DALI: Digits, architectures et logiciels informatiques
L’équipe DALI développe une thématique de recherche unifiée afin d’améliorer la qualité numérique et la haute performance des calculs. DALI permet l’interaction, rare en France au sein d’une même équipe, d’experts en micro-architecture et en arithmétique des ordinateurs.
Côté performances, nos travaux portent sur l’exploitation du potentiel de calcul toujours croissant des processeurs : élargissement des chemins (micro-architecture vectorielle), multiplication des cœurs (parallélisme de tâches), augmentation du parallélisme d’instructions. Côté arithmétique, la qualité numérique des applications de calcul scientifique et la sûreté de fonctionnement d’applications embarquées dépendent crucialement de la maîtrise de la précision finie et de l’arithmétique flottante en particulier. Il s’agit de contrôler et certifier les calculs (algorithmes, codes) mais aussi d’optimiser la précision des résultats. De nombreux logiciels, scientifiques ou embarqués, nécessitent d’améliorer la qualité numérique sans pour autant sacrifier la rapidité d’exécution. Ainsi se rejoignent amélioration de la performance et de la qualité numérique.
Membres
Permanents
- Philippe Langlois, Professeur des Universités UPerpignan
- Sylvia Munoz, Technicien UPerpignan
- Christophe Negre, Maître de Conférences UPerpignan
- David Parello, Maître de Conférences UPerpignan
- Guillaume Revy, Maître de Conférences UPerpignan
- Vincent Zucca, Maître de Conférences UPerpignan
Non permanents
- Harris Ludwig Boudi, Doctorant UPerpignan
- Bernard Goossens, Invité longue durée, Professeur Emérite UPerpignan
- Kenelm Louetsi, Doctorant UPerpignan
Thématiques de recherche
La cohérence thématique des travaux de recherche sur l’amélioration de la qualité et de la performance des calculs est une des forces de l’équipe DALI.
L’amélioration de la performance des calculs est étroitement liée aux améliorations apportées aux micro-architectures. Celle-ci est réalisée suivant plusieurs directions, par élargissement des chemins (micro-architecture vectorielle), multiplication des cœurs (parallélisme de tâches), ou encore augmentation du parallélisme d’instructions (ILP). La qualité numérique des applications de calcul scientifique ou la sûreté de fonctionnement d’applications embarquées critiques dépendent crucialement de la maîtrise des effets de la précision finie des calculs — et de l’arithmétique flottante en particulier. Il s’agit alors de contrôler et valider les calculs (algorithmes, codes) mais aussi d’améliorer et optimiser la précision des calculs et des résultats.
Les travaux développés sur la période 2008-2013 sont organisés autour de 4 actions de recherche :
- Action 1. Mesure reproductible et analyse du potentiel de parallélisme et des performances.
- Action 2. Meilleure exploitation des nouvelles architectures multicœurs.
- Action 3. Implantation sûre et efficace de protocoles cryptographiques.
- Action 4. Synthèse de code pour l’implémentation de calculs précis, rapides et certifiés.
Partenariats et collaborations académiques
Partenaires industriels : Actility, Airbus, Inpixal, Prover, Rockwell-Collins, Thales, Total.
Projets collaboratifs : ANR Blanc EvaFlo (2006-2010), Projet DPAC MASSANE (2007-2010), ANR ARPEGE (2009-2011), Projet FNRAE SARDANES (2009-2012), Projet Chercheur d’avenir Compil’HD (2010-2012), ANR INS DEFIS (2011-2014), ANR INS CAFEIN (2012-2015), PEPS QUARENUM (2013).
Collaborations nationales : CEA-LIST, ENSTA, EXASCALE, IRISA, LIENS, LIP, LIP6, LRI, LSIS, ONERA.
Collaborations internationales : Universitat de Girona, Technical University Hamburg, University of Malaysia Sabah, Mississippi State University, Microsoft Research Redmond, Rice University, Tokyo WCU, University of Waterloo (Canada), University of Wollongong, University of Waseda.
Publications majeures
- Bernard Goossens and David Parello, Limits of Instruction-Level Parallelism Capture, International Conference on Computational Science (ALCHEMY Workshop), 2013, to appear.
- Stef Graillat, Philippe Langlois, and Nicolas Louvet. Algorithms for accurate, validated and fast computations with polynomials. Japan Journal of Industrial and Applied Mathematics, 26(2,3):191-214, 2009.
- Sylvain Collange, Marc Daumas and David Defour, Interval Arithmetic, in GPU Computing Gems, Jade Edition, ISBN 978-0-12-385963-1, 2011.
- Anwar Hasan, Nicolas Meloni, Ashkan Namin, and Christophe Negre,Block Recombination Approach for Subquadratic Space Complexity Binary Field Multiplication based on Toeplitz Matrix-Vector Product ,IEEE Transactions on Computers, Vol 61(2), pages 151-163, 2012.
- Arnault Ioualalen and Matthieu Martel, A New Abstract Domain for the Representation of Mathematically Equivalent Expressions, Static Analysis Symposium, SAS'12, Lecture Notes in Computer Science, Volume 7460, pages 75-93, Springer-Verlag, 2012.
- Claude-Pierre Jeannerod, Hervé Knochel, Christophe Monat, and Guillaume Revy, Computing floating-point square roots via bivariate polynomial evaluation, IEEE Transactions on Computers, Vol. 60(2), pages 214-227, February 2011.
Publications depuis 2014 - Evaluation 2019
Articles de revues internationales
2020
- Optimizing Cherenkov photons generation and propagation in CORSIKA for CTA Monte-Carlo simulationsLuisa Arrabito, Matthieu Carrère, Konrad Bernlöhr, Johan Bregeon, Adnane Khattabi, Philippe Langlois, David Parello, Guillaume RevyComputing and Software for Big Science, Springer, 2020, 4, pp.9.
2019
- Efficient Fixed Base Exponentiation and Scalar Multiplication based on a Multiplicative Splitting Exponent RecodingJean-Marc Robert, Christophe Negre, Thomas PlantardJournal of Cryptographic Engineering, Springer, 2019, 9 (2), pp.115-136.
- Hierarchical approach for deriving a reproducible unblocked LU factorizationRoman Iakymchuk, Stef Graillat, David Defour, Enrique Quintana-OrtíInternational Journal of High Performance Computing Applications, SAGE Publications, 2019, pp.#1094342019832968.
2017
- Midpoint-Radius Interval-based Method to Deal with Uncertainty in Power Flow AnalysisManuel Marin, Federico Milano, David DefourElectric Power Systems Research, Elsevier, 2017, 147, pp.81-87.
- Trade-offs of certified fixed-point code synthesis for linear algebra basic blocksMatthieu Martel, Mohamed Amine Najahi, Guillaume RevyJournal of Systems Architecture, Elsevier, 2017, 76, pp.133-148.
- Computing On Many CoresBernard Goossens, David Parello, Katarzyna Porada, Djallal RahmouneConcurrency and Computation: Practice and Experience, Wiley, 2017, 29 (15), pp.e4120.
- Automatic source-to-source error compensation of floating-point programs: code synthesis to optimize accuracy and timeLaurent Thévenoux, Philippe Langlois, Matthieu MartelConcurrency and Computation: Practice and Experience, Wiley, 2017, Concurrency and Computation: Practice and Experience, 29 (7), pp.e3953.
- Exact Lookup Tables for the Evaluation of Trigonometric and Hyperbolic FunctionsHugues de Lassus Saint-Geniès, David Defour, Guillaume RevyIEEE Transactions on Computers, Institute of Electrical and Electronics Engineers, 2017, 66 (12), pp.2058-2071.
- Efficient Regular Modular Exponentiation Using Multiplicative Half-Size SplittingChristophe Negre, Thomas PlantardJournal of Cryptographic Engineering, Springer, 2017, 7 (3), pp.245-253.
2016
- An efficient representation format for fuzzy intervals based on symmetric membership functionsManuel Marin, David Defour, Federico MilanoACM Transactions on Mathematical Software, Association for Computing Machinery, 2016, 43 (3), pp.23:1--23:22.
- A software scheduling solution to avoid corrupted units on GPUsDavid Defour, Eric PetitJournal of Parallel and Distributed Computing, Elsevier, 2016, 90-91, pp.1--8.
2015
- Transformation of a PID Controller for Numerical AccuracyNasrine Damouche, Matthieu Martel, Alexandre ChapoutotElectronic Notes in Theoretical Computer Science, Elsevier, 2015, 317, pp.47-54.
- Numerical Reproducibility for the Parallel Reduction on Multi- and Many-Core ArchitecturesSylvain Collange, David Defour, Stef Graillat, Roman IakymchukParallel Computing, Elsevier, 2015, 49, pp.83-97.
- New Parallel Approaches for Scalar Multiplication in Elliptic Curve over Fields of Small CharacteristicChristophe Negre, Jean-Marc RobertIEEE Transactions on Computers, Institute of Electrical and Electronics Engineers, 2015, 64 (10), pp.2875-2890.
2014
- Efficient Subquadratic Space Complexity Binary Polynomial Multipliers Based On Block RecombinationMurat Cenk, Anwar Hasan, Christophe NegreIEEE Transactions on Computers, Institute of Electrical and Electronics Engineers, 2014, 63 (9), pp.2273-2287.
- First steps towards more numerical reproducibilityFabienne Jézéquel, Philippe Langlois, Nathalie RevolESAIM: Proceedings and Surveys, EDP Sciences, 2014, ESAIM: Proceedings and Surveys, 45, pp.229-238.
- Efficient Binary Polynomial Multiplication Based on Optimized Karatsuba ReconstructionChristophe NegreJournal of Cryptographic Engineering, Springer, 2014, 4 (2), pp.91--106.
- A Fast Chaos-Based Pseudo-Random Bit Generator Using Binary64 Floating-Point ArithmeticMichael François, David Defour, Christophe NegreInformatica, Slovene Society Informatika, Ljubljana, 2014, 38 (2), pp.115-124.
- Pseudo-random number generator based on mixing of three chaotic mapsMichael François, Thomas Grosges, Dominique Barchiesi, Robert ErraCommunications in Nonlinear Science and Numerical Simulation, Elsevier, 2014, 19 (4), pp.887--895.
Communications internationales
2019
- Reproducible and Accurate Parallel Triangular SolverChemseddine Chohra, Philippe Langlois, David Parello9th International Congress on Industrial and Applied Mathematics (ICIAM), Jul 2019, Valencia, Spain. <https://www.siam.org/conferences/cm/conference/iciam19>
2018
- Numerical Accuracy Stuff: Tools. . . and PrerequisitesPhilippe LangloisCTAOptSim General Workshop, Dec 2018, Montpellier, France.
- Meta-implementation of vectorized logarithm function in binary floating-point arithmeticHugues de Lassus Saint-Geniès, Nicolas Brunie, Guillaume RevyASAP: Application-specific Systems, Architectures and Processors, Jul 2018, Milan, Italy. <https://asap18.necst.it>
- Performance optimization of the air shower simulation program for the Cherenkov Telescope ArrayLuisa Arrabito, Konrad Bernlöhr, Johan Bregeon, Gernot Maier, Philippe Langlois, David Parello, Guillaume RevyCHEP: Computing in High Energy and Nuclear Physics, Jul 2018, Sofia, Bulgaria. pp.05041.
2017
- Efficient Leak Resistant Modular Exponentiation in RNSAndrea Lesavourey, Christophe Negre, Thomas PlantardARITH: Computer Arithmetic, Jul 2017, London, United Kingdom. pp.156-163.
- Asynchronous Power Flow on Graphic Processing UnitsManuel Marin, David Defour, Federico MilanoPDP: Parallel, Distributed and network-Based Processing, Mar 2017, St Petersburg, Russia.
- Reproducible Parallel Simulations in HPCChemseddine Chohra, Philippe Langlois, Rafife Nheilli, David ParelloCSE: Computational Science and Engineering, Feb 2017, Altanta, Georgia, United States. <https://archive.siam.org/meetings/cse17/>
2016
- Hierarchical Approach for Deriving a Reproducible LU factorization on GPUsRoman Iakymchuk, Stef Graillat, David Defour, Enrique Quintana-OrtíThe Numerical Reproducibility at Exascale (NRE16) workshop held as part of the Supercomputing Conference (SC16), Nov 2016, Salt Lake City, UT, United States.
- First improvements toward a reproducible Telemac-2DRafife Nheili, Philippe Langlois, Christophe DenisXXIIIrd TELEMAC-MASCARET User Conference , Oct 2016, Paris, France. <http://www.opentelemac.org/index.php/user-conference26>
- Towards Fast, Accurate and Reproducible LU FactorizationRoman Iakymchuk, David Defour, Stef GraillatSCAN 2016, 17th international symposium on Scientific Computing, Computer Arithmetic and Validated Numerics, Sep 2016, Uppsala, Sweden. pp.59-60.
- Parallel experiments with RARE-BLASChemseddine Chohra, Philippe Langlois, David ParelloSYNASC: Symbolic and Numeric Algorithms for Scientific Computing, Sep 2016, Timisoara, Romania. pp.135-138.
- Reproducible, Accurately Rounded and Efficient BLASChemseddine Chohra, Philippe Langlois, David ParelloEuro-Par: Parallel Processing Workshops., Aug 2016, Grenoble, France. pp.609-620.
- Efficient Randomized Regular Modular Exponentiation using Combined Montgomery and Barrett MultiplicationsAndrea Lesavourey, Christophe Negre, Thomas PlantardICETE: International Joint Conference on e-Business and Telecommunications, Jul 2016, Lisbon, Portugal. pp.368-375.
- Enhanced Digital Signature using RNS Digit Exponent RepresentationThomas Plantard, Jean-Marc RobertWAIFI: Workshop on the Arithmetic of Finite Fields, Department of Mathematics of Ghent University, Jul 2016, Gand, Belgium. pp.177-192.
- Recovering numerical reproducibility in hydrodynamic simulationsPhilippe Langlois, Rafife Nheili, Christophe DenisARITH: Computer Arithmetic, Jul 2016, Silicon Valley, Santa Clara, CA, United States. pp.63-70.
- Automated design of floating-point logarithm functions on integer processorsGuillaume RevyARITH: Computer Arithmetic, Jul 2016, Silicon Valley, Santa Clara, CA, United States. pp.172-180.
- Reproducible and Accurate Algorithms for Numerical Linear AlgebraRoman Iakymchuk, David Defour, Sylvain Collange, Stef GraillatPP: Parallel Processing for Scientific Computing, Apr 2016, Paris, France.
- Parallel Locality and Parallelization QualityBernard Goossens, David Parello, Katarzyna Porada, Djallal RahmounePMAM: Programming Models and Applications for Multicores and Manycores, Mar 2016, Barcelona, Spain. pp.59-68.
2015
- ExBLAS: Reproducible and Accurate BLAS LibraryRoman Iakymchuk, Sylvain Collange, David Defour, Stef GraillatNRE: Numerical Reproducibility at Exascale, Nov 2015, Austin, TX, United States.
- Automatic Source-to-Source Error Compensation of Floating-Point ProgramsLaurent Thévenoux, Philippe Langlois, Matthieu MartelComputational Science and Engineering (CSE), Oct 2015, Porto, Portugal. pp.9--16.
- Measuring predictability of Nvidia’s GPU warp and block schedulers: Application to the summation problemDavid DefourMCSoC: Embedded Multicore/Many-core Systems-on-Chip, Sep 2015, Turin, Italy. pp.17-24.
- Reproducible floating-point atomic addition in data-parallel environmentDavid Defour, Sylvain CollangeACSIS, Sep 2015, Lodz, Poland. pp.721-728.
- Toward a Core Design to Distribute an Execution on a Many-Core ProcessorBernard Goossens, David Parello, Katarzyna Porada, Djallal RahmounePaCT: Parallel Computing Technologies, Aug 2015, Petrozavodsk, Russia. pp.390-404.
- Numerical Reproducibility: Feasibility IssuesPhilippe Langlois, Rafife Nheili, Christophe DenisNTMS: New Technologies, Mobility and Security, Jul 2015, Paris, France. pp.1-5.
- Range Reduction Based on Pythagorean Triples for Trigonometric Function EvaluationHugues de Lassus Saint-Geniès, David Defour, Guillaume RevyASAP: Application-specific Systems, Architectures and Processors, Jul 2015, Toronto, Canada. pp.74-81.
- Parallel Approaches for Efficient Scalar Multiplication over Elliptic CurveChristophe Negre, Jean-Marc RobertMultiplication over Elliptic Curve. SECRYPT, Jul 2015, Colmar, France. pp.202-209.
- Numerical Reproducibility in open TELEMAC: A Case Study within the Tomawac LibraryRafife Nheili, Philippe Langlois, Christophe DenisHPCSET: High Performance Computing Simulation in Energy/Transport Domains, Jul 2015, Frankfurt, Germany.
- Trade-off Approaches for Leak Resistant Modular Arithmetic in RNSChristophe Negre, Guilherme PerinACISP: Australasian Conference on Information Security and Privacy, Jun 2015, Brisbane, Australia. pp.107-124.
- Intra-procedural Optimization of the Numerical Accuracy of ProgramsNasrine Damouche, Matthieu Martel, Alexandre ChapoutotFMICS: Formal Methods for Industrial Critical Systems, Jun 2015, Oslo, Norway. pp.31-46.
- Efficient Modular Exponentiation Based on Multiple Multiplications by a Common OperandChristophe Negre, Thomas Plantard, Jean-Marc RobertARITH: Computer Arithmetic, INRIA, Jun 2015, Lyon, France. pp.144-151.
- Optimizing the Accuracy of a Rocket Trajectory Simulation by Program TransformationNasrine Damouche, Matthieu Martel, Alexandre ChapoutotCF: Computing Frontiers, May 2015, Ischia, Italy. pp.40.
- Reproducible Triangular Solvers for High-Performance ComputingRoman Iakymchuk, David Defour, Sylvain Collange, Stef GraillatITNG: Information Technology - New Generations, Apr 2015, Las Vegas, NV, United States. pp.353-358.
- An efficient midpoint-radius implementation to handle symmetric fuzzy intervalsManuel Marin, David Defour, Federico MilanoRAIM: Rencontres Arithmétiques de l’Informatique Mathématique, Apr 2015, Rennes, France.
2014
- Software Implementation of Parallelized ECSM over Binary and Prime FieldsJean-Marc RobertInscrypt: Information Security and Cryptology, Dec 2014, Beijing, China. pp.445-462.
- Toward the synthesis of fixed-point code for matrix inversion based on Cholesky decompositionMatthieu Martel, Mohamed Amine Najahi, Guillaume RevyDASIP: Design and Architectures for Signal and Image Processing, Oct 2014, Madrid, Spain. pp.1-8.
- Automated Synthesis of Target-Dependent Programs for Polynomial Evaluation in Fixed-Point ArithmeticChristophe Mouilleron, Mohamed Amine Najahi, Guillaume RevySYNASC: Symposium on Symbolic and Numeric Algorithms for Scientific Computing, Sep 2014, Timisoara, Romania. pp.141-148.
- Efficiency of Reproducible Level 1 BLASChemseddine Chohra, Philippe Langlois, David ParelloSCAN: Scientific Computing, Computer Arithmetic, and Validated Numerics, Sep 2014, Würzburg, Germany. pp.99-108.
- Level 1 Parallel RTN-BLAS: Implementation and Efficiency AnalysisChemseddine Chohra, Philippe Langlois, David ParelloSCAN: Scientific Computing, Computer Arithmetic and Validated Numerics, Sep 2014, Wurzburg, Germany. <http://www.scan2014.uni-wuerzburg.de/talks/>
- Reproducible and Accurate Matrix MultiplicationRoman Iakymchuk, David Defour, Sylvain Collange, Stef GraillatSCAN: Scientific Computing, Computer Arithmetic and Validated Numerics, Sep 2014, Wurzburg, Germany. pp.126-137.
- Reproducible and Accurate Matrix Multiplication for High-Performance ComputingSylvain Collange, David Defour, Stef Graillat, Roman IakymchukSCAN: Scientific Computing, Computer Arithmetic and Validated Numerics, Sep 2014, Wuerzburg, Germany. pp.42-43.
- Power Flow Analysis under Uncertainty using Symmetric Fuzzy ArithmeticManuel Marin, David Defour, Federico MilanoPES General Meeting 2014 | Conference & Exposition, Jul 2014, National Harbor, MD, United States. pp.1-5.
- A Reproducible Accurate Summation Algorithm for High-Performance ComputingSylvain Collange, David Defour, Stef Graillat, Roman IakymchukEX: Exascale Applied Mathematics Challenges and Opportunities, Jul 2014, Chicago, United States. <http://www.siam.org/meetings/ex14/>
- A Pseudo-Random Bit Generator Based on Three Chaotic Logistic Maps and IEEE 754-2008 Floating-Point ArithmeticMichael François, David Defour, Pascal BerthoméTheory and Applications of Models of Computation, Apr 2014, Chennai, India. pp.229-247.
- FuzzyGPU : a fuzzy arithmetic library for GPUManuel Marin, David DefourPDP: Parallel, Distributed and Network-Based Processing, Feb 2014, Torino, Italy. pp.624-631.
- Code Size and Accuracy-Aware Synthesis of Fixed-Point Programs for Matrix MultiplicationMatthieu Martel, Mohamed Amine Najahi, Guillaume RevyPECCS: Pervasive and Embedded Computing and Communication Systems, Jan 2014, Lisbonne, Portugal. <10.5220/0004884802040214>
Dernière mise à jour le 02/12/2020
Département : Informatique
Responsable : Philippe LANGLOIS
Site de l'équipe : http://webdali.univ-perp.fr
Fiche-équipe DALI
Télécharger la fiche-équipe DALI du rapport d'activité 2008-2013 :