An approach for dynamically adaptable SIMD vectorization of FEM kernels
The paper focuses on the optimization of the FEM matrix kernels with respect to user-defined parameters such as materials, initial conditions, and boundary conditions that are known during run-time only. Adapting the kernels to specific parameters can save a significant amount of execution time and...
Uložené v:
| Vydané v: | Computer physics communications Ročník 304; s. 109319 |
|---|---|
| Hlavní autori: | , , , |
| Médium: | Journal Article |
| Jazyk: | English |
| Vydavateľské údaje: |
Elsevier B.V
01.11.2024
|
| Predmet: | |
| ISSN: | 0010-4655 |
| On-line prístup: | Získať plný text |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Abstract | The paper focuses on the optimization of the FEM matrix kernels with respect to user-defined parameters such as materials, initial conditions, and boundary conditions that are known during run-time only. Adapting the kernels to specific parameters can save a significant amount of execution time and increase performance. Handling them efficiently is challenging due to the exponential number of potential combinations that the user can specify.
The paper presents an approach that combines (a) cross-element vectorization for the easy-to-write transformation of the original scalar code to vectorized one, (b) meta-programming for utilization of a compiler for building sub-kernels tailored for a particular set of parameters, (c) and dynamic polymorphism allowing run-time selection of sub-kernels.
We show that the above techniques allow (1) straightforward code modifications, (2) efficient handling of required dynamic behavior with a minor performance penalty for most kernels, and (3) achieving up to 8-fold speedups compared to non-adapted kernels.
•High productivity approach for developing and implementing new FEM matrix kernels.•Automated approach that reaches the performance of hand-tuned kernels.•Technique allowing implementation of highly optimized kernels with standard C++ language features.•Approach combining SIMD vectorization with expression optimization with significantly higher impact on performance than pure vectorization. |
|---|---|
| AbstractList | The paper focuses on the optimization of the FEM matrix kernels with respect to user-defined parameters such as materials, initial conditions, and boundary conditions that are known during run-time only. Adapting the kernels to specific parameters can save a significant amount of execution time and increase performance. Handling them efficiently is challenging due to the exponential number of potential combinations that the user can specify.
The paper presents an approach that combines (a) cross-element vectorization for the easy-to-write transformation of the original scalar code to vectorized one, (b) meta-programming for utilization of a compiler for building sub-kernels tailored for a particular set of parameters, (c) and dynamic polymorphism allowing run-time selection of sub-kernels.
We show that the above techniques allow (1) straightforward code modifications, (2) efficient handling of required dynamic behavior with a minor performance penalty for most kernels, and (3) achieving up to 8-fold speedups compared to non-adapted kernels.
•High productivity approach for developing and implementing new FEM matrix kernels.•Automated approach that reaches the performance of hand-tuned kernels.•Technique allowing implementation of highly optimized kernels with standard C++ language features.•Approach combining SIMD vectorization with expression optimization with significantly higher impact on performance than pure vectorization. |
| ArticleNumber | 109319 |
| Author | Brzobohatý, Tomáš Kadlubiak, Kristian Meca, Ondřej Říha, Lubomír |
| Author_xml | – sequence: 1 givenname: Kristian orcidid: 0000-0001-7846-7848 surname: Kadlubiak fullname: Kadlubiak, Kristian email: kristian.kadlubiak@vsb.cz – sequence: 2 givenname: Ondřej surname: Meca fullname: Meca, Ondřej – sequence: 3 givenname: Lubomír surname: Říha fullname: Říha, Lubomír – sequence: 4 givenname: Tomáš surname: Brzobohatý fullname: Brzobohatý, Tomáš |
| BookMark | eNp9z71OwzAQwHEPRaItPACbXyDFjuPEFlNV-iW1YgBm6-LYwiW1IzuqFJ6eVGVmOt3wP91vhiY-eIPQEyULSmj5fFroTi9ykhfjLhmVEzQlhJKsKDm_R7OUToSQqpJsirZLj6HrYgD9hW2IuBk8nJ2Gth0wNND1ULcGv--Pr_hidB-i-4HeBY-DxZv1EX-b6E2bHtCdhTaZx785R5-b9cdqlx3etvvV8pDpvJB9ZgXUhIvcsII1FlhdEMYkB15ZxrmwwGsppMhBcsNKBiDKqtG2sLyUWlDC5oje7uoYUorGqi66M8RBUaKuenVSo15d9eqmH5uXWzP-aS7ORJW0M16bxsWRpJrg_ql_AT3mZWk |
| Cites_doi | 10.1016/j.cma.2004.01.015 10.1109/50.827514 10.1145/2687415 10.1016/j.cpc.2015.02.008 10.1177/1094342020945005 10.1016/j.camwa.2020.01.007 10.1145/1498765.1498785 10.1016/j.cpc.2021.108193 10.1016/j.camwa.2020.06.009 10.1002/spe.1149 10.1137/S0036144504444711 |
| ContentType | Journal Article |
| Copyright | 2024 Elsevier B.V. |
| Copyright_xml | – notice: 2024 Elsevier B.V. |
| DBID | AAYXX CITATION |
| DOI | 10.1016/j.cpc.2024.109319 |
| DatabaseName | CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Physics |
| ExternalDocumentID | 10_1016_j_cpc_2024_109319 S001046552400242X |
| GroupedDBID | --K --M -~X .DC .~1 0R~ 1B1 1RT 1~. 1~5 29F 4.4 457 4G. 5GY 5VS 7-5 71M 8P~ 9JN AACTN AAEDT AAEDW AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AARLI AAXUO AAYFN ABBOA ABFNM ABMAC ABNEU ABQEM ABQYD ABXDB ACDAQ ACFVG ACGFS ACLVX ACNNM ACRLP ACSBN ACZNC ADBBV ADECG ADEZE ADJOM ADMUD AEBSH AEKER AENEX AFKWA AFTJW AFZHZ AGHFR AGUBO AGYEJ AHHHB AHZHX AI. AIALX AIEXJ AIKHN AITUG AIVDX AJOXV AJSZI AKRWK ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AOUOD ASPBG ATOGT AVWKF AXJTR AZFZN BBWZM BKOJK BLXMC CS3 DU5 EBS EFJIC EJD EO8 EO9 EP2 EP3 F5P FDB FEDTE FGOYB FIRID FLBIZ FNPLU FYGXN G-2 G-Q GBLVA GBOLZ HLZ HME HMV HVGLF HZ~ IHE IMUCA J1W KOM LG9 LZ4 M38 M41 MO0 N9A NDZJH O-L O9- OAUVE OGIMB OZT P-8 P-9 P2P PC. Q38 R2- RIG ROL RPZ SBC SCB SDF SDG SES SEW SHN SPC SPCBC SPD SPG SSE SSK SSQ SSV SSZ T5K TN5 UPT VH1 WUQ ZMT ~02 ~G- 9DU AATTM AAXKI AAYWO AAYXX ABJNI ABWVN ACLOT ACRPL ACVFH ADCNI ADNMO AEIPS AEUPX AFJKZ AFPUW AGQPQ AIGII AIIUN AKBMS AKYEP ANKPU APXCP CITATION EFKBS EFLBG ~HD |
| ID | FETCH-LOGICAL-c249t-f8ab0582e343dfa3b403395a57f3558fa5b98982a95e363aa867dcf4f569c8103 |
| ISICitedReferencesCount | 0 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001279658300001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0010-4655 |
| IngestDate | Sat Nov 29 03:58:26 EST 2025 Sat Aug 24 15:41:22 EDT 2024 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | FEM kernels SIMD Dynamic polymorphism Meta-programming |
| Language | English |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c249t-f8ab0582e343dfa3b403395a57f3558fa5b98982a95e363aa867dcf4f569c8103 |
| ORCID | 0000-0001-7846-7848 |
| ParticipantIDs | crossref_primary_10_1016_j_cpc_2024_109319 elsevier_sciencedirect_doi_10_1016_j_cpc_2024_109319 |
| PublicationCentury | 2000 |
| PublicationDate | November 2024 2024-11-00 |
| PublicationDateYYYYMMDD | 2024-11-01 |
| PublicationDate_xml | – month: 11 year: 2024 text: November 2024 |
| PublicationDecade | 2020 |
| PublicationTitle | Computer physics communications |
| PublicationYear | 2024 |
| Publisher | Elsevier B.V |
| Publisher_xml | – name: Elsevier B.V |
| References | Bauer, Eibl, Godenschwager, Kohl, Kuron, Rettinger, Schornbaum, Schwarzmeier, Thönnes, Köstler, Rüde (br0240) 2021; 81 Anderson, Andrej, Barker, Bramwell, Camier, Cerveny, Dobrev, Dudouit, Fisher, Kolev, Pazner, Stowell, Tomov, Akkerman, Dahm, Medina, Zampini (br0190) 2021; 81 (br0170) 2018 Williams, Waterman, Patterson (br0280) 2009; 52 Obayya, Rahman, El-Mikati (br0130) 2000; 18 Akkurt, Witherden, Vincent (br0230) 2022; 271 Jansson (br0200) 2021 Intel corporation (br0210) 2012 Gebremedhin, Manne, Pothen (br0260) 2005; 47 Karpiński, McDonald (br0150) 2017 Reguly, Giles (br0220) 04 2013; 43 Luporini, Ham, Kelly (br0110) 04 2016; 44 (br0020) 2023 (br0030) 2023 (br0040) 2023 (br0050) 2023 (br0060) 2023 (br0070) 2023 Sun, Mitchell, Kulkarni, Klöckner, Ham, Kelly (br0140) 2020; 34 Kronbichler, Kormann (br0100) 2017 Luporini, Varbanescu, Rathgeber, Bercea, Ramanujam, Ham, Kelly (br0120) 2015; 11 (br0270) 2023 Cantwell, Moxey, Comerford, Bolis, Rocco, Mengaldo, De Grazia, Yakovlev, Lombard, Ekelschot, Jordi, Xu, Mohamied, Eskilsson, Nelson, Vos, Biotto, Kirby, Sherwin (br0180) 2015; 192 do Carmo, Alvarez (br0080) 2004; 193 Jubertie, Dupros, De Martin (br0090) 2018 Bathe (br0010) 2006 Intel corporation (br0290) 2023 Kretz, Lindenstruth (br0160) 2012; 42 Říha, Merta, Vavřík, Brzobohatý, Markopoulos, Meca, Vysocky, Kozubek, Vondrak (br0250) 2018; 33 Intel corporation (10.1016/j.cpc.2024.109319_br0290) Sun (10.1016/j.cpc.2024.109319_br0140) 2020; 34 Reguly (10.1016/j.cpc.2024.109319_br0220) 2013; 43 Bauer (10.1016/j.cpc.2024.109319_br0240) 2021; 81 Bathe (10.1016/j.cpc.2024.109319_br0010) 2006 do Carmo (10.1016/j.cpc.2024.109319_br0080) 2004; 193 Kretz (10.1016/j.cpc.2024.109319_br0160) 2012; 42 Cantwell (10.1016/j.cpc.2024.109319_br0180) 2015; 192 Jubertie (10.1016/j.cpc.2024.109319_br0090) 2018 Akkurt (10.1016/j.cpc.2024.109319_br0230) 2022; 271 Gebremedhin (10.1016/j.cpc.2024.109319_br0260) 2005; 47 Luporini (10.1016/j.cpc.2024.109319_br0120) 2015; 11 Williams (10.1016/j.cpc.2024.109319_br0280) 2009; 52 Kronbichler (10.1016/j.cpc.2024.109319_br0100) Karpiński (10.1016/j.cpc.2024.109319_br0150) 2017 (10.1016/j.cpc.2024.109319_br0070) 2023 Říha (10.1016/j.cpc.2024.109319_br0250) 2018; 33 Anderson (10.1016/j.cpc.2024.109319_br0190) 2021; 81 Jansson (10.1016/j.cpc.2024.109319_br0200) 2021 Intel corporation (10.1016/j.cpc.2024.109319_br0210) Luporini (10.1016/j.cpc.2024.109319_br0110) 2016; 44 Obayya (10.1016/j.cpc.2024.109319_br0130) 2000; 18 |
| References_xml | – volume: 33 year: 2018 ident: br0250 article-title: A massively parallel and memory-efficient fem toolbox with a hybrid total feti solver with accelerator support publication-title: Int. J. High Perform. Comput. Appl. – volume: 11 year: 2015 ident: br0120 article-title: Cross-loop optimization of arithmetic intensity for finite element local assembly publication-title: ACM Trans. Archit. Code Optim. – volume: 193 start-page: 2383 year: 2004 end-page: 2402 ident: br0080 article-title: A new upwind function in stabilized finite element formulations, using linear and quadratic elements for scalar convection–diffusion problems publication-title: Comput. Methods Appl. Mech. Eng. – volume: 44 year: 04 2016 ident: br0110 article-title: An algorithm for the optimization of finite element integration loops publication-title: ACM Trans. Math. Softw. – volume: 81 start-page: 42 year: 2021 end-page: 74 ident: br0190 article-title: Mfem: a modular finite element methods library publication-title: Comput. Math. Appl. – year: 2023 ident: br0060 article-title: Ansys - engineering simulation software – year: 2018 ident: br0170 – start-page: 32 year: 2021 end-page: 39 ident: br0200 article-title: Spectral element simulations on the nec sx-aurora tsubasa publication-title: The International Conference on High Performance Computing in Asia-Pacific Region – year: 2023 ident: br0050 article-title: MSC Nastran - multidisciplinary structural analysis – year: 2023 ident: br0270 – year: 2023 ident: br0070 article-title: ABAQUS - finite element analysis for mechanical engineering and civil engineering – start-page: 1 year: 2018 end-page: 7 ident: br0090 article-title: Vectorization of a spectral finite-element numerical kernel publication-title: Proceedings of the 2018 4th Workshop on Programming Models for SIMD/Vector Processing – year: 2012 ident: br0210 article-title: Intel® 64 and ia-32 architecturesoptimization reference manual – volume: 47 start-page: 629 year: 2005 end-page: 705 ident: br0260 article-title: What color is your Jacobian? Graph coloring for computing derivatives publication-title: SIAM Rev. – year: 2023 ident: br0020 article-title: Elmer FEM - open source multiphysical simulation software – volume: 43 year: 04 2013 ident: br0220 article-title: Finite element algorithms and data structures on graphical processing units publication-title: Int. J. Parallel Program. – volume: 81 start-page: 478 year: 2021 end-page: 501 ident: br0240 article-title: walberla: a block-structured high-performance framework for multiphysics simulations publication-title: Comput. Math. Appl. – year: 2023 ident: br0290 article-title: Intel® software development emulator (Intel® sde) – year: 2017 ident: br0100 article-title: Fast matrix-free evaluation of discontinuous Galerkin finite element operators – volume: 192 start-page: 205 year: 2015 end-page: 219 ident: br0180 article-title: Nektar++: an open-source spectral/hp element framework publication-title: Comput. Phys. Commun. – year: 2023 ident: br0040 – volume: 18 start-page: 409 year: 2000 end-page: 415 ident: br0130 article-title: New full vectorial numerically efficient propagation algorithm based on the finite element method publication-title: J. Lightwave Technol. – start-page: 21 year: 2017 end-page: 28 ident: br0150 article-title: A high-performance portable abstract interface for explicit simd vectorization publication-title: Proceedings of the 8th International Workshop on Programming Models and Applications for Multicores and Manycores – year: 2006 ident: br0010 article-title: Finite Element Procedures – volume: 34 start-page: 629 year: 2020 end-page: 644 ident: br0140 article-title: A study of vectorization for matrix-free finite element methods publication-title: Int. J. High Perform. Comput. Appl. – year: 2023 ident: br0030 article-title: CALCULIX - A Free Software Three-Dimensional Structural Finite Element Program – volume: 271 year: 2022 ident: br0230 article-title: Cache blocking strategies applied to flux reconstruction publication-title: Comput. Phys. Commun. – volume: 52 start-page: 65 year: 2009 end-page: 76 ident: br0280 article-title: Roofline: an insightful visual performance model for multicore architectures publication-title: Commun. ACM – volume: 42 start-page: 1409 year: 2012 end-page: 1430 ident: br0160 article-title: Vc: a C++ library for explicit vectorization publication-title: Softw. Pract. Exp. – volume: 193 start-page: 2383 issue: 23 year: 2004 ident: 10.1016/j.cpc.2024.109319_br0080 article-title: A new upwind function in stabilized finite element formulations, using linear and quadratic elements for scalar convection–diffusion problems publication-title: Comput. Methods Appl. Mech. Eng. doi: 10.1016/j.cma.2004.01.015 – volume: 18 start-page: 409 year: 2000 ident: 10.1016/j.cpc.2024.109319_br0130 article-title: New full vectorial numerically efficient propagation algorithm based on the finite element method publication-title: J. Lightwave Technol. doi: 10.1109/50.827514 – volume: 11 issue: 4 year: 2015 ident: 10.1016/j.cpc.2024.109319_br0120 article-title: Cross-loop optimization of arithmetic intensity for finite element local assembly publication-title: ACM Trans. Archit. Code Optim. doi: 10.1145/2687415 – ident: 10.1016/j.cpc.2024.109319_br0290 – volume: 192 start-page: 205 year: 2015 ident: 10.1016/j.cpc.2024.109319_br0180 article-title: Nektar++: an open-source spectral/hp element framework publication-title: Comput. Phys. Commun. doi: 10.1016/j.cpc.2015.02.008 – start-page: 32 year: 2021 ident: 10.1016/j.cpc.2024.109319_br0200 article-title: Spectral element simulations on the nec sx-aurora tsubasa – ident: 10.1016/j.cpc.2024.109319_br0100 – volume: 34 start-page: 629 issue: 6 year: 2020 ident: 10.1016/j.cpc.2024.109319_br0140 article-title: A study of vectorization for matrix-free finite element methods publication-title: Int. J. High Perform. Comput. Appl. doi: 10.1177/1094342020945005 – volume: 33 year: 2018 ident: 10.1016/j.cpc.2024.109319_br0250 article-title: A massively parallel and memory-efficient fem toolbox with a hybrid total feti solver with accelerator support publication-title: Int. J. High Perform. Comput. Appl. – volume: 81 start-page: 478 year: 2021 ident: 10.1016/j.cpc.2024.109319_br0240 article-title: walberla: a block-structured high-performance framework for multiphysics simulations publication-title: Comput. Math. Appl. doi: 10.1016/j.camwa.2020.01.007 – volume: 52 start-page: 65 issue: 4 year: 2009 ident: 10.1016/j.cpc.2024.109319_br0280 article-title: Roofline: an insightful visual performance model for multicore architectures publication-title: Commun. ACM doi: 10.1145/1498765.1498785 – year: 2006 ident: 10.1016/j.cpc.2024.109319_br0010 – volume: 44 year: 2016 ident: 10.1016/j.cpc.2024.109319_br0110 article-title: An algorithm for the optimization of finite element integration loops publication-title: ACM Trans. Math. Softw. – volume: 271 year: 2022 ident: 10.1016/j.cpc.2024.109319_br0230 article-title: Cache blocking strategies applied to flux reconstruction publication-title: Comput. Phys. Commun. doi: 10.1016/j.cpc.2021.108193 – volume: 81 start-page: 42 year: 2021 ident: 10.1016/j.cpc.2024.109319_br0190 article-title: Mfem: a modular finite element methods library publication-title: Comput. Math. Appl. doi: 10.1016/j.camwa.2020.06.009 – start-page: 1 year: 2018 ident: 10.1016/j.cpc.2024.109319_br0090 article-title: Vectorization of a spectral finite-element numerical kernel – ident: 10.1016/j.cpc.2024.109319_br0210 – year: 2023 ident: 10.1016/j.cpc.2024.109319_br0070 – start-page: 21 year: 2017 ident: 10.1016/j.cpc.2024.109319_br0150 article-title: A high-performance portable abstract interface for explicit simd vectorization – volume: 42 start-page: 1409 issue: 11 year: 2012 ident: 10.1016/j.cpc.2024.109319_br0160 article-title: Vc: a C++ library for explicit vectorization publication-title: Softw. Pract. Exp. doi: 10.1002/spe.1149 – volume: 43 year: 2013 ident: 10.1016/j.cpc.2024.109319_br0220 article-title: Finite element algorithms and data structures on graphical processing units publication-title: Int. J. Parallel Program. – volume: 47 start-page: 629 issue: 4 year: 2005 ident: 10.1016/j.cpc.2024.109319_br0260 article-title: What color is your Jacobian? Graph coloring for computing derivatives publication-title: SIAM Rev. doi: 10.1137/S0036144504444711 |
| SSID | ssj0007793 |
| Score | 2.4483907 |
| Snippet | The paper focuses on the optimization of the FEM matrix kernels with respect to user-defined parameters such as materials, initial conditions, and boundary... |
| SourceID | crossref elsevier |
| SourceType | Index Database Publisher |
| StartPage | 109319 |
| SubjectTerms | Dynamic polymorphism FEM kernels Meta-programming SIMD |
| Title | An approach for dynamically adaptable SIMD vectorization of FEM kernels |
| URI | https://dx.doi.org/10.1016/j.cpc.2024.109319 |
| Volume | 304 |
| WOSCitedRecordID | wos001279658300001&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: 0010-4655 databaseCode: AIEXJ dateStart: 19950101 customDbUrl: isFulltext: true dateEnd: 99991231 titleUrlDefault: https://www.sciencedirect.com omitProxy: false ssIdentifier: ssj0007793 providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3db9MwELfKBhIvE59iDJAf2AtVUGPHdfyYrR0MsTFpQ-pb5CS2tq4kVZdWG38954-kGQMJkHiJWrd2qrtfL-fz3e8QesuKDJBRkGDIQxIYRq4gE1wGOiNZmIWaDmJtm03w4-N4MhEnvd5NUwuzmvGyjK-vxfy_qhrGQNmmdPYv1N0uCgPwGpQOV1A7XP9I8UnZEoXbHMLC9ZyXs9lNXxZyXttiqdPDo1F_ZUP2vhLTuI0H46P-pVqUjmN5TWHgWz_4OMiVSURfl5Wsj4NkMVtmF_Kyaz1ajarc-qlfymJ3n-0KoabNR_Z9bI_sR-786fMyq765gTZ5eG_xvcqqc1mb8b2RBZr_UmhWSMJuBINEvpSvDavdKa1xphoeEIbcrWuqqWtVfMfsuwjE9H0-N6yUJLIkWd4U32bTPrWERLCsyZ0F92RyD20SzgQYxM3kcDz51D7GOfeMzf53NEfiNjnwpxv92qnpOCpnj9CW32HgxCHjMeqp8gl6cOI09xR9SErc4AMDPnAHH7jFBzb4wLfwgSuNAR_Y4-MZ-nowPtv_GPhuGkEOW-w60LHMBiwmika00JLCf5RSwSTj2lDsa8ky00uUSMEUHVIp4yEvch1pNhR5HA7oc7RRVqV6gfAg0mZmGFING2yYlUcDosATjgRnipFt9K4RRzp3pClpk004TUF2qZFd6mS3jaJGYKn3-pw3l4J2fz_t5b9N20EP1xB8hTbqxVK9RvfzVX1xtXjjMfADM-t0Bw |
| 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=An+approach+for+dynamically+adaptable+SIMD+vectorization+of+FEM+kernels&rft.jtitle=Computer+physics+communications&rft.au=Kadlubiak%2C+Kristian&rft.au=Meca%2C+Ond%C5%99ej&rft.au=%C5%98%C3%ADha%2C+Lubom%C3%ADr&rft.au=Brzobohat%C3%BD%2C+Tom%C3%A1%C5%A1&rft.date=2024-11-01&rft.pub=Elsevier+B.V&rft.issn=0010-4655&rft.volume=304&rft_id=info:doi/10.1016%2Fj.cpc.2024.109319&rft.externalDocID=S001046552400242X |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0010-4655&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0010-4655&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0010-4655&client=summon |