Auto-tuning for floating-point precision with Discrete Stochastic Arithmetic
•An algorithm and a software are proposed for floating-point precision auto-tuning.•Input programs are automatically modified taking into account accuracy requirements.•Discrete Stochastic Arithmetic is used to verify the results accuracy.•Performance results are presented on several benchmarks and...
Uloženo v:
| Vydáno v: | Journal of computational science Ročník 36; s. 101017 |
|---|---|
| Hlavní autoři: | , , , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
Elsevier B.V
01.09.2019
Elsevier |
| Témata: | |
| ISSN: | 1877-7503, 1877-7511 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | •An algorithm and a software are proposed for floating-point precision auto-tuning.•Input programs are automatically modified taking into account accuracy requirements.•Discrete Stochastic Arithmetic is used to verify the results accuracy.•Performance results are presented on several benchmarks and on an industrial code.
The type length chosen for floating-point numbers (e.g. 32 bits or 64 bits) may have an impact on the execution time, especially on SIMD (Single Instruction Multiple Data) units. Furthermore optimizing the types used in a numerical simulation causes a reduction of the data volume that is possibly transferred. In this paper we present PROMISE, a tool that makes it possible to optimize the numerical types in a program by taking into account the requested accuracy on the computed results. With PROMISE the numerical quality of results is verified using DSA (Discrete Stochastic Arithmetic) that enables one to estimate round-off errors. The search for a suitable type configuration is performed with a reasonable complexity thanks to the delta debugging algorithm. The PROMISE tool has been successfully tested on programs implementing several numerical algorithms including linear system solving and also on an industrial code that solves the neutron transport equations. |
|---|---|
| AbstractList | The type length chosen for floating-point numbers (e.g. 32 bits or 64 bits) may have an impact on the execution time, especially on SIMD (Single Instruction Multiple Data) units. Furthermore optimizing the types used in a numerical simulation causes a reduction of the data volume that is possibly transferred. In this paper we present PROMISE, a tool that makes it possible to optimize the numerical types in a program by taking into account the requested accuracy on the computed results. With PROMISE the numerical quality of results is verified using DSA (Discrete Stochastic Arithmetic) that enables one to estimate round-off errors. The search for a suitable type configuration is performed with a reasonable complexity thanks to the delta debugging algorithm. The PROMISE tool has been successfully tested on programs implementing several numerical algorithms including linear system solving and also on an industrial code that solves the neutron transport equations. •An algorithm and a software are proposed for floating-point precision auto-tuning.•Input programs are automatically modified taking into account accuracy requirements.•Discrete Stochastic Arithmetic is used to verify the results accuracy.•Performance results are presented on several benchmarks and on an industrial code. The type length chosen for floating-point numbers (e.g. 32 bits or 64 bits) may have an impact on the execution time, especially on SIMD (Single Instruction Multiple Data) units. Furthermore optimizing the types used in a numerical simulation causes a reduction of the data volume that is possibly transferred. In this paper we present PROMISE, a tool that makes it possible to optimize the numerical types in a program by taking into account the requested accuracy on the computed results. With PROMISE the numerical quality of results is verified using DSA (Discrete Stochastic Arithmetic) that enables one to estimate round-off errors. The search for a suitable type configuration is performed with a reasonable complexity thanks to the delta debugging algorithm. The PROMISE tool has been successfully tested on programs implementing several numerical algorithms including linear system solving and also on an industrial code that solves the neutron transport equations. |
| ArticleNumber | 101017 |
| Author | Févotte, François Jézéquel, Fabienne Graillat, Stef Picot, Romain Lathuilière, Bruno |
| Author_xml | – sequence: 1 givenname: Stef surname: Graillat fullname: Graillat, Stef email: Stef.Graillat@lip6.fr organization: Sorbonne Université, CNRS, LIP6, F-75005 Paris, France – sequence: 2 givenname: Fabienne surname: Jézéquel fullname: Jézéquel, Fabienne email: Fabienne.Jezequel@lip6.fr organization: Sorbonne Université, CNRS, LIP6, F-75005 Paris, France – sequence: 3 givenname: Romain surname: Picot fullname: Picot, Romain email: Romain.Picot@lip6.fr organization: Sorbonne Université, CNRS, LIP6, F-75005 Paris, France – sequence: 4 givenname: François orcidid: 0000-0003-2402-0535 surname: Févotte fullname: Févotte, François email: Francois.Fevotte@edf.fr organization: EDF R&D, 7 boulevard Gaspard Monge, 91120 Palaiseau, France – sequence: 5 givenname: Bruno surname: Lathuilière fullname: Lathuilière, Bruno email: Bruno.Lathuiliere@edf.fr organization: EDF R&D, 7 boulevard Gaspard Monge, 91120 Palaiseau, France |
| BackLink | https://hal.science/hal-01331917$$DView record in HAL |
| BookMark | eNp9kDtPwzAQgC1UJErpH2DKypDgq5PakVii8ihSJAZgthzHpo7SuLLdIv49joo6MNTLPXzfSfddo8lgB4XQLeAMMCzvu6yz0mcLDGWGaYZxfoGmwChNaQEwOeWYXKG59x2OjzBWApmiutoHm4b9YIavRFuX6N6KEIt0Z80Qkp1T0nhjh-TbhE3yaLx0KqjkPVi5ET4YmVQu_mxVTG_QpRa9V_O_OEOfz08fq3Vav728rqo6lYSykOZkCawpGAFN2rallGidLwuKVStyLEnDCAGcN8CkahcNgdgpdVEy1kpcForM0N1x70b0fOfMVrgfboXh66rmYw9D3FACPUCcXRxnpbPeO6VPAGA-6uMdH_XxUR_HlEd9EWL_IGlC1GKH4ITpz6MPR1RFAQejHPfSqCFeYqLKwFtrzuG_GRONKA |
| CitedBy_id | crossref_primary_10_1109_TETC_2021_3070422 crossref_primary_10_3390_e23010086 crossref_primary_10_3390_electronics10182209 crossref_primary_10_1016_j_microrel_2021_114099 |
| Cites_doi | 10.1016/j.cpc.2010.07.006 10.1007/s10009-016-0435-0 10.1016/j.cpc.2008.11.005 10.1016/0378-4754(93)90003-D 10.1145/3014426 10.1109/32.988498 10.1016/j.cpc.2008.02.003 10.1016/j.cpc.2007.01.005 10.1023/B:NUMA.0000049483.75679.ce 10.1007/s11786-011-0103-4 10.1016/j.matcom.2005.11.014 |
| ContentType | Journal Article |
| Copyright | 2019 Elsevier B.V. Distributed under a Creative Commons Attribution 4.0 International License |
| Copyright_xml | – notice: 2019 Elsevier B.V. – notice: Distributed under a Creative Commons Attribution 4.0 International License |
| DBID | AAYXX CITATION 1XC VOOES |
| DOI | 10.1016/j.jocs.2019.07.004 |
| DatabaseName | CrossRef Hyper Article en Ligne (HAL) Hyper Article en Ligne (HAL) (Open Access) |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Sciences (General) Business Computer Science |
| EISSN | 1877-7511 |
| ExternalDocumentID | oai:HAL:hal-01331917v1 10_1016_j_jocs_2019_07_004 S1877750318309475 |
| GroupedDBID | --K --M .~1 0R~ 1B1 1~. 1~5 4.4 457 4G. 5VS 7-5 71M 8P~ AACTN AAEDT AAEDW AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AAXUO AAYFN ABBOA ABFRF ABJNI ABMAC ABXDB ABYKQ ACDAQ ACGFO ACGFS ACNNM ACRLP ACZNC ADBBV ADEZE ADMUD AEBSH AEFWE AEKER AENEX AFKWA AFTJW AGHFR AGUBO AGYEJ AHZHX AIALX AIEXJ AIKHN AITUG AJBFU AJOXV ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AOUOD AXJTR BKOJK BLXMC EBS EFJIC EFLBG EJD EP3 FDB FEDTE FIRID FNPLU FYGXN GBLVA GBOLZ HVGLF HZ~ J1W KOM M41 MO0 N9A O-L O9- OAUVE P-8 P-9 P2P PC. Q38 RIG ROL SDF SES SPC SPCBC SSV SSZ T5K UNMZH ~G- 9DU AATTM AAXKI AAYWO AAYXX ABWVN ACLOT ACRPL ACVFH ADCNI ADNMO AEIPS AEUPX AFJKZ AFPUW AIGII AIIUN AKBMS AKRWK AKYEP ANKPU APXCP CITATION EFKBS ~HD 1XC VOOES |
| ID | FETCH-LOGICAL-c378t-43618b5831f3ddd773ff46570eda40c3b833104b18ced2b313b89f5988dc095e3 |
| ISICitedReferencesCount | 19 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000497247400008&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1877-7503 |
| IngestDate | Tue Oct 14 20:26:39 EDT 2025 Tue Nov 18 21:51:49 EST 2025 Sat Nov 29 06:57:34 EST 2025 Fri Feb 23 02:22:07 EST 2024 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Round-off errors Numerical validation Discrete Stochastic Arithmetic Auto-tuning Floating-point arithmetic floating-point arithmetic round-off errors numerical validation auto-tuning |
| Language | English |
| License | Distributed under a Creative Commons Attribution 4.0 International License: http://creativecommons.org/licenses/by/4.0 |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c378t-43618b5831f3ddd773ff46570eda40c3b833104b18ced2b313b89f5988dc095e3 |
| ORCID | 0000-0003-2402-0535 0000-0002-8782-7566 |
| OpenAccessLink | https://hal.science/hal-01331917 |
| ParticipantIDs | hal_primary_oai_HAL_hal_01331917v1 crossref_primary_10_1016_j_jocs_2019_07_004 crossref_citationtrail_10_1016_j_jocs_2019_07_004 elsevier_sciencedirect_doi_10_1016_j_jocs_2019_07_004 |
| PublicationCentury | 2000 |
| PublicationDate | 2019-09-01 |
| PublicationDateYYYYMMDD | 2019-09-01 |
| PublicationDate_xml | – month: 09 year: 2019 text: 2019-09-01 day: 01 |
| PublicationDecade | 2010 |
| PublicationTitle | Journal of computational science |
| PublicationYear | 2019 |
| Publisher | Elsevier B.V Elsevier |
| Publisher_xml | – name: Elsevier B.V – name: Elsevier |
| References | Lattner, Adve (bib0045) 2004 Vignes (bib0055) 1993; 35 Zeller, Hildebrandt (bib0025) 2002; 28 Scott, Jézéquel, Denis, Chesneaux (bib0095) 2007; 176 Chesneaux, Vignes (bib0070) 1992; 315 Darulova, Kuncak (bib0190) 2017; 39 Chiang, Baranowski, Briggs, Solovyev, Gopalakrishnan, Rakamarić (bib0165) 2017 IEEE Computer Society (bib0005) 2008 Rubio-González, Nguyen, Nguyen, Demmel, Kahan, Sen, Bailey, Iancu, Hough (bib0040) 2013 Damouche, Martel, Chapoutot (bib0180) 2017; 19 Halsall (bib0145) 1980 Darulova, Izycheva, Nasir, Ritter, Becker, Bastian (bib0200) 2018 Brajard, Li, Jézéquel, Benavidès, Thiria (bib0100) 2013 Jul Jézéquel, Rico, Chesneaux, Charikhi (bib0090) 2006; 71 Lamotte, Chesneaux, Jézéquel (bib0085) 2010; 181 Vignes (bib0065) 1986; 303 Graillat, Jézéquel, Picot, Févotte, Lathuilière (bib0015) 2016 Moustafa, Févotte, Lathuilière, Plagne (bib0150) 2013 Panchekha, Sanchez-Stern, Wilcox, Tatlock (bib0170) 2015; 2015 Nathan, Naeimi, Sorin, Sun (bib0155) 2016 Contributors of Center for Manycore Programming, Seoul (bib0135) 2010 Chesneaux (bib0060) 1995 Graillat, Jézéquel, Wang, Zhu (bib0120) 2011; 5 Lam, Hollingsworth, de Supinski, Legendre (bib0035) 2013 Eberhart, Landreau, Brajard, Fortin, Jézéquel (bib0210) 2018 May Vignes (bib0030) 2004; 37 Li (bib0125) 2012 Jul Baboulin, Buttari, Dongarra, Kurzak, Langou, Langou, Luszczek, Tomov (bib0010) 2009; 180 Jézéquel, Lamotte, Said (bib0110) 2015 Zeller (bib0020) 2009 Sanchez-Stern, Panchekha, Lerner, Tatlock (bib0175) 2018 Damouche, Martel (bib0185) 2018 Jézéquel, Chesneaux (bib0080) 2008; 178 Rump (bib0115) 1988 Jézéquel, Lamotte, Chubach (bib0105) 2013 Graillat, Jézéquel, Picot, Févotte, Lathuilière (bib0205) 2018; vol. 8 Nguyen, Rubio-González, Mehne, Sen, Demmel, Kahan, Iancu, Lavrijsen, Bailey, Hough (bib0050) 2015 Contributors (bib0130) 2010 Darulova, Horn, Sharma (bib0195) 2018 Févotte, Lathuilière (bib0140) 2013 May Solovyev, Jacobsen, Rakamarić, Gopalakrishnan (bib0160) 2015 Eberhart, Brajard, Fortin, Jézéquel (bib0075) 2015; 21 Lamotte (10.1016/j.jocs.2019.07.004_bib0085) 2010; 181 Li (10.1016/j.jocs.2019.07.004_bib0125) 2012 Vignes (10.1016/j.jocs.2019.07.004_bib0065) 1986; 303 Contributors of Center for Manycore Programming, Seoul (10.1016/j.jocs.2019.07.004_bib0135) 2010 Darulova (10.1016/j.jocs.2019.07.004_bib0195) 2018 Lam (10.1016/j.jocs.2019.07.004_bib0035) 2013 Panchekha (10.1016/j.jocs.2019.07.004_bib0170) 2015; 2015 Chiang (10.1016/j.jocs.2019.07.004_bib0165) 2017 Vignes (10.1016/j.jocs.2019.07.004_bib0030) 2004; 37 Contributors (10.1016/j.jocs.2019.07.004_bib0130) 2010 Scott (10.1016/j.jocs.2019.07.004_bib0095) 2007; 176 Chesneaux (10.1016/j.jocs.2019.07.004_bib0060) 1995 Damouche (10.1016/j.jocs.2019.07.004_bib0180) 2017; 19 Rump (10.1016/j.jocs.2019.07.004_bib0115) 1988 Graillat (10.1016/j.jocs.2019.07.004_bib0015) 2016 Sanchez-Stern (10.1016/j.jocs.2019.07.004_bib0175) 2018 Eberhart (10.1016/j.jocs.2019.07.004_bib0210) 2018 Baboulin (10.1016/j.jocs.2019.07.004_bib0010) 2009; 180 Chesneaux (10.1016/j.jocs.2019.07.004_bib0070) 1992; 315 Brajard (10.1016/j.jocs.2019.07.004_bib0100) 2013 Solovyev (10.1016/j.jocs.2019.07.004_bib0160) 2015 Darulova (10.1016/j.jocs.2019.07.004_bib0190) 2017; 39 Jézéquel (10.1016/j.jocs.2019.07.004_bib0105) 2013 Vignes (10.1016/j.jocs.2019.07.004_bib0055) 1993; 35 Jézéquel (10.1016/j.jocs.2019.07.004_bib0110) 2015 IEEE Computer Society (10.1016/j.jocs.2019.07.004_bib0005) 2008 Nathan (10.1016/j.jocs.2019.07.004_bib0155) 2016 Jézéquel (10.1016/j.jocs.2019.07.004_bib0090) 2006; 71 Graillat (10.1016/j.jocs.2019.07.004_bib0120) 2011; 5 Nguyen (10.1016/j.jocs.2019.07.004_bib0050) 2015 Zeller (10.1016/j.jocs.2019.07.004_bib0025) 2002; 28 Moustafa (10.1016/j.jocs.2019.07.004_bib0150) 2013 Halsall (10.1016/j.jocs.2019.07.004_bib0145) 1980 Zeller (10.1016/j.jocs.2019.07.004_bib0020) 2009 Lattner (10.1016/j.jocs.2019.07.004_bib0045) 2004 Darulova (10.1016/j.jocs.2019.07.004_bib0200) 2018 Jézéquel (10.1016/j.jocs.2019.07.004_bib0080) 2008; 178 Févotte (10.1016/j.jocs.2019.07.004_bib0140) 2013 Graillat (10.1016/j.jocs.2019.07.004_bib0205) 2018; vol. 8 Eberhart (10.1016/j.jocs.2019.07.004_bib0075) 2015; 21 Damouche (10.1016/j.jocs.2019.07.004_bib0185) 2018 Rubio-González (10.1016/j.jocs.2019.07.004_bib0040) 2013 |
| References_xml | – start-page: 109 year: 1988 end-page: 126 ident: bib0115 article-title: ch. Algorithms for verified inclusions – theory and practice publication-title: Reliability in Computing: The Role of Interval Methods in Scientific Computing – year: 2016 ident: bib0155 article-title: Profile-Driven Automated Mixed Precision – volume: 71 start-page: 44 year: 2006 end-page: 61 ident: bib0090 article-title: Reliable computation of a multiple integral involved in the neutron star theory publication-title: Math. Comput. Simul. – volume: 176 start-page: 507 year: 2007 end-page: 521 ident: bib0095 article-title: Numerical ‘health check’ for scientific codes: the CADNA approach publication-title: Comput. Phys. Commun. – year: 1980 ident: bib0145 article-title: CACTUS, a Characteristics Solution to the Neutron Transport Equations in Complicated Geometries, Ser. AEEW /R: AEEW – year: 2016 ident: bib0015 article-title: PROMISE: floating-point precision tuning with stochastic arithmetic publication-title: 17th International Symposium on Scientific Computing, Computer Arithmetic and Validated Numerics (SCAN 2016) – year: 2013 ident: bib0105 article-title: Parallelization of Discrete Stochastic Arithmetic on multicore architectures publication-title: 10th International Conference on Information Technology: New Generations (ITNG) – volume: 5 start-page: 359 year: 2011 end-page: 375 ident: bib0120 article-title: Stochastic arithmetic in multiprecision publication-title: Math. Comput. Sci. – year: 2010 ident: bib0135 article-title: SNU NPB Suite – year: 2004 ident: bib0045 article-title: LLVM: a compilation framework for lifelong program analysis & transformation publication-title: Proceedings of the 2004 International Symposium on Code Generation and Optimization (CGO’04) – volume: 19 start-page: 427 year: 2017 end-page: 448 ident: bib0180 article-title: Improving the numerical accuracy of programs by automatic transformation publication-title: STTT – start-page: 63 year: 2018 end-page: 76 ident: bib0185 article-title: Salsa: an automatic tool to improve the numerical accuracy of programs publication-title: 6th International Workshop on Automated Formal Methods (AFM), vol. 5 – volume: 180 start-page: 2526 year: 2009 end-page: 2533 ident: bib0010 article-title: Accelerating scientific computations with mixed precision algorithms publication-title: Comput. Phys. Commun. – volume: 181 start-page: 1925 year: 2010 end-page: 1926 ident: bib0085 article-title: CADNA_C: a version of CADNA for use with C or C++ programs publication-title: Comput. Phys. Commun. – year: 2010 ident: bib0130 article-title: GSL Project, “GSL – GNU Scientific Library – GNU Project – Free Software Foundation (FSF) – start-page: 532 year: 2015 end-page: 550 ident: bib0160 article-title: Rigorous estimation of floating-point round-off errors with symbolic Taylor expansions publication-title: FM 2015: Formal Methods – year: 1995 ident: bib0060 article-title: L’arithmétique stochastique et le logiciel CADNA, Habilitation à diriger des recherches – volume: 303 start-page: 997 year: 1986 end-page: 1000 ident: bib0065 article-title: Zéro mathématique et zéro informatique publication-title: C. R. l’Acad. Sci. – Ser. I – Math. – year: 2012 Jul ident: bib0125 article-title: Numerical Accuracy Analysis in Simulations on Hybrid High-Performance Computing Systems – start-page: 256 year: 2018 end-page: 269 ident: bib0175 article-title: Finding root causes of floating point error publication-title: Proceedings of the 39th ACM SIGPLAN Conference on Programming Language Design and Implementation, ser. PLDI 2018 – volume: 2015 start-page: 1 year: 2015 end-page: 11 ident: bib0170 article-title: Automatically improving accuracy for floating point expressions publication-title: Proceedings of the 36th ACM SIGPLAN Conference on Programming Language Design and Implementation – PLDI – year: 2013 ident: bib0040 article-title: Precimonious: tuning assistant for floating-point precision publication-title: Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis, ser. SC’13 – year: 2015 ident: bib0050 article-title: Floating-Point Precision Tuning Using Blame Analysis – year: 2013 May ident: bib0140 article-title: MICADO: parallel implementation of a 2D-1D iterative algorithm for the 3D neutron transport problem in prismatic geometries publication-title: Proceedings of Mathematics, Computational Methods & Reactor Physics – start-page: 1016 year: 2018 May end-page: 1025 ident: bib0210 article-title: Improving CADNA performance on GPUs publication-title: 2018 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW) – start-page: 270 year: 2018 end-page: 287 ident: bib0200 article-title: Daisy – framework for analysis and optimization of numerical programs (tool paper) publication-title: 24th International Conference on Tools and Algorithms for the Construction and Analysis of Systems (TACAS), vol. 10805 LNCS – volume: 178 start-page: 933 year: 2008 end-page: 955 ident: bib0080 article-title: CADNA: a library for estimating round-off error propagation publication-title: Comput. Phys. Commun. – start-page: 208 year: 2018 end-page: 219 ident: bib0195 article-title: Sound mixed-precision optimization with rewriting publication-title: Proceedings of the 9th ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS’18) – volume: 35 start-page: 233 year: 1993 end-page: 261 ident: bib0055 article-title: A stochastic arithmetic for reliable scientific computation publication-title: Math. Comput. Simul. – volume: 315 start-page: 1435 year: 1992 end-page: 1440 ident: bib0070 article-title: Les fondements de l’arithmétique stochastique publication-title: C. R. l’Acad. Sci. – Ser. I – Math. – volume: vol. 8 start-page: 38 year: 2018 end-page: 53 ident: bib0205 article-title: Numerical validation in quadruple precision using stochastic arithmetic publication-title: TNC’18. Trusted Numerical Computations, ser. Kalpa Publications in Computing – start-page: 300 year: 2017 end-page: 315 ident: bib0165 article-title: Rigorous floating-point mixed-precision tuning publication-title: Proceedings of the 44th ACM SIGPLAN Symposium on Principles of Programming Languages, ser. POPL 2017 – year: 2008 ident: bib0005 article-title: IEEE Standard for Floating-Point Arithmetic – year: 2013 ident: bib0150 article-title: Vectorization of a 2D-1D iterative algorithm for the 3D neutron transport problem in prismatic geometries publication-title: Joint International Conference on Supercomputing in Nuclear Applications and Monte Carlo 2013 (SNA + MC 2013) – volume: 21 start-page: 35 year: 2015 end-page: 52 ident: bib0075 article-title: High performance numerical validation using stochastic arithmetic publication-title: Reliab. Comput. – volume: 39 year: 2017 ident: bib0190 article-title: Towards a compiler for reals publication-title: ACM Trans. Program. Lang. Syst. (TOPLAS) – year: 2009 ident: bib0020 article-title: Why Programs Fail – volume: 28 start-page: 183 year: 2002 end-page: 200 ident: bib0025 article-title: Simplifying and isolating failure-inducing input publication-title: IEEE Trans. Softw. Eng. – year: 2013 Jul ident: bib0100 article-title: Numerical validation of data assimilation codes generated by the YAO software publication-title: SIAM Annual Meeting – volume: 37 start-page: 377 year: 2004 end-page: 390 ident: bib0030 article-title: Discrete Stochastic Arithmetic for validating results of numerical software publication-title: Numer. Algorithms – start-page: 675 year: 2015 end-page: 680 ident: bib0110 article-title: Estimation of numerical reproducibility on CPU and GPU publication-title: Annals of Computer Science and Information Systems, Proceedings of the 2015 Federated Conference on Computer Science and Information Systems (FedCSIS), vol. 5 – start-page: 369 year: 2013 end-page: 378 ident: bib0035 article-title: Automatically adapting programs for mixed-precision floating-point computation publication-title: Proceedings of the 27th International ACM Conference on International Conference on Supercomputing, ser. ICS’13 – start-page: 532 year: 2015 ident: 10.1016/j.jocs.2019.07.004_bib0160 article-title: Rigorous estimation of floating-point round-off errors with symbolic Taylor expansions – volume: 181 start-page: 1925 issue: 11 year: 2010 ident: 10.1016/j.jocs.2019.07.004_bib0085 article-title: CADNA_C: a version of CADNA for use with C or C++ programs publication-title: Comput. Phys. Commun. doi: 10.1016/j.cpc.2010.07.006 – year: 2016 ident: 10.1016/j.jocs.2019.07.004_bib0015 article-title: PROMISE: floating-point precision tuning with stochastic arithmetic – start-page: 369 year: 2013 ident: 10.1016/j.jocs.2019.07.004_bib0035 article-title: Automatically adapting programs for mixed-precision floating-point computation – year: 1995 ident: 10.1016/j.jocs.2019.07.004_bib0060 – year: 2004 ident: 10.1016/j.jocs.2019.07.004_bib0045 article-title: LLVM: a compilation framework for lifelong program analysis & transformation – year: 1980 ident: 10.1016/j.jocs.2019.07.004_bib0145 – volume: 19 start-page: 427 issue: 4 year: 2017 ident: 10.1016/j.jocs.2019.07.004_bib0180 article-title: Improving the numerical accuracy of programs by automatic transformation publication-title: STTT doi: 10.1007/s10009-016-0435-0 – volume: 180 start-page: 2526 issue: 12 year: 2009 ident: 10.1016/j.jocs.2019.07.004_bib0010 article-title: Accelerating scientific computations with mixed precision algorithms publication-title: Comput. Phys. Commun. doi: 10.1016/j.cpc.2008.11.005 – year: 2015 ident: 10.1016/j.jocs.2019.07.004_bib0050 – year: 2012 ident: 10.1016/j.jocs.2019.07.004_bib0125 – volume: vol. 8 start-page: 38 year: 2018 ident: 10.1016/j.jocs.2019.07.004_bib0205 article-title: Numerical validation in quadruple precision using stochastic arithmetic – volume: 35 start-page: 233 issue: 3 year: 1993 ident: 10.1016/j.jocs.2019.07.004_bib0055 article-title: A stochastic arithmetic for reliable scientific computation publication-title: Math. Comput. Simul. doi: 10.1016/0378-4754(93)90003-D – year: 2010 ident: 10.1016/j.jocs.2019.07.004_bib0135 – volume: 2015 start-page: 1 year: 2015 ident: 10.1016/j.jocs.2019.07.004_bib0170 article-title: Automatically improving accuracy for floating point expressions publication-title: Proceedings of the 36th ACM SIGPLAN Conference on Programming Language Design and Implementation – PLDI – volume: 39 issue: 2 year: 2017 ident: 10.1016/j.jocs.2019.07.004_bib0190 article-title: Towards a compiler for reals publication-title: ACM Trans. Program. Lang. Syst. (TOPLAS) doi: 10.1145/3014426 – start-page: 270 year: 2018 ident: 10.1016/j.jocs.2019.07.004_bib0200 article-title: Daisy – framework for analysis and optimization of numerical programs (tool paper) – volume: 28 start-page: 183 issue: February (2) year: 2002 ident: 10.1016/j.jocs.2019.07.004_bib0025 article-title: Simplifying and isolating failure-inducing input publication-title: IEEE Trans. Softw. Eng. doi: 10.1109/32.988498 – start-page: 1016 year: 2018 ident: 10.1016/j.jocs.2019.07.004_bib0210 article-title: Improving CADNA performance on GPUs – year: 2013 ident: 10.1016/j.jocs.2019.07.004_bib0150 article-title: Vectorization of a 2D-1D iterative algorithm for the 3D neutron transport problem in prismatic geometries publication-title: Joint International Conference on Supercomputing in Nuclear Applications and Monte Carlo 2013 (SNA + MC 2013) – volume: 178 start-page: 933 issue: 12 year: 2008 ident: 10.1016/j.jocs.2019.07.004_bib0080 article-title: CADNA: a library for estimating round-off error propagation publication-title: Comput. Phys. Commun. doi: 10.1016/j.cpc.2008.02.003 – year: 2008 ident: 10.1016/j.jocs.2019.07.004_bib0005 – start-page: 109 year: 1988 ident: 10.1016/j.jocs.2019.07.004_bib0115 article-title: ch. Algorithms for verified inclusions – theory and practice – year: 2013 ident: 10.1016/j.jocs.2019.07.004_bib0100 article-title: Numerical validation of data assimilation codes generated by the YAO software – year: 2013 ident: 10.1016/j.jocs.2019.07.004_bib0140 article-title: MICADO: parallel implementation of a 2D-1D iterative algorithm for the 3D neutron transport problem in prismatic geometries publication-title: Proceedings of Mathematics, Computational Methods & Reactor Physics – start-page: 208 year: 2018 ident: 10.1016/j.jocs.2019.07.004_bib0195 article-title: Sound mixed-precision optimization with rewriting – volume: 303 start-page: 997 year: 1986 ident: 10.1016/j.jocs.2019.07.004_bib0065 article-title: Zéro mathématique et zéro informatique publication-title: C. R. l’Acad. Sci. – Ser. I – Math. – start-page: 63 year: 2018 ident: 10.1016/j.jocs.2019.07.004_bib0185 article-title: Salsa: an automatic tool to improve the numerical accuracy of programs publication-title: 6th International Workshop on Automated Formal Methods (AFM), vol. 5 – start-page: 300 year: 2017 ident: 10.1016/j.jocs.2019.07.004_bib0165 article-title: Rigorous floating-point mixed-precision tuning – start-page: 256 year: 2018 ident: 10.1016/j.jocs.2019.07.004_bib0175 article-title: Finding root causes of floating point error – year: 2016 ident: 10.1016/j.jocs.2019.07.004_bib0155 – volume: 315 start-page: 1435 year: 1992 ident: 10.1016/j.jocs.2019.07.004_bib0070 article-title: Les fondements de l’arithmétique stochastique publication-title: C. R. l’Acad. Sci. – Ser. I – Math. – volume: 176 start-page: 507 issue: April (8) year: 2007 ident: 10.1016/j.jocs.2019.07.004_bib0095 article-title: Numerical ‘health check’ for scientific codes: the CADNA approach publication-title: Comput. Phys. Commun. doi: 10.1016/j.cpc.2007.01.005 – start-page: 675 year: 2015 ident: 10.1016/j.jocs.2019.07.004_bib0110 article-title: Estimation of numerical reproducibility on CPU and GPU – volume: 21 start-page: 35 year: 2015 ident: 10.1016/j.jocs.2019.07.004_bib0075 article-title: High performance numerical validation using stochastic arithmetic publication-title: Reliab. Comput. – year: 2010 ident: 10.1016/j.jocs.2019.07.004_bib0130 – year: 2009 ident: 10.1016/j.jocs.2019.07.004_bib0020 – year: 2013 ident: 10.1016/j.jocs.2019.07.004_bib0105 article-title: Parallelization of Discrete Stochastic Arithmetic on multicore architectures – volume: 37 start-page: 377 issue: Deceber (1–4) year: 2004 ident: 10.1016/j.jocs.2019.07.004_bib0030 article-title: Discrete Stochastic Arithmetic for validating results of numerical software publication-title: Numer. Algorithms doi: 10.1023/B:NUMA.0000049483.75679.ce – year: 2013 ident: 10.1016/j.jocs.2019.07.004_bib0040 article-title: Precimonious: tuning assistant for floating-point precision – volume: 5 start-page: 359 issue: 4 year: 2011 ident: 10.1016/j.jocs.2019.07.004_bib0120 article-title: Stochastic arithmetic in multiprecision publication-title: Math. Comput. Sci. doi: 10.1007/s11786-011-0103-4 – volume: 71 start-page: 44 issue: 1 year: 2006 ident: 10.1016/j.jocs.2019.07.004_bib0090 article-title: Reliable computation of a multiple integral involved in the neutron star theory publication-title: Math. Comput. Simul. doi: 10.1016/j.matcom.2005.11.014 |
| SSID | ssj0000388913 |
| Score | 2.2654386 |
| Snippet | •An algorithm and a software are proposed for floating-point precision auto-tuning.•Input programs are automatically modified taking into account accuracy... The type length chosen for floating-point numbers (e.g. 32 bits or 64 bits) may have an impact on the execution time, especially on SIMD (Single Instruction... |
| SourceID | hal crossref elsevier |
| SourceType | Open Access Repository Enrichment Source Index Database Publisher |
| StartPage | 101017 |
| SubjectTerms | Auto-tuning Computer Science Discrete Stochastic Arithmetic Floating-point arithmetic Numerical validation Round-off errors |
| Title | Auto-tuning for floating-point precision with Discrete Stochastic Arithmetic |
| URI | https://dx.doi.org/10.1016/j.jocs.2019.07.004 https://hal.science/hal-01331917 |
| Volume | 36 |
| WOSCitedRecordID | wos000497247400008&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 customDbUrl: eissn: 1877-7511 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0000388913 issn: 1877-7503 databaseCode: AIEXJ dateStart: 20100501 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3Pa9swFBZpO8YuY-022v1CjB02ikcc2ZZ0NFtDN0rZoYPcjC3L1CG1Q-qE0v-h__Pes2Q5KUtZD7s4QYkUkff56fnp0_cI-YSSZApCeW_IZOCBl4w8yYT2GEtHWaohAtFBW2yCn5-LyUT-GgzuurMwqxmvKnFzI-f_1dTQBsbGo7OPMLcbFBrgPRgdrmB2uP6T4eNlU3vNsuooksWsTpHb7M3rsmpQFMBU1TEp2O8l-A0InFGVW12mqNp8HC_gkys83bgldFVtKYgujWgX0Z7Jk2Ilo8ZyyArH0TF78rfmBZajNvk8TrOWcuZcNCCzMZTvq7R0yB2bXqu6MeX8MN5um3hdbmQu_J6a1TlbwbmH-6jr3pitu1MfFfD4Xz29STpMv05rharrvmw1WE0p401Z7XvLnSMhdvy2aYJjJDhGMsTd-GCH7I14KMFJ7sU_TiY_XdIOpXNkW3Lbzd0exDKcwfuT2Rbs7Fx2afs2jLl4QZ5bI9LY4GafDHR1QJ52xx8OyL519tf0s1Uk__KSnK1higKm6CamqMMURUzRDlO0xxTtMfWK_B6fXHw79WwZDk8xLhovYJEvslAwv2B5nnPOiiJAxpTO02CoWCYYPCMEmS-UzkcZ86FFFqEUIlcQwGv2muxWdaUPCVUFU9CY-TxSQaSkTCMYWIcyKyBw5OKI-N3_lSirUY-lUmbJdmMdkWPXZ24UWh78dtiZIbG3h4kdEwDWg_0-gs3cD6Ao-2l8lmAbPEQxzHqs_DePmspb8qy_K96R3Wax1O_JE7VqyuvFBwu9P9L8pow |
| 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=Auto-tuning+for+floating-point+precision+with+Discrete+Stochastic+Arithmetic&rft.jtitle=Journal+of+computational+science&rft.au=Graillat%2C+Stef&rft.au=J%C3%A9z%C3%A9quel%2C+Fabienne&rft.au=Picot%2C+Romain&rft.au=F%C3%A9votte%2C+Fran%C3%A7ois&rft.date=2019-09-01&rft.issn=1877-7503&rft.volume=36&rft.spage=101017&rft_id=info:doi/10.1016%2Fj.jocs.2019.07.004&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_jocs_2019_07_004 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1877-7503&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1877-7503&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1877-7503&client=summon |