Book chapter

  1. S. Collange, M. Daumas & D. Defour, Chapter 9 – Interval Arithmetic in CUDA, GPU Computing Gems Jade Edition, Morgan Kaufmann, pp. 99-107, ISBN 9780123859631.

Book

  1. D. Defour, B. Goossens & C. Rochange, Architecture des ordinateurs , Technique et Science Informatiques, 27(6), Hermès Science, Juin 2008.

Articles in international journal

  1. D. Defour, P de Oliveira Castro, M Iştoan, E Petit, A Study of the Effects and Benefits of Custom-Precision Mathematical Libraries for HPC Codes, IEEE Transactions on Emerging Topics in Computing, April 2021.
  2. R. Iakymchuk, S. Graillat, D. Defour, E. Laure & E. Quintana-Orti, Hierarchical Approach for Deriving a Reproducible Unblocked LU Factorization, International Journal of High Performance Computing Applications, 2019.
  3. H. De Lassus Saint-Geniès, D. Defour & G. Revy, Exact Lookup Tables for the Evaluation of Trigonometric and Hyperbolic Functions , IEEE Transaction on Computer (IEEE TC), Vol. 66, no. 12, pp. 2058-2071, December 2017.
  4. M. Marin, D. Defour & F. Milano,  Midpoint-Radius Interval-based Method to Deal with Uncertainty in Power Flow Analysis , Electric Power Systems Research, Elsevier, Vol. 147, pp.81-87, June 2017.
  5. M. Marin, D. Defour & F. Milano,  An efficient representation format for fuzzy intervals based on symmetric membership functions, ACM Transaction on Mathematical Software, Vol. 43(3), Art. 23, December 2016.
  6. D. Defour & E. Petit, A software scheduling solution to avoid corrupted units on GPUs, Journal of Parallel and Distributed Computing, Elsevier, pp. 1-8, Vol.90-91, 2016.
  7. R. Iakymchuk, D. Defour, S. Collange & S. Graillat, Numerical Reproducibility for the Parallel Reduction on Multi- and Many-Core Architectures, Journal on Parallel Computing, Elsevier, pp. 83-97, November 2015.
  8. M. François, D. Defour & C. Negre, A Fast Chaos-Based Pseudo-Random Bit Generator Using Binary64 Floating-Point Arithmetic, Informatica, 38(2014), pp. 115-124, 2014.
  9. S. Collange, M. Daumas & D. Defour, Line-by-line spectroscopic simulations on graphics processing units, Computer Physics Communications, 178(2), pp. 135-143, 2008.
  10. B. Goossens & D. Defour, The instruction register file micro-architecture, Future Generation Computer Systems, 21(5), pp. 767-773, 2006.
  11. N. Brisebarre, D. Defour, P. Kornerup, J.-M. Muller & N. Revol, A New Range-Reduction Algorithm, IEEE Trans. Computers, 54(3), pp. 331-339, 2005. 
  12. D. Defour, G. Hanrot, V. Lefèvre, J.-M. Muller, N. Revol & P. Zimmermann, Proposal for a Standardization of Mathematical Function Implementation in Floating-Point Arithmetic, Numerical Algorithms, Vol. 37, Number 1-4, pp. 367-375, 2004.

Peer-reviewed articles in international conference proceedings

  1. D. Defour, P de Oliveira Castro, M Iştoan, E Petit, Shadow computation with BFloat16 to estimate the numerical accuracy of summations, 2021 IEEE 28th Symposium on Computer Arithmetic (ARITH), to appear, June 2021.
  2. D. Defour, P de Oliveira Castro, M Iştoan, E Petit, Custom-Precision Mathematical Library Explorations for Code Profiling and Optimization, 2020 IEEE 27th Symposium on Computer Arithmetic (ARITH), pp 121-124, June 2020.
  3. Y. Chatelain, E. Petit, P. de Oliveira Castro, G. Lartigue,D. Defour Automatic exploration of reduced floating-point representations in iterative methods , International Conference on Parallel Processing (Euro-Par 2019), August 2019.
  4. D. Defour, FP-ANR: A representation format to handle floating-point cancellation at run-time , 25th IEEE Symposium on Computer Arithmetic (ARITH’23), June 2018.
  5. Chatelain, P. de Oliveira Castro, E. Petit,D. Defour, J. Bieder, M. Torrent, VeriTracer: Context-enriched tracer for floating-point arithmetic analysis , 25th IEEE Symposium on Computer Arithmetic (ARITH’23), June 2018.
  6. R. Iakymchuk, S. Graillat , D. Defour, E. Laure & E. Quintana-Orti, Towards A Reproducible Solution of Linear Systems , Workshop on Computational Reproducibility at Exascale, Supercomputing 2017, November 2017.
  7. M. Marin, D. Defour & F. Milano,   Asynchronous Power Flow on Graphic Processing Units , 25th Euromicro International Conference on Parallel, Distributed and network-Based Processing, PDP2017, March 2017.
  8. R. Iakymchuk, S. Graillat, D. Defour & E. Quintana-Orti, Hierarchical Approach for deriving a reproducible LU factorization on GPUs , Workshop on Numerical reproducibility at Exascale, Supercomputing 2016, November 2016.
  9. R. Iakymchuk, D. Defour, S. Collange & S. Graillat, Reproducible and Accurate matrix multiplication , Proceedings of the 16th GAMM-IMACS International Symposium on scientific computing, Computer arithmetic and validated numerics, LNCS 9553. 
  10. R. Iakymchuk, D. Defour, S. Collange & S. Graillat, ExBLAS: Reproducible and Accurate BLAS Library , Workshop on Numerical reproducibility at Exascale, Supercomputing 2015, November 2015. 
  11. D. Defour, Measuring predictability of Nvidia’s GPU warp and block schedulers: Application to the summation problem , IEEE 9th International Symposium on Embedded Multicore/Many-core Systems-on-Chip (MCSoC-15), September 2015. 
  12. D. Defour & S. Collange, Reproducible floating-point atomic addition in data-parallel environment, Federated Conference on Computer Science and Information Systems (FedCSIS), September 2015. 
  13. H. De Lassus Saint-Geniès, D. Defour & G. Revy, Range Reduction Based on Pythagorean Triples for Trigonometric Function Evaluation , 26th IEEE International Conference on Application-specific Systems, Architectures and Processors, (ASAP), July 2015. 
  14. R. Iakymchuk, D. Defour, S. Collange & S. Graillat, Reproducible Triangular Solvers for High-Performance Computing , International Conference on Information Technology – New Generations, ITNG, April 2015. 
  15. M. Marin, D. Defour & F. Milano,  Power Flow Analysis under Uncertainty using Symmetric Fuzzy Arithmetic, IEEE Power & Energy Society General Meeting, 2014. (Best Paper Session)
  16. D. Defour & M. Marin,  FuzzyGPU : a fuzzy arithmetic library for GPU , PDP, 22nd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing, pp 624 – 631, 2014. (Best Paper Award for GPU applications) 
  17. M. François, D. Defour & P. Berthomé,   A pseudo-random bit generator based on three chaotic logistic maps and IEEE-754-2008 floating-point arithmetic , 11th Annual conference on Theory and Application of Models of Computation, Lecture Notes in Computer Science, Vol. 8402, pp. 229-247, Elsevier, 2014.
  18. D. Defour & M. Marin,  Regularity Versus Load-Balancing on GPU for Treefix Computations , ICCS, Lecture Notes in Computer Science, Vol. 18, pp. 309-318, Elsevier, 2013.
  19. D. Defour & E. Petit,  GPUburn: A system to test and mitigate GPU hardware failures , ICSAMOS, pp. 263-270, IEEE, 2013.
  20. M. G. Arnold, S. Collange & D. Defour,  Implementing LNS using filtering units of GPUs ,  ICASSP, pp. 1542-1545, IEEE, 2010.
  21. S. Collange, M. Daumas, D. Defour & David Parello,  Barra: A Parallel Functional Simulator for GPGPU , MASCOTS, pp. 351-360, IEEE, 2010.
  22. S. Collange, D. Defour & Y. Zhang,  Dynamic Detection of Uniform and Affine Vectors in GPGPU Computations , Euro-Par Workshops, LNCS, Vol. 6043, pp. 46-55, Springer, 2009.
  23. S. Collange, Y. S. Dandass, M. Daumas & D. Defour,  Using Graphics Processors for Parallelizing Hash-Based Data Carving , HICSS, pp. 1-10, IEEE Computer Society, 2009.
  24. S. Collange, D. Defour & A. Tisserand,  Power Consumption of GPUs from a Software Perspective , ICCS, LNCS, Vol. 5544, pp. 914-923, Springer, 2009.
  25. S. Collange, J. Flóres, & D. Defour,  A gpu interval library based on boost interval , Real Numbers and  Computers, RNC, pp. 61-72, July 2008.
  26. S. Collange, M. Daumas & D. Defour,  Graphics processors to speed-up simulations for the design of high performance solar receptors , IEEE 18th International Conference on Application-specific Systems (ASAP), Architectures and processors, pp. 377-382, 2007.
  27. G. Da Graca & D. Defour,  Implementation of float-float operators on graphics hardware , Real Numbers and  Computers, RNC, pp. 23-32, 2006. arXiv
  28. D. Defour,  Collapsing dependent floating-point operations , IMACS World Congress Scientific Computation, Applied Mathematics and Simulation, 2005
  29. C. Daramy, D. Defour, F. De Dinechin & J.-M. Muller,  CR-LIBM: a correctly rounded elementary function library , Proc. SPIE 5205, Advanced Signal Processing Algorithms, Architectures and Implementations XIII, 458, 2003
  30. D. Defour & F. de Dinechin,  Software Carry-Save: A Case Study for Instruction-Level Parallelism  Parallel Computing Technologies (7th PaCT’03), LNCS, Vol. 2763, pp. 207-214, Springer-Verlag (New York), September 2003.
  31. D. Defour & F. de Dinechin,  Software carry-save for fast multiple-precision algorithms , 35th International Congress of Mathematical Software, pp. 29-40, 2002.
  32. D. Defour, F. de Dinechin & J.-M. Muller,  A new scheme for table-based evaluation of functions , 36th Asilomar Conference on Signals, Systems, and Computers, pp. 1608-1613, 2001.
  33. D. Defour, P. Kornerup, J.-M. Muller & Nathalie Revol,  A new range reduction algorithm , 35th Asilomar Conference on Signals, Systems, and Computers, pp. 1656-1661, 2001.
  34. D. Defour & J. M. Muller,  Correctly Rounded Exponential Function in Double Precision Arithmetic , SPIE, 46th Annual Meeting International Symposium on Optical Science and Technology, pp. 156-157, 2001.

Articles in national journal

  1. D. Etiemble, D. Defour 2007-2017: les processeurs graphiques totalement programmables (GPU) , Techniques de l’Ingénieur, Réf : H1013 v1, Nov. 2018..
  2. D. Defour & M. Marin, Simulation temps réel de réseaux électriques
à l’aide des architectures multicœurs , UPVD Magazine Hors-Série recherche N°3, pp. 42-44, 2014.
  3. D. Defour & M. Marin,  Optimiser la représentation des flottants , HPC Magazine, Volume 4, pp. 65-70, 2013.
  4. S. Collange, M. Daumas & D. Defour,  État de l’intégration de la virgule flottante dans les processeurs graphiques , Technique et Science Informatiques, 27(6), pp. 719-733, 2008.
  5. B. Goossens & D. Defour,  Ordonnancement distribué d’instructions , Technique et Science Informatiques, 25(7), pp. 827-844, 2006.
  6. D. Defour & J.-M. Muller,  Évaluation de fonctions élémentaires , Réseaux et systèmes répartis, calculateurs parallèles, vol. 13, no. 4-5, pp. 449-465, 2001.

Communications in international conferences (summary)

  1. D. Defour, F. Fevotte, S. Graillat, F. Jezequel, W. Kirschenmann, J.-L. Lamotte, B. Lathuiliere, Y. Lhuiller, P. de Oliveira Castro, E. Petit, J. Signoles, D. Sohier & F. Vedrine, InterFLOP, Interoperable Tools for Computing, Debugging, Validation and Optimization of Floating-Point Programs , ISC 2021, ICS High Performance 2021, June 2021.
  2. R. Iakymchuk, D. Defour & S. Graillat, Toward fast, Accurate and Reproducible LU factorization , SCAN 2016, 17th CAMM-IMACS International Symposium on Scientific Computing, Computer Arithmetic, and Verified Numerical Computation, September 2016.
  3. R. Iakymchuk, D. Defour, S. Collange & S. Graillat, Reproducible and Accurate Algorithms for Numerical Linear Algebra ,  Workshop MS16: Numerical Reproducibility for High-Performance Computing at  SIAM Conference on Parallel Processing for Scientific Computing, PP16, April 2016.
  4. S. Collange, D. Defour, R. Iakymchuck & S. Graillat,   Reproducible and Accurate Matrix Multiplication for High-Performance , SCAN 2014, 16th CAMM-IMACS International Symposium on Scientific Computing, Computer Arithmetic, and Verified Numerical Computation, 2014.
  5. S. Collange, D. Defour, R. Iakymchuck & S. Graillat,   A reproducible Accurate Summation Algorithm for High-Performance Computing , SIAM EX14, SIAM Workshop on Exascale Applied mathematics Challenges and opportunites, 2014.
  6. D. Defour & M. Marin, “Real-time simulation of electrical network using multicore archtiectures”, Conférence DERBI 2012.

Communications in national conferences

  1. H. De Lassus Saint-Geniès, D. Defour & G. Revy,  Réduction d’argument basée sur les triplets pythagoriciens pour l’évaluation de fonctions trigonométriques , Conférence d’informatique en Parallélisme, Architecture et Système, Compas, June 2015. 
  2. D. Defour, [ Impact des ordonnanceurs sur la prédictibilité dans les GPU ], Conférence d’informatique en Parallélisme, Architecture et Système, 2014. 
  3. D. Defour & E. Petit,  Températures, erreurs matérielles et GPU , Conférence d’informatique en Parallélisme, Architecture et Système, 2013. 
  4. S. Collange, M. Daumas, D. Defour & D. Parello,  Étude comparée et simulation d’algorithmes de branchements pour le gpgpu , Symposium en Architectures nouvelles de machines, 2009.
  5. S. Collange, M. Daumas, D. Defour & R. Olivès,  Fonctions élémentaires sur GPU exploitant la localité de valeurs , Symposium en Architectures nouvelles de machines, 2008.
  6. M. Daumas, G. Da Graca, & D. Defour,  Caractéristiques arithmétiques des processeurs graphiques , Symposium en Architectures nouvelles de machines, 2006. arXiv
  7. B. Goossens & D. Defour, “Ordonnancement dynamique distribué”, Symposium en Architectures nouvelles de machines, 2005.

Unpublished research reports

  1. D. Defour, P. de Oliveira Castro, M. Istoan, E. Petit Shadow computation with BFloat16 to computenumerical accuracy Hal-03159965, 2021
  2. D. DefourImpacting predictability of GPU’s , Hal-951920, 2014.
  3. D. DefourAccuracy of maximum likelihood phylogeny reconstruction , Hal-726469, 2010
  4. D. Defour & B. Goossens,  Implémentation de l’opérateur add2 , Research Report 3, DALI Research Team, LP2A, 2004.
  5. F. de Dinechin, D. Defour & C. Lauter,  Fast correct rounding of elementary functions in double precision using double-extended arithmetic , Tech. Rep. RR-5137, INRIA, 2004.
  6. D. Defour,  Cache-optimised methods for the evaluation of elementary functions , Tech. Rep. RR2002-38, LIP, École Normale Supérieure de Lyon, 2002.
  7. D. Defour, F. de Dinechin & J.-M. Muller, “Proof of correct rounding for the exponential function”, Tech. Rep. RR2003-37, LIP, École Normale Supérieure de Lyon, 2002.

Thesis

  1. D. Defour, Contribution au calcul sur GPU:  considérations arithmétiques et architecturales , Habilitation de l’Université de Perpignan Via Domitia, 2014.
  2. D. Defour,  Fonction élémentaires : algorithmes et implémentations efficaces pour l’arrondi correct en double précision , Thèse de doctorat de l’Ecole Normale Supérieure de Lyon, 2003.