Azzam HAIDAR

Research Director,
INNOVATIVE COMPUTING LABORATORY
@
The University of Tennessee
Department of Electrical Engineering
and Computer Science

I can be reached at

Mail: 1122 Volunteer Blvd
203 Claxton Complex
37996-3450 TN, USA
Phone: (+1) 865- 974-9308
Fax : (+1) 865- 974-8296
E-mail: haidar@icl.utk.edu





Outstanding Research Award and Honors:

  • Nvidia CUDA Centers of Excellence (CCOEs 2012-2014)

  • Intel Parallel Computing Center (IPCC 2014, 2015)

  • ACTAPACK-ACcelerated Tensor Algebra PACKage with Lawrence Livermore National Laboratory

  • Best paper finalist SC-2015 IEEE International Conference for High Performance Computing, Networking, Storage and Analysis

  • Best paper awards from 2015 IEEE High Performance Extreme Computing Conference

  • Best paper awards from 2015 High Performance Computing Symposium

  • Best paper awards from 2014 IEEE IPDPS/PDSEC 2014, International Parallel and Distributed Processing Symposium

  • Lauréat du label de compétences C3I (Certificat de Compétences en Calcul Intensif) HPC certification Laureate

  • Invited plenary speaker to 2009 International Conference On Preconditioning Techniques

  • Best work awards from 2007 International Conference On Preconditioning Techniques

    • Publications:

  • R. Solca, A. Kozhevnikov, A. Haidar, S. Tomov and T. Schulthess.
    Towards extreme-scale all-electron quantum simulations for materials design.
    The International Conference for High Performance Computing, Networking, Storage and Analysis, IEEE-SC 2015. Austin, TX, USA.
    Best paper finalist decision in November 2015

  • A. Haidar, P. Luszczek and J. Dongarra.
    Performance Engineering and Modeling for Compute- and Bandwidth-Bound Kernels.
    In preparation for the SIAM Journal on Scientific Computing (SISC).

  • A. Haidar, S. Tomov, P. Luszczek, and J. Dongarra.
    MAGMA Embedded: Towards a Dense Linear Algebra Library for Energy Efficient Extreme Computing
    IEEE High Performance Extreme Computing Conference IEEE-HPEC 2015, Waltham, MA USA.
    Best paper awards

  • A. Haidar, T. Dong, S. Tomov, P. Luszczek, and J. Dongarra.
    A Framework for Batched and GPU-resident Factorization Algorithms Applied to Block Householder Transformations
    International SuperComputing Conference IEEE-ISC 2015, Frankfurt, Germany.
    Springer, Lecture Notes in Computer Science Volume 9137, 2015, pp 31-47

  • K. Kabir, A. Haidar, S. Tomov, and J. Dongarra
    On the Design, Development and Analysis of Optimized Matrix-Vector Multiplication Routines for coprocessors.
    International SuperComputing Conference IEEE-ISC 2015, Frankfurt, Germany
    Springer, Lecture Notes in Computer Science Volume 9137, 2015, pp 58-73

  • A. Haidar, A. YarKhan, C. Cao, P. Luszczek, S. Tomov, and J. Dongarra.
    Flexible Linear Algebra Development and Scheduling with Cholesky Factorization.
    17th IEEE International Conference on High Performance Computing and Communications, IEEE HPCC 2015, New York, USA.

  • K. Kabir, A. Haidar, S. Tomov, and J. Dongarra
    Performance Analysis and Optimisation of Two-Sided Factorization Algorithms for Heterogeneous Platform
    International Conference on Computational Science ICCS 2015, Reykjavík, Iceland.
    Elsevier, Procedia Computer Science, Volume 51, 2015, Pages 180¿190

  • A. Haidar, T. Dong, S. Tomov, P. Luszczek, and J. Dongarra.
    Optimization for Performance and Energy for Batched Matrix Computations on GPUs
    20th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, Workshop on General Purpose Processing Using GPUs. GPGPU/PPoPP 2015, San Francisco, CA, USA, Pages 59-69, 2015

  • K. Kabir, A. Haidar, S. Tomov, and J. Dongarra
    Performance Analysis and Design of a Hessenberg Reduction using Stabilized Blocked Elementary Transformations for New Architectures.
    23rd High Performance Computing Symposium HPC 2015, Spring Simulation Multi-Conference. Alexandria, VA, USA.
    Best paper award.

  • G. Pichon, A. Haidar, M. Faverge, J. Kurzak
    Divide and Conquer Symmetric Tridiagonal Eigensolver for Multicore Architectures
    IEEE IPDPS 2015, 29th IEEE International Parallel and Distributed Processing Symposium, Hyderabad, India, Pages 51-60.

  • J. Dongarra, M. Gates, A. Haidar, Y. Jia, K. Kabir, P. Luszczek and S. Tomov.
    HPC Programming on Intel Many-Integrated-Core Hardware with MAGMA Port to Xeon Phi.
    Scientific Programming, vol. 2015, Article ID 502593, 11 pages. doi:10.1155/2015/502593

  • J. Dongarra, M. Gates, A. Haidar, J. Kurzak, P. Luszczek, S. Tomov, and I. Yamazaki
    Accelerating Numerical Dense Linear Algebra Calculations with GPUs
    Chapter in Numerical Computations with GPUs Springer 2014, 2014, pp 3-28

  • J. Dongarra, A. Haidar, J. Kurzak, P. Luszczek, S. Tomov, and A. YarKhan
    Model-Driven One-Sided Factorizations on Multicore Accelerated Systems.
    Supercomputing frontiers and innovations journal 2014, Vol 1, no 1, pp 85-115.

  • T. Dong, A. Haidar, P. Luszczek, J. Harris, S. Tomov, and J. Dongarra.
    Towards Batched Linear Solvers on Accelerated Hardware Platforms
    20th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming PPoPP 2015, San Francisco, CA, USA, Pages 261-262

  • A. Haidar, T. Dong, P. Luszczek, S. Tomov and J. Dongarra.
    Batched Matrix Computations on Hardware Accelerators Based on GPUs.
    International Journal of High Performance Computing Applications, 2015 vol. 29 no. 2 pp. 193-208

  • C. Cao, M. Gates, A. Haidar, P. Luszczek, S. Tomov, and I. Yamazaki.
    Performance and Portability with OpenCL for Throughput-Oriented HPC Workloads Across Accelerators, Coprocessors, and Multicore Processors.
    ScalA '14 the 5th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems, IEEE-SC 2014, New Orleans, LA, USA, Pages 61-68.

  • A. Haidar, P. Luszczek and J. Dongarra.
    New Algorithm for Computing Eigenvectors of the Symmetric Eigenvalue problem.
    IEEE IPDPS/PDSEC 2014, 28th IEEE International Parallel and Distributed Processing Symposium, Phoenix, Arizona USA, Pages 1150-1159
    Best paper award.

  • T. Dong, A. Haidar, S. Tomov, and J. Dongarra.
    A Fast Batched Cholesky Factorization on a GPU
    ICPP 2014, The 43rd International Conference on Parallel Processing, Minneapolis, USA, Pages 432-440

  • T. Dong, A. Haidar, P. Luszczek, J. Harris, S. Tomov, and J. Dongarra.
    LU Factorization of Small Matrices: Accelerating Batched DGETRF on the GPU.
    HPCC 2014, The 16th IEEE International Conferences on high performance computing and communications, Paris, France, Pages 157-160.

  • A. Haidar, C. Cao, A. YarKhan, P. Luszczek, S. Tomov, K. Kabir, and J. Dongarra.
    Unified Development for Mixed Multi-GPU and Multi-Coprocessor Environments using a Lightweight Runtime Environment.
    IEEE IPDPS 2014, 28th IEEE International Parallel and Distributed Processing Symposium, Phoenix, Arizona USA, Pages 491-500.

  • M. Gates, A. Haidar, and J. Dongarra.
    Accelerating computation of eigenvectors in the nonsymmetric eigenvalue problem.
    VecPar 2014 the 11th International Meeting High Performance Computing for Computational Science, Eugene, Oregon, USA.
    Springer, Lecture Notes in Computer Science Volume 8969, 2015, pp 182-191

  • A. Haidar, P. Luszczek, S. Tomov, and J. Dongarra.
    Heterogenous Acceleration for Linear Algebra in Mulit-Coprocessor Environments.
    VecPar 2014 the 11th International Meeting High Performance Computing for Computational Science, Eugene, Oregon, USA.
    Springer, Lecture Notes in Computer Science Volume 8969, 2015, pp 31-42

  • A. Haidar, P. Luszczek, J. Kurzak and J. Dongarra.
    An Improved Parallel Singular Value Algorithm and Its Implementation for Multicore Hardware.
    International Conference for High Performance Computing, Networking, Storage and Analysis, IEEE-SC 2013, Denver, CO, USA.

  • A. Haidar, R. Solca, M. Gates, S. Tomov, T. Schulthess and J. Dongarra.
    Leading edge multi-GPU algorithms for generalized eigenproblems for electronic structure calculations.
    International Supercomputing Conference IEEE-ISC 2013, Leipzig, Germany.
    Springer, Lecture Notes in Computer Science Volume 7905, 2013, pp 67-80

  • A. Haidar, M. Gates, S. Tomov, and J. Dongarra.
    Toward a scalable multi-GPU eigensolver via compute-intensive kernels and efficient communication.
    International Conference on Supercomputing, ICS 2013, Eugene, Oregon, USA Pages 223-232

  • J. Dongarra, M. Gates, A. Haidar, Y. Jia, K. Kabir, P. Luszczek and S. Tomov.
    Portable HPC Programming on Intel Many-Integrated-Core Hardware with MAGMA Port to Xeon Phi.
    International Conference on Parallel Processing and Applied Mathematics, 2013, Warsaw, Poland.
    Springer, Lecture Notes in Computer Science Volume 8384, pp. 571¿581.

  • A. Haidar, R. Solca, S. Tomov, I. Yamazaki, T. Dong, T. Schulthess and J. Dongarra.
    MAGMA: a breakthrough in solvers for eigenvalue problems committee Proposal for the Nvidia CCOEs award competition 2012.

  • A. Haidar, R. Solca, S. Tomov, T. Schulthess and J. Dongarra.
    A novel hybrid CPU-GPU generalized eigensolver for electronic structure calculations based on fine-grained memory aware tasks.
    International Journal of High Performance Computing Applications. May 2014 vol. 28 no. 2 196-209

  • H. Ltaief, P. Luszczek, A. Haidar and J. Dongarra.
    Solving the Generalized Symmetric Eigenvalue Problem using Tile Algorithms on Multicore Architectures.
    Advances in Parallel Computing Volume 22, pp 397-404, 2012

  • A. Haidar, H. Ltaief, P. Luszczek and J. Dongarra.
    A Comprehensive Study of Task Coalescing for Selecting Parallelism Granularity in a Two-Stage Bidiagonal Reduction A Comprehensive Study of Task Coalescing for Selecting Parallelism Granularity in a Two-Stage Bidiagonal Reduction.
    IEEE IPDPS 2012, 26th IEEE International Parallel and Distributed Processing Symposium, Shanghai, China, Pages 25-35

  • A. Haidar, H. Ltaief and J. Dongarra
    Toward High Performance Divide and Conquer Eigensolver for Dense Symmetric Matrices.
    SIAM Journal on Scientific Computing (SISC) 2012, Volume 34, Issue 6, pp. 249¿274
    Keywords: symmetric eigensolver, eigenvalue, band matrices, bulgechasing, tile algorithms, dynamic scheduling.

  • A. Haidar, H. Ltaief and J. Dongarra.
    Parallel Reduction to Condensed Forms for Symmetric Eigenvalue Problems using Aggregated Fine-Grained and Memory-Aware Kernels.
    International Conference for High Performance Computing, Networking, Storage and Analysis, IEEE-SC 2011, Seattle, WA, USA.

  • G. Bosilca, A. Bouteiller, A. Danalis, M. Faverge, A. Haidar, T. Herault, J. Kurzak, J. Langou, P. Lemarinier, H. Ltaief, P. Luszczek, A. YarKhan, J. Dongarra
    Flexible Development of Dense Linear Algebra Algorithms on Massively Parallel Architectures with DPLASMA
    IEEE IPDPS/PDSEC 2011, 25th IEEE International Parallel & Distributed Processing Symposium, Vol., no., pp.1432, 1441, 16-20 May 2011

  • A. Haidar, H. Ltaief, A. YarKhan, J. Dongarra,
    Analysis of Dynamically Scheduled Tile Algorithms for Dense Linear Algebra on Multicore Architectures,
    Concurrency and Computations: Practice and Experience, Volume 24 Issue 3, Pages 305-321, March 2011
    Keywords:dynamic scheduler, dense linear algebra kernels, multicore architecture, parallel algorithms.

  • F. Sourbier, A. Haidar, L. Giraud, H. Ben HadjAli, S. Operto and J. Virieux
    Three-dimensional parallel frequency-domain visco-acoustic wave modelling based on a hybrid direct/iterative solver
    Geophysical Prospecting journal 2011, Vol 59, pp. 834¿856
    Keywords: seismic modeling, Helmholtz equation, Schur complement, parallel computing, massively parallel algorithms.

  • J. Virieux, S. Operto, H. Ben Hadj Ali, R. Brossier, V. Etienne, F. Sourbier, L. Giraud, A. Haidar,
    Seismic wave modeling for seismic imaging,
    The Leading Edge SEG (Society of Exploration Geophysicists) Volume 28 (5), p.538-544, March 2009.

  • F. Sourbier, A. Haidar, L. Giraud, S. Operto, and J. Virieux,
    Frequency-domain full-waveform modeling using a hybrid direct-iterative solver based on a domain decomposition method: a tool for 3D full-waveform inversion?
    SEG (Society of Exploration Geophysicists) p. 2147-2151, September 2008.

  • F. Sourbier, A. Haidar, L. Giraud, R. Brossier, S. Operto, and J. Virieux,
    Combining direct and iterative solvers for improving efficiency of solving wave equations when considering multi-sources problems,
    EAGE (European Association for Geoscientists and Engineers) workshop on High Performance Computing, W08, June 2008.

  • E. Agullo, L. Giraud, A. Guermouche, A. Haidar, J. Roman,
    Parallel algebraic domain decomposition solver for the solution of augmented systems.
    Special issue of Advances in Engineering Software, 2013, Volumes 60¿61, Pages 23¿30

  • L. Giraud, A. Haidar and S. Pralet,
    Using multiple levels of parallelism to enhance the performance of hybrid linear solvers,
    Parallel Computing vol. 36, p 285-296, 2010.
    Keywords:multiple levels of parallelism, parallel computing, hybrid linear solvers, structural mechanics problems.

  • L. Giraud, A. Haidar and Y. Saad,
    Sparse approximations of the Schur complement for parallel algebraic hybrid solvers in 3D
    Numerical Mathematics vol 3, p. 276-294, 2010.
    Keywords:sparse approximations, , hybrid linear solvers, structural mechanics problems.

  • L. Giraud and A. Haidar,
    Parallel algebraic hybrid solvers for large 3D convection-diffusion problems,
    Numerical Algorithms, vol. 51, p. 151-177, 2009.

  • A. Haidar,
    On the parallel scalability of hybrid linear solvers for large 3D problems,
    PhD thesis 2008 http://www.cerfacs.fr/

  • L. Giraud, A. Haidar, and L. T. Watson,
    Parallel scalability study of hybrid preconditioners in three dimensions,
    Parallel Computing: vol. 34, p. 363-379, 2008.
    Keywords:Large sparse, linear systems, sparse direct solvers, iterative methods, preconditioned Krylov solvers, mixed arithmetic algorithms, parallel computing, convection-diffusion equations.

  • L. Giraud, A. Haidar, and L. T. Watson,
    Mixed-precision preconditioners in parallel domain decomposition solvers.,
    Springer, Lectures notes computational science and engineering vol. 60, p.357-364, 2008.
    Keywords:Large sparse, linear systems, sparse direct solvers, iterative methods, preconditioned Krylov solvers, mixed arithmetic algorithms, parallel computing, convection-diffusion equations.

  • A. Haidar, and J. Erhel.
    Transport réactif en milieux poreux
    Master Thesis 2004
    Keywords: Non-linear systems, reactive transport modelling,homogeneous and heterogeneous porous media, convection diffusion, global coupling algorithm.

  • L. Giraud, A. Haidar and Y. Saad,
    Sparse approximations of the Schur complement for parallel algebraic hybrid solvers in 3D
    Numerical Mathematics vol 3, p. 276-294, 2010.

  • E. Agullo, L. Giraud, A. Guermouche, A. Haidar, J. Roman,
    Parallel algebraic domain decomposition solver for the solution of augmented systems,
    Parallel, Distributed, Grid and Cloud Computing for Engineering, 12-15 April 2011.

  • E. Agullo, L. Giraud, A. Guermouche, A. Haidar, S. Lanteri, J. Roman,
    Algebraic Schwarz Preconditioning for the Schur Complement: Application to the Time-Harmonic Maxwell Equations Discretized by a Discontinuous Galerkin Method.
    20th International Conference on Domain Decomposition Methods, UC San Diego, in La Jolla, California, February 7-11, 2011.

  • G. Bosilca, A. Bouteiller, A. Danalis, M. Faverge, A. Haidar, T. Herault, J. Kurzak, J. Langou, P. Lemariner, H. Ltaief, P. Luszczek, A. YarKhan, J. Dongarra
    Distibuted Dense Numerical Linear Algebra Algorithms on Massively Parallel Architectures: DPLASMA.
    12th IEEE International Workshop on Parallel and Distributed Scientific and Engineering Computing (PDSEC-11), May 16-20, 2011, Anchorage, Alaska, USA.

  • E. Agullo, L. Giraud, A. Guermouche, A. Haidar, J. Roman,
    Towards a Complexity Analysis of Sparse Hybrid Linear Solvers,
    PARA 2010, Reykjavik, Iceland, June 6-9, 2010

  • F. Sourbier, A. Haidar, L. Giraud, S. Operto, and J. Virieux,
    Frequency-domain acoustic wave modeling using a hybrid direct-iterative solver based on a domain decomposition method,
    EAGE (European Association for Geoscientists and Engineers), H036, June 2008.