (MPI) is a communication interface that has been introduced in MPI-2 (1997) that enables new more efficient programming models. In MPI-3, some new more flexible and efficient primitives have been in...
Uloženo v:
| Vydáno v: | Future generation computer systems Ročník 132; s. 25 - 39 |
|---|---|
| Hlavní autoři: | , , , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
Elsevier B.V
01.07.2022
|
| Témata: | |
| ISSN: | 0167-739X, 1872-7115 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | One-sided communication (also known as remote memory access, or RMA) in the Message Passing Interface (MPI) is a communication interface that has been introduced in MPI-2 (1997) that enables new more efficient programming models. In MPI-3, some new more flexible and efficient primitives have been introduced, which makes it easier to use and more deployable.
However, compared to traditional two-sided communication, little work has been performed on analyzing one-sided communication costs, which urgently requires formal analysis. The communication performance model is a formal analysis of communication and its cost. We focus on the software model, whose core idea is that the transmission can be represented as a sequence of implicit transfers and data movements. This approach is a suitable solution for concurrent communication modeling.
We propose LogSC, which consists of the window cost, transmission cost, synchronization cost, and computational cost in atomic operations. In this paper, LogSC is used to model most of the operations in one-sided communication, including the put/get operation, atomic operation, and shared memory programming of MPI. We model and evaluate the parallel tests of IMB, collectives designed by combining MPI and the MPI shared memory (MPI+MPI), and the communication in the scalable universal matrix multiplication algorithm (SUMMA), which is a common matrix multiplication algorithm. Experiments show that our modeling has high accuracy, which makes up for the lack of existing models.
•We present a MPI performance model named LogSC for modeling one-sided communication.•The LogSC goal is to help optimize parallel applications and parallel algorithms.•LogSC can model most operation in RMA, including put/get atomic operations, and shared memory programming.•We provide the modeling of concurrent communication, including parallel algorithms and a MPI+MPI code. |
|---|---|
| AbstractList | One-sided communication (also known as remote memory access, or RMA) in the Message Passing Interface (MPI) is a communication interface that has been introduced in MPI-2 (1997) that enables new more efficient programming models. In MPI-3, some new more flexible and efficient primitives have been introduced, which makes it easier to use and more deployable.
However, compared to traditional two-sided communication, little work has been performed on analyzing one-sided communication costs, which urgently requires formal analysis. The communication performance model is a formal analysis of communication and its cost. We focus on the software model, whose core idea is that the transmission can be represented as a sequence of implicit transfers and data movements. This approach is a suitable solution for concurrent communication modeling.
We propose LogSC, which consists of the window cost, transmission cost, synchronization cost, and computational cost in atomic operations. In this paper, LogSC is used to model most of the operations in one-sided communication, including the put/get operation, atomic operation, and shared memory programming of MPI. We model and evaluate the parallel tests of IMB, collectives designed by combining MPI and the MPI shared memory (MPI+MPI), and the communication in the scalable universal matrix multiplication algorithm (SUMMA), which is a common matrix multiplication algorithm. Experiments show that our modeling has high accuracy, which makes up for the lack of existing models.
•We present a MPI performance model named LogSC for modeling one-sided communication.•The LogSC goal is to help optimize parallel applications and parallel algorithms.•LogSC can model most operation in RMA, including put/get atomic operations, and shared memory programming.•We provide the modeling of concurrent communication, including parallel algorithms and a MPI+MPI code. |
| Author | Wang, Ziheng Dong, Xiaoshe Cai, Weilin Chen, Heng Zhang, Xingjun |
| Author_xml | – sequence: 1 givenname: Ziheng orcidid: 0000-0001-5064-2376 surname: Wang fullname: Wang, Ziheng email: wzh009888@stu.xjtu.edu.cn – sequence: 2 givenname: Heng orcidid: 0000-0001-6665-1787 surname: Chen fullname: Chen, Heng email: hengchen@xjtu.edu.cn – sequence: 3 givenname: Xiaoshe surname: Dong fullname: Dong, Xiaoshe email: xsdong@xjtu.edu.cn – sequence: 4 givenname: Weilin orcidid: 0000-0002-6369-6389 surname: Cai fullname: Cai, Weilin email: withinmiaov@stu.xjtu.edu.cn – sequence: 5 givenname: Xingjun surname: Zhang fullname: Zhang, Xingjun email: xjzhang@xjtu.edu.cn |
| BookMark | eNp9UF1LxDAQDHKCd-o_8KF_IHXTpk3rg3AcfsGJDyr4FprNVnJcmyPpCf57c57PwsAuy84wMws2G_1IjF0JyAWI-nqT9_tpHygvoChySAB5wuaiUQVXQlQzNk9viquy_Thjixg3ACBUKeZsufafr6ub7Nlb2nLTRbJZEufR2bShH4b96LCbnB-zHYXeh6EbkTKKkxt-zxfstO-2kS7_5jl7v797Wz3y9cvD02q55lhU1cQRG9OWxgoEi2UPkpTpZCnLum6UbQXJou6VVRJMq6zpASswXUVIDUnssDxn8qiLwccYqNe7kCyEby1AH2rQG32sQR9q0JAAMtFujzRK3r4cBR3RUYpgXSCctPXuf4EfBbxrjg |
| Cites_doi | 10.1016/j.cpc.2010.04.018 10.1631/FITEE.1900075 10.1007/s11227-021-03759-8 10.1109/TPDS.2006.112 10.1109/TC.2007.38 10.1002/(SICI)1096-9128(199704)9:4<255::AID-CPE250>3.0.CO;2-2 10.1016/j.parco.2015.02.006 10.1109/71.920589 10.1006/jpdc.2001.1761 10.1177/1094342010391989 10.1145/240455.240477 10.1504/IJHPCN.2004.007569 10.1016/j.jpdc.2012.09.016 10.1016/S0167-8191(06)80021-9 10.1007/s00607-013-0324-2 10.1007/s11432-009-0161-2 10.1007/s11227-016-1779-7 10.1007/s11227-009-0296-3 10.1145/3284358 10.1016/j.future.2016.02.021 10.1002/cpe.3758 10.1109/TPDS.2017.2715809 10.1145/2780584 10.1007/s11390-015-1520-7 10.1109/HPCASIA.2005.75 10.1016/j.parco.2020.102669 10.1016/j.jpdc.2014.06.008 |
| ContentType | Journal Article |
| Copyright | 2022 Elsevier B.V. |
| Copyright_xml | – notice: 2022 Elsevier B.V. |
| DBID | AAYXX CITATION |
| DOI | 10.1016/j.future.2022.02.004 |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Computer Science |
| EISSN | 1872-7115 |
| EndPage | 39 |
| ExternalDocumentID | 10_1016_j_future_2022_02_004 S0167739X22000486 |
| GroupedDBID | --K --M -~X .DC .~1 0R~ 1B1 1~. 1~5 29H 4.4 457 4G. 5GY 5VS 7-5 71M 8P~ 9JN AACTN AAEDT AAEDW AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AAXUO AAYFN ABBOA ABFNM ABJNI ABMAC ABXDB ABYKQ ACDAQ ACGFS ACNNM ACRLP ACZNC ADBBV ADEZE ADJOM ADMUD AEBSH AEKER AFKWA AFTJW AGHFR AGUBO AGYEJ AHHHB AHZHX AIALX AIEXJ AIKHN AITUG AJBFU AJOXV ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AOUOD ASPBG AVWKF AXJTR AZFZN BKOJK BLXMC CS3 EBS EFJIC EFLBG EJD EO8 EO9 EP2 EP3 F5P FDB FEDTE FGOYB FIRID FNPLU FYGXN G-Q G8K 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- RIG ROL RPZ SBC SDF SDG SES SEW SPC SPCBC SSV SSZ T5K UHS WUQ XPP ZMT ~G- 9DU AATTM AAXKI AAYWO AAYXX ABDPE ABWVN ACLOT ACRPL ADNMO AEIPS AFJKZ AGQPQ AIIUN ANKPU APXCP CITATION EFKBS ~HD |
| ID | FETCH-LOGICAL-c255t-cc8b93bd1c0dc3f04e7ba43436687d91e426f7d740b97dbf0c50ba5ece8e4cac3 |
| ISICitedReferencesCount | 1 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000770655700003&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 07:22:35 EST 2025 Fri Feb 23 02:39:29 EST 2024 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Parallel performance models Performance analysis One-sided communication Hybrid programming MPI-3 |
| Language | English |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c255t-cc8b93bd1c0dc3f04e7ba43436687d91e426f7d740b97dbf0c50ba5ece8e4cac3 |
| ORCID | 0000-0002-6369-6389 0000-0001-5064-2376 0000-0001-6665-1787 |
| PageCount | 15 |
| ParticipantIDs | crossref_primary_10_1016_j_future_2022_02_004 elsevier_sciencedirect_doi_10_1016_j_future_2022_02_004 |
| PublicationCentury | 2000 |
| PublicationDate | July 2022 2022-07-00 |
| PublicationDateYYYYMMDD | 2022-07-01 |
| PublicationDate_xml | – month: 07 year: 2022 text: July 2022 |
| PublicationDecade | 2020 |
| PublicationTitle | Future generation computer systems |
| PublicationYear | 2022 |
| Publisher | Elsevier B.V |
| Publisher_xml | – name: Elsevier B.V |
| References | Frank, Agarwal, Vernon (b32) 1997 Kielmann, Bal, Verstoep (b56) 2000; vol. 1800 Goglin, Moreaud (b46) 2013; 73 Jing Chen, Linbo Zhang, Yunquan Zhang, Wei Yuan, Performance evaluation of Allgather algorithms on terascale Linux cluster with fast Ethernet, in: Eighth International Conference on High-Performance Computing in Asia-Pacific Region, HPCASIA’05, 2005, pp. 6 pp.–442. Jin, Sur, Chai, Panda (b45) 2005 Lastovetsky, Mkwawa, O’Flynn (b38) 2006 Rico-Gallego, Martín, Lastovetsky (b39) 2016; 61 Oehmen, Nieplocha (b4) 2006; 17 Hockney (b29) 1994; 20 Cui, Olsen, Jordan, Lee, Zhou, Small, Roten, Ely, Panda, Chourasia, Levesque, Day, Maechling (b5) 2010 MPI Forum (b7) 2012 Moritz, Frank (b34) 2001; 12 Culler, Karp, Patterson, Sahay, Santos, Schauser, Subramonian, von Eicken (b30) 1996; 39 Michael Woodacre, Derek Robb, Dean Roe, Karl Feind, The SGI Altix TM 3000 global shared-memory architecture (2003), SGI HPC White Papers. Smith, Bull (b26) 2001; 9 Nelson, Palmieri (b42) 2020 Valiev, Bylaska, Govind, Kowalski, Straatsma, Dam, Wang, Nieplocha, Aprà, Windus, de Jong (b3) 2010; 181 Dongarra, Beckman, Moore, Aerts, Aloisio, Andre, Barkai, Berthou, Boku, Braunschweig (b1) 2011; 25 Ino, Fujimoto, Hagihara (b18) 2001 Rico-Gallego, Martín, Manumachu, Lastovetsky (b9) 2019; 51 Chen, Zhai, Zhang, Zheng (b24) 2009; 52 Yoo, Jette, Grondona (b58) 2003; vol. 2862 Bonachea, Duell (b14) 2004; 1 Rico-Gallego, Martín (b25) 2015; 46 Blackford, Choi, Cleary, Demmel, Dhillon, Dongarra, Hammarling, Henry, Petitet, Stanley, Walker, Whaley (b62) 1996 Hoefler, Dinan, Buntinas, Balaji, Barrett, Brightwell, Gropp, Kale, Thakur (b13) 2012; vol. 7490 Tu, Fan, Zhan, Zhao (b23) 2012; 60 Kim, Lee (b36) 2001; 61 Rico-Gallego, Lastovetsky, Martín (b40) 2017; 28 Hoefler, Mehlan, Mietke, Rehm (b20) 2006 Bell, Bonachea, Cote, Duell, Hargrove, Husbands, Iancu, Welcome, Yelick (b47) 2003 Hoefler, Cerquetti, Mehlan, Mietke, Rehm (b12) 2005 Casanova, Giersch, Legrand, Quinson, Suter (b11) 2014; 74 Moritz, Frank (b33) 1998 Liao, Pang, Wang, Lu, Xie, Xia, Dong, Suo (b54) 2015; 30 Alexandrov, Ionescu, Schauser, Scheiman (b31) 1995 Gerstenberger, Besta, Hoefler (b8) 2013 Nishtala, Hargrove, Bonachea, Yelick (b49) 2009 van de Geijn, Watts (b61) 1997; 9 Dinan, Balaji, Buntinas, Goodell, Gropp, Thakur (b2) 2016; 28 Rabenseifner, Hager, Jost (b27) 2009 Zhang (b51) 2015 Dan, Lam, Hoefler, Vechev (b44) 2016 Hjelm (b41) 2016 Ibrahim, Hargrove, Iancu, Yelick (b43) 2014 Culler, Karp, Patterson, Sahay, Schauser, Santos, Subramonian, von Eicken (b17) 1993 Woodall, Shipman, Bosilca, Graham, Maccabe (b50) 2006; vol. 4192 You, Yang, Luan, Liu, Qian (b52) 2019; vol. 11416 Zhou, Gracia, Zhou, Schneider (b59) 2020; 99 Bosque, Perez (b21) 2004 Li, Chen, Liu, Yang, Gong, Gan, Li, Xu (b55) 2020; 21 Yuan, Zhang, Tang, Rao, Sun (b19) 2010 Hoefler, Dinan, Buntinas, Balaji, Barrett, Brightwell, Gropp, Kale, Thakur (b28) 2013; 95 Hasanov, Lastovetsky (b10) 2017; 73 Hoefler, Dinan, Thakur, Barrett, Balaji, Gropp, Underwood (b16) 2015; 2 Chen, Dong, Chen, Wang, Yu, Zhang (b53) 2021 Intel (b57) 2004 León, Maccabe, Brightwell (b37) 2002 Geist, Gropp, Huss-Lederman, Lumsdaine, Lusk, Saphir, Skjellum, Snir (b6) 1996; vol. 1123 Cameron, Ge, Sun (b22) 2007; 56 Hoefler, Schneider, Lumsdaine (b35) 2008 Bell, Bonachea, Nishtala, Yelick (b48) 2006 Intel (10.1016/j.future.2022.02.004_b57) 2004 Nishtala (10.1016/j.future.2022.02.004_b49) 2009 Smith (10.1016/j.future.2022.02.004_b26) 2001; 9 Zhou (10.1016/j.future.2022.02.004_b59) 2020; 99 Dongarra (10.1016/j.future.2022.02.004_b1) 2011; 25 Valiev (10.1016/j.future.2022.02.004_b3) 2010; 181 Bell (10.1016/j.future.2022.02.004_b47) 2003 Liao (10.1016/j.future.2022.02.004_b54) 2015; 30 Oehmen (10.1016/j.future.2022.02.004_b4) 2006; 17 Bonachea (10.1016/j.future.2022.02.004_b14) 2004; 1 Rico-Gallego (10.1016/j.future.2022.02.004_b40) 2017; 28 van de Geijn (10.1016/j.future.2022.02.004_b61) 1997; 9 Hasanov (10.1016/j.future.2022.02.004_b10) 2017; 73 Hoefler (10.1016/j.future.2022.02.004_b13) 2012; vol. 7490 Hoefler (10.1016/j.future.2022.02.004_b16) 2015; 2 Nelson (10.1016/j.future.2022.02.004_b42) 2020 You (10.1016/j.future.2022.02.004_b52) 2019; vol. 11416 Goglin (10.1016/j.future.2022.02.004_b46) 2013; 73 Chen (10.1016/j.future.2022.02.004_b24) 2009; 52 Hjelm (10.1016/j.future.2022.02.004_b41) 2016 Dan (10.1016/j.future.2022.02.004_b44) 2016 Culler (10.1016/j.future.2022.02.004_b30) 1996; 39 Ino (10.1016/j.future.2022.02.004_b18) 2001 Hoefler (10.1016/j.future.2022.02.004_b20) 2006 MPI Forum (10.1016/j.future.2022.02.004_b7) 2012 Lastovetsky (10.1016/j.future.2022.02.004_b38) 2006 Yoo (10.1016/j.future.2022.02.004_b58) 2003; vol. 2862 Hoefler (10.1016/j.future.2022.02.004_b12) 2005 Hoefler (10.1016/j.future.2022.02.004_b35) 2008 Blackford (10.1016/j.future.2022.02.004_b62) 1996 Dinan (10.1016/j.future.2022.02.004_b2) 2016; 28 10.1016/j.future.2022.02.004_b15 10.1016/j.future.2022.02.004_b60 Alexandrov (10.1016/j.future.2022.02.004_b31) 1995 Moritz (10.1016/j.future.2022.02.004_b33) 1998 Cameron (10.1016/j.future.2022.02.004_b22) 2007; 56 Kim (10.1016/j.future.2022.02.004_b36) 2001; 61 Rico-Gallego (10.1016/j.future.2022.02.004_b25) 2015; 46 León (10.1016/j.future.2022.02.004_b37) 2002 Casanova (10.1016/j.future.2022.02.004_b11) 2014; 74 Geist (10.1016/j.future.2022.02.004_b6) 1996; vol. 1123 Rico-Gallego (10.1016/j.future.2022.02.004_b9) 2019; 51 Hockney (10.1016/j.future.2022.02.004_b29) 1994; 20 Cui (10.1016/j.future.2022.02.004_b5) 2010 Bell (10.1016/j.future.2022.02.004_b48) 2006 Culler (10.1016/j.future.2022.02.004_b17) 1993 Kielmann (10.1016/j.future.2022.02.004_b56) 2000; vol. 1800 Yuan (10.1016/j.future.2022.02.004_b19) 2010 Zhang (10.1016/j.future.2022.02.004_b51) 2015 Li (10.1016/j.future.2022.02.004_b55) 2020; 21 Hoefler (10.1016/j.future.2022.02.004_b28) 2013; 95 Rabenseifner (10.1016/j.future.2022.02.004_b27) 2009 Jin (10.1016/j.future.2022.02.004_b45) 2005 Chen (10.1016/j.future.2022.02.004_b53) 2021 Frank (10.1016/j.future.2022.02.004_b32) 1997 Woodall (10.1016/j.future.2022.02.004_b50) 2006; vol. 4192 Bosque (10.1016/j.future.2022.02.004_b21) 2004 Tu (10.1016/j.future.2022.02.004_b23) 2012; 60 Moritz (10.1016/j.future.2022.02.004_b34) 2001; 12 Ibrahim (10.1016/j.future.2022.02.004_b43) 2014 Gerstenberger (10.1016/j.future.2022.02.004_b8) 2013 Rico-Gallego (10.1016/j.future.2022.02.004_b39) 2016; 61 |
| References_xml | – start-page: 184 year: 2005 end-page: 191 ident: b45 article-title: Limic: Support for high-performance MPI intra-node communication on linux cluster publication-title: 34th International Conference on Parallel Processing – start-page: 5 year: 1996 ident: b62 article-title: Scalapack: A portable linear algebra library for distributed memory computers - design issues and performance publication-title: Proceedings of the 1996 ACM/IEEE Conference on Supercomputing – volume: 9 start-page: 255 year: 1997 end-page: 274 ident: b61 article-title: SUMMA: scalable universal matrix multiplication algorithm publication-title: Concurr. Pract. Exp. – volume: 28 start-page: 3215 year: 2017 end-page: 3228 ident: b40 article-title: Model-based estimation of the communication cost of hybrid data-parallel applications on heterogeneous clusters publication-title: IEEE Trans. Parallel Distrib. Syst. – volume: 12 start-page: 404 year: 2001 end-page: 415 ident: b34 article-title: LoGPC: Modeling network contention in message-passing programs publication-title: IEEE Trans. Parallel Distrib. Syst. – start-page: 28 year: 2003 ident: b47 article-title: An evaluation of current high-performance networks publication-title: 17th International Parallel and Distributed Processing Symposium – volume: 61 start-page: 66 year: 2016 end-page: 82 ident: b39 article-title: Extending publication-title: Future Gener. Comput. Syst. – reference: Jing Chen, Linbo Zhang, Yunquan Zhang, Wei Yuan, Performance evaluation of Allgather algorithms on terascale Linux cluster with fast Ethernet, in: Eighth International Conference on High-Performance Computing in Asia-Pacific Region, HPCASIA’05, 2005, pp. 6 pp.–442. – start-page: 403 year: 2004 end-page: 410 ident: b21 article-title: HLogGP: a new parallel computational model for heterogeneous clusters publication-title: 4th IEEE/ACM International Symposium on Cluster Computing and the Grid – volume: 61 start-page: 1692 year: 2001 end-page: 1704 ident: b36 article-title: Measurement and prediction of communication delays in myrinet networks publication-title: J. Parallel Distrib. Comput. – volume: 20 start-page: 389 year: 1994 end-page: 398 ident: b29 article-title: The communication challenge for MPP: intel paragon and meiko CS-2 publication-title: Parallel Comput. – start-page: 184 year: 2016 end-page: 187 ident: b41 article-title: An evaluation of the one-sided performance in open MPI publication-title: Proceedings of the 23rd European MPI Users’ Group Meeting – volume: vol. 11416 start-page: 86 year: 2019 end-page: 105 ident: b52 article-title: Performance evaluation and analysis of linear algebra kernels in the prototype tianhe-3 cluster publication-title: Supercomputing Frontiers - 5th Asian Conference – start-page: 129 year: 2016 end-page: 144 ident: b44 article-title: Modeling and analysis of remote memory access programming publication-title: Proceedings of the 2016 ACM SIGPLAN International Conference on Object-Oriented Programming, Systems, Languages, and Applications – volume: 25 start-page: 3 year: 2011 end-page: 60 ident: b1 article-title: The international exascale software project roadmap publication-title: Int. J. High Perform. Comput. Appl. – reference: Michael Woodacre, Derek Robb, Dean Roe, Karl Feind, The SGI Altix TM 3000 global shared-memory architecture (2003), SGI HPC White Papers. – start-page: 427 year: 2009 end-page: 436 ident: b27 article-title: Hybrid MPI/OpenMP parallel programming on clusters of multi-core SMP nodes publication-title: Proceedings of the 17th Euromicro International Conference on Parallel, Distributed and Network-Based Processing – volume: 99 year: 2020 ident: b59 article-title: Collectives in hybrid MPI+MPI code: Design, practice and performance publication-title: Parallel Comput. – start-page: 53:1 year: 2013 end-page: 53:12 ident: b8 article-title: Enabling highly-scalable remote memory access programming with MPI-3 one sided publication-title: International Conference for High Performance Computing, Networking, Storage and Analysis – start-page: 268 year: 2010 end-page: 274 ident: b19 article-title: LogGPH: A parallel computational model with hierarchical communication awareness publication-title: 13th IEEE International Conference on Computational Science and Engineering – volume: 28 start-page: 4385 year: 2016 end-page: 4404 ident: b2 article-title: An implementation and evaluation of the MPI 3.0 one-sided communication interface publication-title: Concurr. Comput. Pract. Exp. – year: 2004 ident: b57 article-title: Intel MPI benchmarks – start-page: 95 year: 1995 end-page: 105 ident: b31 article-title: Loggp: Incorporating long messages into the logp model - one step closer towards a realistic model for parallel computation publication-title: 7th Annual ACM Symposium on Parallel Algorithms and Architectures – volume: 181 start-page: 1477 year: 2010 end-page: 1489 ident: b3 article-title: NWChem: A comprehensive and scalable open-source solution for large scale molecular simulations publication-title: Comput. Phys. Comm. – year: 2012 ident: b7 article-title: MPI: A message-passing interface standard. Version 3.0 – volume: vol. 4192 start-page: 76 year: 2006 end-page: 85 ident: b50 article-title: High performance RDMA protocols in HPC publication-title: Recent Advances in Parallel Virtual Machine and Message Passing Interface, 13th European PVM/MPI User’s Group Meeting, Bonn, Germany, September 17-20, 2006, Proceedings – start-page: 276 year: 1997 end-page: 287 ident: b32 article-title: Lopc: Modeling contention in parallel algorithms publication-title: Proceedings of the Sixth ACM SIGPLAN Symposium on Principles & Practice of Parallel Programming – year: 2021 ident: b53 article-title: Performance evaluation of convolutional neural network on tianhe-3 prototype publication-title: J. Supercomput. – year: 2006 ident: b48 article-title: Optimizing bandwidth limited problems using one-sided communication and overlap publication-title: 20th International Parallel and Distributed Processing Symposium – volume: 1 start-page: 91 year: 2004 end-page: 99 ident: b14 article-title: Problems with using MPI 1.1 and 2.0 as compilation targets for parallel language implementations publication-title: Int. J. High Perform. Comput. Netw. – volume: 39 start-page: 78 year: 1996 end-page: 85 ident: b30 article-title: Logp: A practical model of parallel computation publication-title: Commun. ACM – start-page: 254 year: 1998 end-page: 263 ident: b33 article-title: Logpc: Modeling network contention in message-passing programs publication-title: Proceedings of the 1998 ACM SIGMETRICS Joint International Conference on Measurement and Modeling of Computer Systems – volume: 74 start-page: 2899 year: 2014 end-page: 2917 ident: b11 article-title: Versatile, scalable, and accurate simulation of distributed applications and platforms publication-title: J. Parallel Distrib. Comput. – volume: 2 start-page: 9:1 year: 2015 end-page: 9:26 ident: b16 article-title: Remote memory access programming in MPI-3 publication-title: ACM Trans. Parallel Comput. – volume: 17 start-page: 740 year: 2006 end-page: 749 ident: b4 article-title: Scalablast: A scalable implementation of BLAST for high-performance data-intensive bioinformatics analysis publication-title: IEEE Trans. Parallel Distrib. Syst. – start-page: 1 year: 2010 end-page: 20 ident: b5 article-title: Scalable earthquake simulation on petascale supercomputers publication-title: Conference on High Performance Computing Networking, Storage and Analysis – start-page: 15 year: 2006 end-page: 20 ident: b38 article-title: An accurate communication model of a heterogeneous cluster based on a switch-enabled ethernet network publication-title: 12th International Conference on Parallel and Distributed Systems – start-page: 1115 year: 2014 end-page: 1125 ident: b43 article-title: An evaluation of one-sided and two-sided communication paradigms on relaxed-ordering interconnect publication-title: 2014 IEEE 28th International Parallel and Distributed Processing Symposium – volume: vol. 1800 start-page: 1176 year: 2000 end-page: 1183 ident: b56 article-title: Fast measurement of logp parameters for message passing platforms publication-title: Parallel and Distributed Processing, 15 IPDPS 2000 Workshops, Cancun, Mexico, May 1-5, 2000, Proceedings – year: 2006 ident: b20 article-title: Logfp - a model for small messages in InfiniBand publication-title: 20th International Parallel and Distributed Processing Symposium – volume: 60 start-page: 141 year: 2012 end-page: 162 ident: b23 article-title: Performance analysis and optimization of MPI collective operations on multi-core clusters publication-title: J. Supercomput. – volume: vol. 2862 start-page: 44 year: 2003 end-page: 60 ident: b58 article-title: SLURM: simple linux utility for resource management publication-title: Job Scheduling Strategies for Parallel Processing, 9th International Workshop, JSSPP 2003, Seattle, WA, USA, June 24, 2003, Revised Papers – volume: 21 start-page: 939 year: 2020 end-page: 949 ident: b55 article-title: OHTMA: an optimized heuristic topology-aware mapping algorithm on the tianhe-3 exascale supercomputer prototype publication-title: Front. Inf. Technol. Electron. Eng. – volume: 56 start-page: 314 year: 2007 end-page: 327 ident: b22 article-title: Log publication-title: IEEE Trans. Comput. – start-page: 1 year: 2009 end-page: 12 ident: b49 article-title: Scaling communication-intensive applications on BlueGene/P using one-sided communication and overlap publication-title: 23rd IEEE International Symposium on Parallel and Distributed Processing – volume: 30 start-page: 259 year: 2015 end-page: 272 ident: b54 article-title: High performance interconnect network for tianhe system publication-title: J. Comput. Sci. Technol. – start-page: 116 year: 2008 end-page: 125 ident: b35 article-title: Multistage switches are not crossbars: Effects of static routing in high-performance networks publication-title: Proceedings of the 2008 IEEE International Conference on Cluster Computing – start-page: 1 year: 2015 end-page: 23 ident: b51 article-title: Mars: A 64-core ARMv8 processor publication-title: 2015 IEEE Hot Chips 27 Symposium – volume: 51 start-page: 126:1 year: 2019 end-page: 126:36 ident: b9 article-title: A survey of communication performance models for high-performance computing publication-title: ACM Comput. Surv. – volume: 46 start-page: 14 year: 2015 end-page: 31 ident: b25 article-title: -lop: Modeling performance of shared memory MPI publication-title: Parallel Comput. – start-page: 1 year: 1993 end-page: 12 ident: b17 article-title: Logp: Towards a realistic model of parallel computation publication-title: Proceedings of the Fourth ACM SIGPLAN Symposium on Principles & Practice of Parallel Programming – volume: 95 start-page: 1121 year: 2013 end-page: 1136 ident: b28 article-title: MPI + MPI: a new hybrid approach to parallel programming with MPI plus shared memory publication-title: Computing – start-page: 288 year: 2020 end-page: 298 ident: b42 article-title: Performance evaluation of the impact of NUMA on one-sided RDMA interactions publication-title: International Symposium on Reliable Distributed Systems – volume: vol. 1123 start-page: 128 year: 1996 end-page: 135 ident: b6 article-title: MPI-2: extending the message-passing interface publication-title: Euro-Par ’96 Parallel Processing, Second International Euro-Par Conference, Lyon, France, August 26-29, 1996, Proceedings, Volume I – volume: vol. 7490 start-page: 132 year: 2012 end-page: 141 ident: b13 article-title: Leveraging mpi’s one-sided communication interface for shared-memory programming publication-title: Recent Advances in the Message Passing Interface - 19th European MPI Users’ Group Meeting – volume: 52 start-page: 1785 year: 2009 end-page: 1791 ident: b24 article-title: Loggpo: An accurate communication model for performance prediction of MPI programs publication-title: Sci. China Ser. F Inf. Sci. – volume: 73 start-page: 713 year: 2017 end-page: 725 ident: b10 article-title: Hierarchical redesign of classic MPI reduction algorithms publication-title: J. Supercomput. – volume: 9 start-page: 83 year: 2001 end-page: 98 ident: b26 article-title: Development of mixed mode MPI / OpenMP applications publication-title: Sci. Program. – start-page: 648 year: 2002 end-page: 657 ident: b37 article-title: Instrumenting logp parameters in GM: implementation and validation publication-title: 27th Annual IEEE Conference on Local Computer Networks – start-page: 133 year: 2001 end-page: 142 ident: b18 article-title: Loggps: a parallel computational model for synchronization analysis publication-title: Proceedings of the 2001 ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming – start-page: 562 year: 2005 end-page: 569 ident: b12 article-title: A practical approach to the rating of barrier algorithms using the logP model and open MPI publication-title: 34th International Conference on Parallel Processing Workshops – volume: 73 start-page: 176 year: 2013 end-page: 188 ident: b46 article-title: KNEM: a generic and scalable kernel-assisted intra-node MPI communication framework publication-title: J. Parallel Distrib. Comput. – volume: vol. 7490 start-page: 132 year: 2012 ident: 10.1016/j.future.2022.02.004_b13 article-title: Leveraging mpi’s one-sided communication interface for shared-memory programming – start-page: 427 year: 2009 ident: 10.1016/j.future.2022.02.004_b27 article-title: Hybrid MPI/OpenMP parallel programming on clusters of multi-core SMP nodes – start-page: 276 year: 1997 ident: 10.1016/j.future.2022.02.004_b32 article-title: Lopc: Modeling contention in parallel algorithms – start-page: 129 year: 2016 ident: 10.1016/j.future.2022.02.004_b44 article-title: Modeling and analysis of remote memory access programming – volume: 9 start-page: 83 issue: 2–3 year: 2001 ident: 10.1016/j.future.2022.02.004_b26 article-title: Development of mixed mode MPI / OpenMP applications publication-title: Sci. Program. – volume: vol. 1123 start-page: 128 year: 1996 ident: 10.1016/j.future.2022.02.004_b6 article-title: MPI-2: extending the message-passing interface – start-page: 116 year: 2008 ident: 10.1016/j.future.2022.02.004_b35 article-title: Multistage switches are not crossbars: Effects of static routing in high-performance networks – volume: 181 start-page: 1477 issue: 9 year: 2010 ident: 10.1016/j.future.2022.02.004_b3 article-title: NWChem: A comprehensive and scalable open-source solution for large scale molecular simulations publication-title: Comput. Phys. Comm. doi: 10.1016/j.cpc.2010.04.018 – start-page: 1 year: 2009 ident: 10.1016/j.future.2022.02.004_b49 article-title: Scaling communication-intensive applications on BlueGene/P using one-sided communication and overlap – start-page: 648 year: 2002 ident: 10.1016/j.future.2022.02.004_b37 article-title: Instrumenting logp parameters in GM: implementation and validation – volume: 21 start-page: 939 issue: 6 year: 2020 ident: 10.1016/j.future.2022.02.004_b55 article-title: OHTMA: an optimized heuristic topology-aware mapping algorithm on the tianhe-3 exascale supercomputer prototype publication-title: Front. Inf. Technol. Electron. Eng. doi: 10.1631/FITEE.1900075 – volume: vol. 4192 start-page: 76 year: 2006 ident: 10.1016/j.future.2022.02.004_b50 article-title: High performance RDMA protocols in HPC – year: 2021 ident: 10.1016/j.future.2022.02.004_b53 article-title: Performance evaluation of convolutional neural network on tianhe-3 prototype publication-title: J. Supercomput. doi: 10.1007/s11227-021-03759-8 – volume: 17 start-page: 740 issue: 8 year: 2006 ident: 10.1016/j.future.2022.02.004_b4 article-title: Scalablast: A scalable implementation of BLAST for high-performance data-intensive bioinformatics analysis publication-title: IEEE Trans. Parallel Distrib. Syst. doi: 10.1109/TPDS.2006.112 – start-page: 1115 year: 2014 ident: 10.1016/j.future.2022.02.004_b43 article-title: An evaluation of one-sided and two-sided communication paradigms on relaxed-ordering interconnect – start-page: 133 year: 2001 ident: 10.1016/j.future.2022.02.004_b18 article-title: Loggps: a parallel computational model for synchronization analysis – volume: 56 start-page: 314 issue: 3 year: 2007 ident: 10.1016/j.future.2022.02.004_b22 article-title: Lognp and log3p: Accurate analytical models of point-to-point communication in distributed systems publication-title: IEEE Trans. Comput. doi: 10.1109/TC.2007.38 – start-page: 28 year: 2003 ident: 10.1016/j.future.2022.02.004_b47 article-title: An evaluation of current high-performance networks – volume: vol. 11416 start-page: 86 year: 2019 ident: 10.1016/j.future.2022.02.004_b52 article-title: Performance evaluation and analysis of linear algebra kernels in the prototype tianhe-3 cluster – volume: 9 start-page: 255 issue: 4 year: 1997 ident: 10.1016/j.future.2022.02.004_b61 article-title: SUMMA: scalable universal matrix multiplication algorithm publication-title: Concurr. Pract. Exp. doi: 10.1002/(SICI)1096-9128(199704)9:4<255::AID-CPE250>3.0.CO;2-2 – volume: 46 start-page: 14 year: 2015 ident: 10.1016/j.future.2022.02.004_b25 article-title: τ-lop: Modeling performance of shared memory MPI publication-title: Parallel Comput. doi: 10.1016/j.parco.2015.02.006 – volume: 12 start-page: 404 issue: 4 year: 2001 ident: 10.1016/j.future.2022.02.004_b34 article-title: LoGPC: Modeling network contention in message-passing programs publication-title: IEEE Trans. Parallel Distrib. Syst. doi: 10.1109/71.920589 – volume: 61 start-page: 1692 issue: 11 year: 2001 ident: 10.1016/j.future.2022.02.004_b36 article-title: Measurement and prediction of communication delays in myrinet networks publication-title: J. Parallel Distrib. Comput. doi: 10.1006/jpdc.2001.1761 – volume: 25 start-page: 3 issue: 1 year: 2011 ident: 10.1016/j.future.2022.02.004_b1 article-title: The international exascale software project roadmap publication-title: Int. J. High Perform. Comput. Appl. doi: 10.1177/1094342010391989 – volume: 39 start-page: 78 issue: 11 year: 1996 ident: 10.1016/j.future.2022.02.004_b30 article-title: Logp: A practical model of parallel computation publication-title: Commun. ACM doi: 10.1145/240455.240477 – volume: 1 start-page: 91 issue: 1/2/3 year: 2004 ident: 10.1016/j.future.2022.02.004_b14 article-title: Problems with using MPI 1.1 and 2.0 as compilation targets for parallel language implementations publication-title: Int. J. High Perform. Comput. Netw. doi: 10.1504/IJHPCN.2004.007569 – volume: 73 start-page: 176 issue: 2 year: 2013 ident: 10.1016/j.future.2022.02.004_b46 article-title: KNEM: a generic and scalable kernel-assisted intra-node MPI communication framework publication-title: J. Parallel Distrib. Comput. doi: 10.1016/j.jpdc.2012.09.016 – start-page: 562 year: 2005 ident: 10.1016/j.future.2022.02.004_b12 article-title: A practical approach to the rating of barrier algorithms using the logP model and open MPI – start-page: 403 year: 2004 ident: 10.1016/j.future.2022.02.004_b21 article-title: HLogGP: a new parallel computational model for heterogeneous clusters – volume: 20 start-page: 389 issue: 3 year: 1994 ident: 10.1016/j.future.2022.02.004_b29 article-title: The communication challenge for MPP: intel paragon and meiko CS-2 publication-title: Parallel Comput. doi: 10.1016/S0167-8191(06)80021-9 – start-page: 184 year: 2016 ident: 10.1016/j.future.2022.02.004_b41 article-title: An evaluation of the one-sided performance in open MPI – start-page: 1 year: 2015 ident: 10.1016/j.future.2022.02.004_b51 article-title: Mars: A 64-core ARMv8 processor – volume: 95 start-page: 1121 issue: 12 year: 2013 ident: 10.1016/j.future.2022.02.004_b28 article-title: MPI + MPI: a new hybrid approach to parallel programming with MPI plus shared memory publication-title: Computing doi: 10.1007/s00607-013-0324-2 – volume: 52 start-page: 1785 issue: 10 year: 2009 ident: 10.1016/j.future.2022.02.004_b24 article-title: Loggpo: An accurate communication model for performance prediction of MPI programs publication-title: Sci. China Ser. F Inf. Sci. doi: 10.1007/s11432-009-0161-2 – start-page: 95 year: 1995 ident: 10.1016/j.future.2022.02.004_b31 article-title: Loggp: Incorporating long messages into the logp model - one step closer towards a realistic model for parallel computation – start-page: 254 year: 1998 ident: 10.1016/j.future.2022.02.004_b33 article-title: Logpc: Modeling network contention in message-passing programs – start-page: 288 year: 2020 ident: 10.1016/j.future.2022.02.004_b42 article-title: Performance evaluation of the impact of NUMA on one-sided RDMA interactions – volume: 73 start-page: 713 issue: 2 year: 2017 ident: 10.1016/j.future.2022.02.004_b10 article-title: Hierarchical redesign of classic MPI reduction algorithms publication-title: J. Supercomput. doi: 10.1007/s11227-016-1779-7 – year: 2006 ident: 10.1016/j.future.2022.02.004_b20 article-title: Logfp - a model for small messages in InfiniBand – year: 2012 ident: 10.1016/j.future.2022.02.004_b7 – volume: 60 start-page: 141 issue: 1 year: 2012 ident: 10.1016/j.future.2022.02.004_b23 article-title: Performance analysis and optimization of MPI collective operations on multi-core clusters publication-title: J. Supercomput. doi: 10.1007/s11227-009-0296-3 – volume: 51 start-page: 126:1 issue: 6 year: 2019 ident: 10.1016/j.future.2022.02.004_b9 article-title: A survey of communication performance models for high-performance computing publication-title: ACM Comput. Surv. doi: 10.1145/3284358 – volume: 61 start-page: 66 year: 2016 ident: 10.1016/j.future.2022.02.004_b39 article-title: Extending τ-lop to model concurrent MPI communications in multicore clusters publication-title: Future Gener. Comput. Syst. doi: 10.1016/j.future.2016.02.021 – volume: 28 start-page: 4385 issue: 17 year: 2016 ident: 10.1016/j.future.2022.02.004_b2 article-title: An implementation and evaluation of the MPI 3.0 one-sided communication interface publication-title: Concurr. Comput. Pract. Exp. doi: 10.1002/cpe.3758 – volume: 28 start-page: 3215 issue: 11 year: 2017 ident: 10.1016/j.future.2022.02.004_b40 article-title: Model-based estimation of the communication cost of hybrid data-parallel applications on heterogeneous clusters publication-title: IEEE Trans. Parallel Distrib. Syst. doi: 10.1109/TPDS.2017.2715809 – start-page: 5 year: 1996 ident: 10.1016/j.future.2022.02.004_b62 article-title: Scalapack: A portable linear algebra library for distributed memory computers - design issues and performance – volume: 2 start-page: 9:1 issue: 2 year: 2015 ident: 10.1016/j.future.2022.02.004_b16 article-title: Remote memory access programming in MPI-3 publication-title: ACM Trans. Parallel Comput. doi: 10.1145/2780584 – volume: 30 start-page: 259 issue: 2 year: 2015 ident: 10.1016/j.future.2022.02.004_b54 article-title: High performance interconnect network for tianhe system publication-title: J. Comput. Sci. Technol. doi: 10.1007/s11390-015-1520-7 – ident: 10.1016/j.future.2022.02.004_b60 doi: 10.1109/HPCASIA.2005.75 – year: 2006 ident: 10.1016/j.future.2022.02.004_b48 article-title: Optimizing bandwidth limited problems using one-sided communication and overlap – start-page: 1 year: 2010 ident: 10.1016/j.future.2022.02.004_b5 article-title: Scalable earthquake simulation on petascale supercomputers – start-page: 1 year: 1993 ident: 10.1016/j.future.2022.02.004_b17 article-title: Logp: Towards a realistic model of parallel computation – year: 2004 ident: 10.1016/j.future.2022.02.004_b57 – volume: vol. 2862 start-page: 44 year: 2003 ident: 10.1016/j.future.2022.02.004_b58 article-title: SLURM: simple linux utility for resource management – start-page: 15 year: 2006 ident: 10.1016/j.future.2022.02.004_b38 article-title: An accurate communication model of a heterogeneous cluster based on a switch-enabled ethernet network – ident: 10.1016/j.future.2022.02.004_b15 – start-page: 184 year: 2005 ident: 10.1016/j.future.2022.02.004_b45 article-title: Limic: Support for high-performance MPI intra-node communication on linux cluster – start-page: 53:1 year: 2013 ident: 10.1016/j.future.2022.02.004_b8 article-title: Enabling highly-scalable remote memory access programming with MPI-3 one sided – volume: 99 year: 2020 ident: 10.1016/j.future.2022.02.004_b59 article-title: Collectives in hybrid MPI+MPI code: Design, practice and performance publication-title: Parallel Comput. doi: 10.1016/j.parco.2020.102669 – volume: vol. 1800 start-page: 1176 year: 2000 ident: 10.1016/j.future.2022.02.004_b56 article-title: Fast measurement of logp parameters for message passing platforms – start-page: 268 year: 2010 ident: 10.1016/j.future.2022.02.004_b19 article-title: LogGPH: A parallel computational model with hierarchical communication awareness – volume: 74 start-page: 2899 issue: 10 year: 2014 ident: 10.1016/j.future.2022.02.004_b11 article-title: Versatile, scalable, and accurate simulation of distributed applications and platforms publication-title: J. Parallel Distrib. Comput. doi: 10.1016/j.jpdc.2014.06.008 |
| SSID | ssj0001731 |
| Score | 2.3449538 |
| Snippet | One-sided communication (also known as remote memory access, or RMA) in the Message Passing Interface (MPI) is a communication interface that has been... |
| SourceID | crossref elsevier |
| SourceType | Index Database Publisher |
| StartPage | 25 |
| SubjectTerms | Hybrid programming MPI-3 One-sided communication Parallel performance models Performance analysis |
| Title | LogSC: Model-based one-sided communication performance estimation |
| URI | https://dx.doi.org/10.1016/j.future.2022.02.004 |
| Volume | 132 |
| WOSCitedRecordID | wos000770655700003&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: ScienceDirect Freedom Collection 2021 customDbUrl: eissn: 1872-7115 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0001731 issn: 0167-739X databaseCode: AIEXJ dateStart: 19950201 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3dS9xAEF_O0wdfaquW-tGSB99kZZNLbnf7dpwWW4oIWgx9CdnNXD255kSv6p_f2ezko1wptlAIISzkbpj5Zb6YD8YOoqJyJIAnBlVgHIHmubEhD3VijVEgB9ovm5BnZypN9Xmv91j3wjzMZFmqpyd9-19FjWcobNc6-xfibn4UD_AZhY53FDvenyX4z_NvF2MX6Ls1ZzPuzFRxOC-Bu8WcVQ9b2xHiphY3fQNu3sb3VlD17s5q6IjbtAwEFkuLIGgKdOOUX1Hq-ev0GsgeVnUDXrGdds6OqQ44nebz--sGWmO_GvsKpjOaB07piKgtXaUc2VKfjE9bojpGsafe6nhVqyT69qFv5mx0MSU7SZsmHbvsZx4taXyffLg58iNYjhxJfghr3Fq4pu7wwhHi6Iiiqpt-uMJWI5lo1Wero48n6afGiIeSVlkS4XXXZVUauPxfv_dqOp7K5Uv2gkKMYOSh8Yr1oNxkG_X6joC0-RYbVUh5H3RwEjQ4CX7BSdDBSdDiZJt9-XByOT7ltFGDWwwdF9xaZfTAFKEVhR1MRAzS5K63eDhUstAhoL82kYWMhdGyMBNhE2HyBCwoiG1uB69Zv0RK3rBgmNsQFYCIDQYQgN-4koDRtjBxgvwwsMN4zZHs1g9OyeqKwpvMczBzHMwEXiLeYbJmW0bOn3fqMpT0H9_c_ec399h6C-F91l_c_YC3bM0-LKb3d-8IEj8BSGWB2A |
| 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=LogSC%3A+Model-based+one-sided+communication+performance+estimation&rft.jtitle=Future+generation+computer+systems&rft.au=Wang%2C+Ziheng&rft.au=Chen%2C+Heng&rft.au=Dong%2C+Xiaoshe&rft.au=Cai%2C+Weilin&rft.date=2022-07-01&rft.pub=Elsevier+B.V&rft.issn=0167-739X&rft.eissn=1872-7115&rft.volume=132&rft.spage=25&rft.epage=39&rft_id=info:doi/10.1016%2Fj.future.2022.02.004&rft.externalDocID=S0167739X22000486 |
| 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 |