Performance Characterization of Python Runtimes for Multi-device Task Parallel Programming

Modern Python programs in high-performance computing call into compiled libraries and kernels for performance-critical tasks. However, effectively parallelizing these finer-grained, and often dynamic, kernels across modern heterogeneous platforms remains a challenge. This paper designs and optimizes...

Full description

Saved in:

Bibliographic Details
Published in:	International journal of parallel programming Vol. 53; no. 2; p. 16
Main Authors:	Ruys, William, Lee, Hochan, You, Bozhi, Talati, Shreya, Park, Jaeyoung, Almgren-Bell, James, Yan, Yineng, Fernando, Milinda, Erez, Mattan, Gligoric, Milos, Burtscher, Martin, Rossbach, Christopher J., Pingali, Keshav, Biros, George
Format:	Journal Article
Language:	English
Published:	New York Springer US 01.04.2025 Springer Nature B.V
Subjects:	Computer Science Graphs Interpreters Libraries Parallel programming Processor Architectures Python Run time (computers) Software Engineering/Programming and Operating Systems Theory of Computation Workloads HPC in Python Task parallel programming GPU tasking systems GPU programming in Python Global Interpreter Lock
ISSN:	0885-7458, 1573-7640
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Abstract	Modern Python programs in high-performance computing call into compiled libraries and kernels for performance-critical tasks. However, effectively parallelizing these finer-grained, and often dynamic, kernels across modern heterogeneous platforms remains a challenge. This paper designs and optimizes a multi-threaded runtime for Python tasks on single-node multi-GPU systems, including tasks that use resources across multiple devices. We perform an experimental study which examines the impact of Python’s Global Interpreter Lock (GIL) on runtime performance and the potential gains under a GIL-less PEP703 future. This work explores tasks with variants for different different device sets, introducing new programming abstractions and runtime mechanisms to simplify their management and enhance portability. Our experimental analysis, using tasks graphs from synthetic and real applications, shows at least a 3 × (and up to 6 × ) performance improvement over its predecessor in scenarios with high GIL contention. Our implementation of multi-device tasks achieves 8 × less overhead per task relative to a multi-process alternative using Ray.
AbstractList	Modern Python programs in high-performance computing call into compiled libraries and kernels for performance-critical tasks. However, effectively parallelizing these finer-grained, and often dynamic, kernels across modern heterogeneous platforms remains a challenge. This paper designs and optimizes a multi-threaded runtime for Python tasks on single-node multi-GPU systems, including tasks that use resources across multiple devices. We perform an experimental study which examines the impact of Python’s Global Interpreter Lock (GIL) on runtime performance and the potential gains under a GIL-less PEP703 future. This work explores tasks with variants for different different device sets, introducing new programming abstractions and runtime mechanisms to simplify their management and enhance portability. Our experimental analysis, using tasks graphs from synthetic and real applications, shows at least a 3× (and up to 6×) performance improvement over its predecessor in scenarios with high GIL contention. Our implementation of multi-device tasks achieves 8× less overhead per task relative to a multi-process alternative using Ray. Modern Python programs in high-performance computing call into compiled libraries and kernels for performance-critical tasks. However, effectively parallelizing these finer-grained, and often dynamic, kernels across modern heterogeneous platforms remains a challenge. This paper designs and optimizes a multi-threaded runtime for Python tasks on single-node multi-GPU systems, including tasks that use resources across multiple devices. We perform an experimental study which examines the impact of Python’s Global Interpreter Lock (GIL) on runtime performance and the potential gains under a GIL-less PEP703 future. This work explores tasks with variants for different different device sets, introducing new programming abstractions and runtime mechanisms to simplify their management and enhance portability. Our experimental analysis, using tasks graphs from synthetic and real applications, shows at least a 3 × (and up to 6 × ) performance improvement over its predecessor in scenarios with high GIL contention. Our implementation of multi-device tasks achieves 8 × less overhead per task relative to a multi-process alternative using Ray.
ArticleNumber	16
Author	Gligoric, Milos Rossbach, Christopher J. Ruys, William Yan, Yineng Fernando, Milinda Erez, Mattan Talati, Shreya Park, Jaeyoung Lee, Hochan You, Bozhi Biros, George Pingali, Keshav Almgren-Bell, James Burtscher, Martin
Author_xml	– sequence: 1 givenname: William surname: Ruys fullname: Ruys, William email: will@oden.utexas.edu organization: The University of Texas at Austin – sequence: 2 givenname: Hochan surname: Lee fullname: Lee, Hochan organization: The University of Texas at Austin – sequence: 3 givenname: Bozhi surname: You fullname: You, Bozhi organization: The University of Texas at Austin – sequence: 4 givenname: Shreya surname: Talati fullname: Talati, Shreya organization: The University of Texas at Austin – sequence: 5 givenname: Jaeyoung surname: Park fullname: Park, Jaeyoung organization: The University of Texas at Austin – sequence: 6 givenname: James surname: Almgren-Bell fullname: Almgren-Bell, James organization: The University of Texas at Austin – sequence: 7 givenname: Yineng surname: Yan fullname: Yan, Yineng organization: The University of Texas at Austin – sequence: 8 givenname: Milinda surname: Fernando fullname: Fernando, Milinda organization: The University of Texas at Austin – sequence: 9 givenname: Mattan surname: Erez fullname: Erez, Mattan organization: The University of Texas at Austin – sequence: 10 givenname: Milos surname: Gligoric fullname: Gligoric, Milos organization: The University of Texas at Austin – sequence: 11 givenname: Martin surname: Burtscher fullname: Burtscher, Martin organization: Texas State University – sequence: 12 givenname: Christopher J. surname: Rossbach fullname: Rossbach, Christopher J. organization: The University of Texas at Austin – sequence: 13 givenname: Keshav surname: Pingali fullname: Pingali, Keshav organization: The University of Texas at Austin – sequence: 14 givenname: George surname: Biros fullname: Biros, George organization: The University of Texas at Austin
BookMark	eNp9kN9LwzAQx4NMcJv-Az4VfI4mTdJkjzL8BROLzBdfQtpets42nUkrzL_eaAXfhIO7g8_nDr4zNHGdA4TOKbmkhMirQInMMkxSgeOqFKZHaEqFZFhmnEzQlCglsORCnaBZCDtCyCJiU_Sag7edb40rIVlujTdlD77-NH3duaSzSX7ot3F6HlxftxCSCCePQ9PXuIKPOkprE96SPIpNA02S-27jTdvWbnOKjq1pApz99jl6ub1ZL-_x6unuYXm9wmVKSI8rwSspIVWZNYKJQgJXmSIZAAPDWVFYWxVVuWC0NJZaZkFwICKWLJThhs3RxXh377v3AUKvd93gXXypGZWKM56RRaTSkSp9F4IHq_e-bo0_aEr0d4Z6zFDHDPVPhppGiY1SiLDbgP87_Y_1BYrHeAI
Cites_doi	10.1109/SC41405.2020.00066 10.1002/cpe.1631 10.1109/SC41404.2022.00056 10.1145/3332186.3332231 10.1109/SBAC-PAD55451.2022.00025 10.1038/s41592-019-0686-2 10.1016/j.parco.2009.12.005 10.1145/3311790.3396656 10.1109/MCSE.2011.37 10.1109/CLUSTERWorkshops61457.2023.00028 10.1109/ICPADS56603.2022.00117 10.1145/3447818.3460376 10.1145/3307681.3325401 10.1177/1094342015594678 10.1177/1094342006064503 10.1145/2818311 10.1109/CCGrid49817.2020.00-66 10.1145/3295500.3356175 10.1016/j.jpdc.2017.06.009 10.1109/MCSE.2013.98 10.1109/IPDPS53621.2022.00073 10.1007/BFb0053987 10.1109/PAW-ATM49560.2019.00011 10.1007/978-0-387-09766-4_54 10.1109/SC.2012.71
ContentType	Journal Article
Copyright	The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025 Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law. Copyright Springer Nature B.V. Apr 2025
Copyright_xml	– notice: The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025 Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law. – notice: Copyright Springer Nature B.V. Apr 2025
DBID	AAYXX CITATION 7SC 8FD JQ2 L7M L~C L~D
DOI	10.1007/s10766-025-00788-1
DatabaseName	CrossRef Computer and Information Systems Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional
DatabaseTitle	CrossRef Computer and Information Systems Abstracts Technology Research Database Computer and Information Systems Abstracts – Academic Advanced Technologies Database with Aerospace ProQuest Computer Science Collection Computer and Information Systems Abstracts Professional
DatabaseTitleList	Computer and Information Systems Abstracts
DeliveryMethod	fulltext_linktorsrc
Discipline	Computer Science
EISSN	1573-7640
ExternalDocumentID	10_1007_s10766_025_00788_1
GrantInformation_xml	– fundername: U.S. Department of Energy grantid: DE-NA0003969; DE-NA0003969; DE-NA0003969; DE-NA0003969; DE-NA0003969; DE-NA0003969; DE-NA0003969; DE-NA0003969; DE-NA0003969; DE-NA0003969; DE-NA0003969; DE-NA0003969; DE-NA0003969
GroupedDBID	-Y2 -~X .4S .86 .DC .VR 06D 0R~ 0VY 199 1N0 2.D 203 28- 29J 2J2 2JN 2JY 2KG 2LR 2P1 2VQ 2~H 30V 4.4 406 408 409 40D 40E 5GY 5QI 5VS 67Z 6NX 78A 7WY 8FE 8FG 8FL 8G5 8TC 8UJ 95- 95. 95~ 96X AABHQ AACDK AAHNG AAIAL AAJBT AAJKR AANZL AAOBN AAPKM AARHV AARTL AASML AATNV AATVU AAUYE AAWCG AAYIU AAYJJ AAYQN AAYTO AAYZH ABAKF ABBBX ABBXA ABDBE ABDBF ABDPE ABDZT ABECU ABFSI ABFTD ABFTV ABHLI ABHQN ABJNI ABJOX ABKCH ABKTR ABMNI ABMQK ABNWP ABQBU ABQSL ABSXP ABTAH ABTEG ABTHY ABTKH ABTMW ABULA ABUWG ABWNU ABXPI ACAOD ACBXY ACDTI ACGFO ACGFS ACHSB ACHXU ACIHN ACKNC ACMDZ ACMLO ACNCT ACOKC ACOMO ACPIV ACREN ACUHS ACZOJ ADHIR ADHKG ADKNI ADKPE ADMLS ADRFC ADTPH ADURQ ADYFF ADYOE ADZKW AEAQA AEBTG AEFIE AEFQL AEGAL AEGNC AEJHL AEJRE AEKMD AEMSY AENEX AEOHA AEPYU AESKC AETLH AEVLU AEXYK AFBBN AFDZB AFEXP AFGCZ AFKRA AFLOW AFQWF AFWTZ AFYQB AFZKB AGAYW AGDGC AGGDS AGJBK AGMZJ AGQEE AGQMX AGRTI AGWIL AGWZB AGYKE AHAVH AHBYD AHKAY AHPBZ AHSBF AHYZX AIAKS AIGIU AIIXL AILAN AITGF AJBLW AJRNO AJZVZ ALMA_UNASSIGNED_HOLDINGS ALWAN AMKLP AMTXH AMXSW AMYLF AOCGG ARAPS ARCSS ARMRJ AXYYD AYFIA AYJHY AZFZN AZQEC B-. B0M BA0 BBWZM BDATZ BENPR BEZIV BGLVJ BGNMA BKOMP BPHCQ BSONS CAG CCPQU COF CS3 CSCUP DDRTE DL5 DNIVK DPUIP DU5 DWQXO E.L EAD EAP EAS EBLON EBS EDO EIOEI EJD EMK EPL ESBYG ESX FEDTE FERAY FFXSO FIGPU FINBP FNLPD FRNLG FRRFC FSGXE FWDCC GGCAI GGRSB GJIRD GNUQQ GNWQR GQ7 GQ8 GROUPED_ABI_INFORM_RESEARCH GUQSH GXS H13 HCIFZ HF~ HG5 HG6 HMJXF HQYDN HRMNR HVGLF HZ~ H~9 I-F I09 IHE IJ- IKXTQ ITM IWAJR IXC IZIGR IZQ I~X I~Z J-C J0Z JBSCW JCJTX JZLTJ K60 K6V K6~ K7- KDC KOV KOW LAK LLZTM M0C M2O M4Y MA- MS~ N2Q NB0 NDZJH NPVJJ NQJWS NU0 O9- O93 O9G O9I O9J OAM OVD P19 P62 P9O PF0 PHGZT PQBIZ PQBZA PQQKQ PROAC PT4 PT5 Q2X QOK QOS R89 R9I RHV RNI RNS ROL RPX RSV RZC RZE RZK S16 S1Z S26 S27 S28 S3B SAP SCJ SCLPG SCO SDH SDM SHX SISQX SJYHP SNE SNPRN SNX SOHCF SOJ SPISZ SRMVM SSLCW STPWE SZN T13 T16 TAE TEORI TN5 TSG TSK TSV TUC TUS U2A U5U UG4 UOJIU UTJUX UZXMN VC2 VFIZW W23 W48 WH7 WK8 YLTOR Z45 ZMTXR ZY4 ~8M ~EX AAYXX ABBRH ABFSG ABRTQ ACSTC AEZWR AFFHD AFHIU AFOHR AGQPQ AHWEU AIXLP ATHPR CITATION PHGZM PQGLB 7SC 8FD JQ2 L7M L~C L~D
ID	FETCH-LOGICAL-c200t-d54d77e286fa535b7e486806ee3ea43bbffdbdc931caf1f3fe54e05e057b8a4a3
IEDL.DBID	RSV
ISICitedReferencesCount	0
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001449019300001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	0885-7458
IngestDate	Wed Nov 05 07:03:30 EST 2025 Sat Nov 29 08:06:45 EST 2025 Sat Apr 05 01:12:49 EDT 2025
IsPeerReviewed	true
IsScholarly	true
Issue	2
Keywords	HPC in Python Task parallel programming GPU tasking systems GPU programming in Python Global Interpreter Lock
Language	English
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c200t-d54d77e286fa535b7e486806ee3ea43bbffdbdc931caf1f3fe54e05e057b8a4a3
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
PQID	3178434609
PQPubID	48389
ParticipantIDs	proquest_journals_3178434609 crossref_primary_10_1007_s10766_025_00788_1 springer_journals_10_1007_s10766_025_00788_1
PublicationCentury	2000
PublicationDate	20250400 2025-04-00 20250401
PublicationDateYYYYMMDD	2025-04-01
PublicationDate_xml	– month: 4 year: 2025 text: 20250400
PublicationDecade	2020
PublicationPlace	New York
PublicationPlace_xml	– name: New York
PublicationTitle	International journal of parallel programming
PublicationTitleAbbrev	Int J Parallel Prog
PublicationYear	2025
Publisher	Springer US Springer Nature B.V
Publisher_xml	– name: Springer US – name: Springer Nature B.V
References	E Tejedor (788_CR17) 2017; 31 A Reuther (788_CR32) 2018; 111 788_CR22 788_CR21 788_CR43 788_CR20 788_CR42 J Nieplocha (788_CR40) 2006; 20 788_CR41 788_CR6 788_CR5 788_CR29 788_CR4 S Tomov (788_CR14) 2010; 36 788_CR28 S van der Walt (788_CR1) 2011 788_CR3 788_CR27 788_CR26 788_CR25 788_CR24 788_CR23 788_CR9 788_CR7 G Bosilca (788_CR8) 2013; 15 788_CR11 788_CR33 788_CR10 788_CR31 788_CR30 P Virtanen (788_CR2) 2020; 17 788_CR19 788_CR18 A Abdelfattah (788_CR15) 2016 788_CR39 788_CR16 788_CR38 788_CR37 788_CR36 788_CR13 788_CR35 788_CR12 788_CR34
References_xml	– ident: 788_CR19 doi: 10.1109/SC41405.2020.00066 – ident: 788_CR7 doi: 10.1002/cpe.1631 – ident: 788_CR18 doi: 10.1109/SC41404.2022.00056 – ident: 788_CR13 – ident: 788_CR11 – ident: 788_CR37 doi: 10.1145/3332186.3332231 – ident: 788_CR38 – ident: 788_CR29 doi: 10.1109/SBAC-PAD55451.2022.00025 – volume: 17 start-page: 261 issue: 3 year: 2020 ident: 788_CR2 publication-title: Nat. Methods doi: 10.1038/s41592-019-0686-2 – ident: 788_CR36 – ident: 788_CR16 – volume: 36 start-page: 232 issue: 5–6 year: 2010 ident: 788_CR14 publication-title: Parallel Comput. doi: 10.1016/j.parco.2009.12.005 – ident: 788_CR20 – ident: 788_CR27 doi: 10.1145/3311790.3396656 – year: 2011 ident: 788_CR1 publication-title: Comput. Sci. Eng. doi: 10.1109/MCSE.2011.37 – ident: 788_CR24 – ident: 788_CR41 doi: 10.1109/CLUSTERWorkshops61457.2023.00028 – ident: 788_CR35 doi: 10.1109/ICPADS56603.2022.00117 – ident: 788_CR5 doi: 10.1145/3447818.3460376 – ident: 788_CR30 doi: 10.1145/3307681.3325401 – volume: 31 start-page: 66 issue: 1 year: 2017 ident: 788_CR17 publication-title: Int. J. High Perf. Comput. Appl. doi: 10.1177/1094342015594678 – ident: 788_CR26 – volume: 20 start-page: 203 issue: 2 year: 2006 ident: 788_CR40 publication-title: Int. J. High Perf. Comput. Appl. doi: 10.1177/1094342006064503 – ident: 788_CR3 – ident: 788_CR28 – ident: 788_CR31 – year: 2016 ident: 788_CR15 publication-title: ACM Trans. Math. Softw. doi: 10.1145/2818311 – ident: 788_CR34 doi: 10.1109/CCGrid49817.2020.00-66 – ident: 788_CR10 – ident: 788_CR43 doi: 10.1145/3295500.3356175 – ident: 788_CR12 – volume: 111 start-page: 76 year: 2018 ident: 788_CR32 publication-title: J. Parallel Distrib. Comput. doi: 10.1016/j.jpdc.2017.06.009 – volume: 15 start-page: 36 issue: 6 year: 2013 ident: 788_CR8 publication-title: Comput. Sci. Eng. doi: 10.1109/MCSE.2013.98 – ident: 788_CR22 doi: 10.1109/IPDPS53621.2022.00073 – ident: 788_CR33 doi: 10.1007/BFb0053987 – ident: 788_CR42 doi: 10.1109/PAW-ATM49560.2019.00011 – ident: 788_CR39 doi: 10.1007/978-0-387-09766-4_54 – ident: 788_CR9 doi: 10.1109/SC.2012.71 – ident: 788_CR23 – ident: 788_CR21 – ident: 788_CR6 – ident: 788_CR25 – ident: 788_CR4
SSID	ssj0009788
Score	2.3468256
Snippet	Modern Python programs in high-performance computing call into compiled libraries and kernels for performance-critical tasks. However, effectively...
SourceID	proquest crossref springer
SourceType	Aggregation Database Index Database Publisher
StartPage	16
SubjectTerms	Computer Science Graphs Interpreters Libraries Parallel programming Processor Architectures Python Run time (computers) Software Engineering/Programming and Operating Systems Theory of Computation Workloads
Title	Performance Characterization of Python Runtimes for Multi-device Task Parallel Programming
URI	https://link.springer.com/article/10.1007/s10766-025-00788-1 https://www.proquest.com/docview/3178434609
Volume	53
WOSCitedRecordID	wos001449019300001&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: PRVAVX databaseName: SpringerLINK Contemporary 1997-Present customDbUrl: eissn: 1573-7640 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0009788 issn: 0885-7458 databaseCode: RSV dateStart: 19970101 isFulltext: true titleUrlDefault: https://link.springer.com/search?facet-content-type=%22Journal%22 providerName: Springer Nature
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1LSwMxEA5aPXixPrFaJQdvGug2zz1KsXiQstRaipcl2U1A7EO6VfDfO9mHq6IHhb1tEsI8Ml_IfDMInRsBegW7JcrokDCZKKJFF2xZAzYPlNYsp0ePb-VgoCaTMCpJYVmV7V49SeYn9SeymxQ-YZYTH9fg7rOONiDcKd-wYXg3rkvtyrzbJLgPJ5JxVVJlfl7jaziqMea3Z9E82vSb_9vnDtou0SW-KsxhF63Z-R5qVp0bcOnI--ghqvkCuPdRs7mgZOKFw9GbrymAh76RxMxmGAbjnKtLUuvPFjzS2ROOYOJ0aqc4KrK8ZrDPA3Tfvx71bkjZZYEk4CErknKWSmm7SjjNKTfSMiVUR1hLrWbUGOdSkyYhDRLtAked5cx2OHzSKM00PUSN-WJujxC2ynAqhE1MGDK4KWqnAD2YIJEAQhxNW-iiEnb8XBTTiOuyyV5sMYgtzsUWBy3UrvQRl46VxQB3FKNMdMIWuqzkX__-fbXjvw0_QVvdXIU-R6eNGqvliz1Fm8nr6jFbnuUG9w580NBc
linkProvider	Springer Nature
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1LS8QwEB50FfTiW1yfOXjTwHbz7FEWRXFdiq4iXkrSJiDurmJXwX_vtNtaFT0o9NYkhHlkJmS-bwD2rUS9ot1SbU1IuUo0NbKNtmwwNw-0MbyAR990Va-nb2_DqASFZVW1e_UkWZzUn8BuSuYFs4LmcQ3vPtMwwzFi5Yz5l1c3NdWuKrpNovsIqrjQJVTm5zW-hqM6x_z2LFpEm5PF_-1zCRbK7JIcTcxhGabcaAUWq84NpHTkVbiLarwA6XxwNk8gmeTRk-gt5xQgl3kjiaHLCA4mBVaXpi4_W0jfZA8kwomDgRuQaFLlNcR9rsH1yXG_c0rLLgs0QQ8Z01TwVCnX1tIbwYRVjmupW9I55gxn1nqf2jQJWZAYH3jmneCuJfBTVhtu2Do0Ro8jtwHEaSuYlC6xYcjxpmi8xuzBBonCJMSztAkHlbDjpwmZRlzTJudii1FscSG2OGjCdqWPuHSsLMZ0R3PGZStswmEl__r376tt_m34Hsyd9i-6cfesd74F8-1CnXm9zjY0xs8vbgdmk9fxffa8WxjfO0rd00A
linkToPdf	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1JS8QwFH7oKOLFXRzXHLxpcDpZexR1UJShuAyDl5K0CYizMTMK_nuTLlZFDyL01jSEt_S9R973PYBDzZ1end1iqVWIqUgkVrzpbFm53DyQStEMHt25Ee227HbD6BOKP-t2L68kc0yDZ2kaTE9GqT35BHwT3DfPMuxjnKuDZmGO-kZ6X6_fdSraXZFNnnSuxLCgTBawmZ_3-Bqaqnzz2xVpFnlay_8_8wosFVknOs3NZBVmzGANlsuJDqhw8HV4jCocATr74HLOoZpoaFH05rkG0K0fMNE3E-QWowzDi1Pj_znoXk2eUeQ-7PVMD0V591ffnXkDHloX92eXuJi-gBPnOVOcMpoKYZqSW8UI08JQyWWDG0OMokRra1OdJiEJEmUDS6xh1DSYe4SWiiqyCbXBcGC2ABmpGeHcJDoMqasglZUuq9BBIlxyYklah6NS8PEoJ9mIKzplL7bYiS3OxBYHddgtdRMXDjeJXRokKaG8EdbhuNRF9fr33bb_tvwAFqLzVnxz1b7egcVmpk3fxrMLten4xezBfPI6fZqM9zM7fAcuC9wk
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=Performance+Characterization+of+Python+Runtimes+for+Multi-device+Task+Parallel+Programming&rft.jtitle=International+journal+of+parallel+programming&rft.date=2025-04-01&rft.pub=Springer+Nature+B.V&rft.issn=0885-7458&rft.eissn=1573-7640&rft.volume=53&rft.issue=2&rft.spage=16&rft_id=info:doi/10.1007%2Fs10766-025-00788-1&rft.externalDBID=HAS_PDF_LINK
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0885-7458&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0885-7458&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0885-7458&client=summon