Presentations

Various presentations from conferences, workshops, and other events regarding the NLAFET project and its results are published here.

Overview presentation of the NLAFET project at conferences / workshops

Title of presentation: NLAFET – Parallel Numerical Linear Algebra for Future Extreme-Scale Systems

Besides the named presenters below, several members of the NLAFET consortium team contributed.

  • EXDCI HPC Workshop, Rome, September 29-20. Presentation by Bo Kågström
  • Swedish e-Science Academy, October 14-15, 2015, Arlandastad. Invited presentation by Bo Kågström
  • SIAM Conference on Parallel Processing for Scientific Computing, April 12-15, 2016, Paris. Poster presentation by Bo Kågström, Lennart Edblom, Lars Karlsson, Laura Grigori, Iain Duff, Jonathan Hogg, Jack Dongarra and Nick Higham
  • EASC2016- Exascale Applications & Software Conference, April 24-29, 2016, Stockholm. Presentation by Bo Kågström
  • European HPC Summit week: EXDCI Workshop, May 9-10, 2016, Prague, Czech Republic. Presentation by Bo Kågström and Lennart Edblom
  • Workshop on Batched, Reproducible, and Reduced Precision BLAS, May 18-19, 2016, Univ. of Tennessee, Knoxville, TN. Invited presentation by Bo Kågström
  • IFIP WG 2.5 Workshop on Numerical Software and Scientific Computing, August 4, 2016, Oxford, UK. Invited presentation by Iain Duff

Presentations etc. related to NLAFET, describing NLAFET work

2016

  • SIAM Conference on Parallel Processing for Scientific Computing, April 12-15, 2016, Paris
    Organizing Committee Co-Chair: Laura Grigori, Inria and Université Pierre et Marie Curie, France
    Invited presentation: Scalability of Sparse Direct Codes, Iain Duff, Science & Technology
    Facilities Council, United Kingdom and CERFACS, Toulouse, France
    Minisymposia organization MS60: Advances in Parallel Dense and Sparse Linear Algebra, Jonathan Hogg, Rutherford Appleton Laboratory, United Kingdom and Lars Karlsson, Umeå University, Sweden
    Minisymposia MS60 presentations:

    • The Challenge of Strong Scaling for Direct Methods, Jonathan Hogg, Rutherford Appleton Laboratory, United Kingdom
    • NUMA-Aware Hessenberg Reduction in the Context of the QR Algorithm, Mahmoud Eljammaly, Lars Karlsson, and Bo Kågström, Umeå University, Sweden
    • Low Rank Approximation of Sparse Matrices for LU Factorization Based on Row and Column Permutations, Laura Grigori and Sebastien Cayrols, Inria, France; James W. Demmel, Univ. of California, Berkeley, USA
    • Assessing Recent Sparse Matrix Storage Formats for Parallel Sparse Matrix-Vector Multiplication, Weifeng Liu, Univ. of Copenhagen, Denmark

    Minisymposia MS4 presentation: Enlarged GMRES for Reducing Communication When Solving Reservoir Simulation Problems, Hussam Al Daas, UPMC-Inria-TOTAL, France; Laura Grigori, Inria, France; Pascal Henon, Total E&P, France; Philippe Ricoux, TOTAL SA, France
    Poster presentation: A Standard for a Batched BLAS routines, Pedro Valero-Lara, Univ. of Manchester, UK; Jack J. Dongarra and UK; Stanimire Tomov, Univ. of Tennessee, Knoxville, USA; Mawussi Zounon, Univ. of Manchester, UK

  • Workshop on Batched, Reproducible, and Reduced Precision BLAS, May 18-19, 2016, Univ. of Tennessee, Knoxville, TN
    • Invited seminar: Scalability of sparse direct codes. Iain Duff, Science & Technology Facilities Council, United Kingdom
  • International Supercomputing (ICS) 2016, June 23, 2016, Frankfurt, Germany. Workshop presentation: Form Follows Function – Software, Algorithms & Application
    Challenges, Jack Dongarra, Univ. of Tennessee, Oak Ridge National Lab, and Univ. of
    Manchester
  • 9th International Workshop on Parallel Matrix Algorithms and Applications (PMAA16), July 6-8, 2016, Bordeaux, France
    Invited presentation: Current Trends in High-Performance Computing and Challenges for the Future, Jack Dongarra, Univ. of Tennessee, Oak Ridge National Lab, and Univ. of Manchester
    Minisymposia presentations – Task-based scientific library on top of runtime system

    • Task-based sparse Cholesky solver on top of runtime system, Florent Lopez, Numerical Analysis Group – STFC Rutherford Appleton Laboratory, UK
    • Towards Highly Parallel and Compute-bound Computation of Selected Eigen-vectors given a Matrix in Schur Form, Lars Karlsson*, Björn Adlerborn, Mahmoud Eljammaly, Carl Christian Kjelgaard Mikkelsen, and Bo Kågström, Umeå University
  • 2016 SIAM Annual Meeting, July 10-15, 2016, Boston, USA
    Minisymposia MS73 – Numerical Linear and Multilinear Algebra, Revisit to the GEMM-Based Level 3 BLAS and Its Impact on High Performance Matrix Computations, Bo Kågström, Umeå University
  • International Congress on Mathematical Software, July 11, 2016, Berlin, Germany
    Invited presentation: An Overview of High-Performance Computing and Changing Rules at Exascale, Jack Dongarra, Univ. of Tennessee, Oak Ridge National Lab, and Univ. of Manchester
  • IMA Conference on Numerical Linear Algebra and Optimization, September 7-9, 2016, Birmingham, UK
  • Invited plenary presentation: Enlarged Krylov subspace solvers for reducing communication, L. Grigori, Inria France
    • A Block Krylov Method for the Frechet Derivative of f(A)b. Samuel Relton, Univ. of Manchester and P. Kandolf, Univ. of Innsbruck.
    • Numerical fault tolerant strategies for resilient parallel eigensolvers. Mawussi Zounon, Emmanuel Agullo, Luc Giraud and Pablo Salas.
  • Workshop on Sustainable Software for Science: Practice and Experiences, Sept 12-14, 2016, Manchester, UK.
    • Creating a standardized set of batched BLAS routines. Jack Dongarra, Sven Hammerling, Nicholas J. Higham, Samuel D. Relton, Pedro Valero-Lara, and Mawussi Zounon.
  • Eleventh Workshop on Mathematical Modeling of Environmental and Life Sciences Problems, Constanta, Romania. 12-16 October 2016.
    • Invited presentation: Highly scalable solution of large sparse equations. Iain Duff, STFC.
  • The Sixth International Conference on Numerical Algebra and Scientific Computing, Zhejiang University, Hangzhou, China. 22-26 October, 2016. Co-chairman of ANA Prize Committee: Iain Duff, STFC
    • Plenary speaker: Highly scalable solution of large sparse equations. Iain Duff, STFC.

2017

  • The Fourth International Conference on Numerical Analysis and Optimization. Sultan Qaboos University, Muscat, Oman. 4-6 January 2017.
    • Plenary invited speaker: Sparse direct solution on parallel computers, Iain Duff, STFC
  • Bath-RAL Numerical Analysis Day. University of Bath, UK. 27th January 2017.
    • Presentation: Experiments with a sparse Cholesky solver using runtime systems, Florent Lopez, STFC
  • International Conference on Domain Decomposition Methods DD XXIV, February 6-10 2017, Longyearbyen, Norway.
    • Invited plenary talk: Communication avoiding iterative solvers and preconditioning, Laura Grigori, Inria, France
    • Minisymposia MS01-3: Adaptive enlarged Krylov Conjugate Gradient, Laura Grigori and Olivier Tissot
  • Second Workshop on Batched, Reproducible, and Reduced Precision BLAS, February 23-25, 2017, Georgia Tech, Atlanta
    • Workshop presentation: Standardizing the Batched BLAS API and Memory Layout, Jack Dongarra, Nick Higham, Samuel Relton, and Mawussi Zounon, Univ. of Manchester
    • Invited presentation: The NLAFET Project: Overview and Status Report, Bo Kågström, Umeå University
  • 2017 SIAM Conference on Computational Science and Engineering, February 27 – March 3, 2017, Atlanta, USA
    • Minisymposia organization MS64: Batched Linear Algebra on Multi / Many-Core Architectures. Mawussi Zounon and Pedro Valero-Lara, Univ. of Manchester.
    • Minisymposia MS64 presentation:
      • Recent Advances in Batched Linear Algebra Computation. Samuel Relton, Mawussi Zounon, and Pedro Valero-Lara, Univ. of Manchester, UK.
    • Minisymposia organization MS69: Recent Advances in Matrix Functions and Applications. Samuel Relton, University of Manchester.
    • Minisymposia organization MS101: Parallel Numerical Linear Algebra for Extreme Scale Systems: Nicholas Higham, Univ. of Manchester and Bo Kågström, Umeå University.
    • Minisymposia MS101 presentations:
      • The Design and Implementation of a Dense Linear Algebra Library for Extreme Parallel Computers. Jack Dongarra, Nick Higham, Samuel Relton, and Mawussi Zounon, Univ. of Manchester, UK.
      • Computing the Low Rank Approximation of a Sparse Matrix. Laura Grigori, Alan Ayala, and Sebastien Cayrols, Inria, France; James W. Demmel, University of California, Berkeley, USA.
      • Sparse Direct Solvers for Extreme Scale Computing. Iain Duff, Jonathan Hogg, Florent Lopez, and Stojce Nakov, Rutherford Appleton Laboratory, STFC.
      • Extreme-Scale Eigenvalue Reordering in the Real Schur Form. Carl Christian Kjelgaard Mikkelsen, Mirko Myllykoski, Lars Karlsson and Bo Kågström, Umeå University.
  • PARNUM 2017 Workshop, Waischenfeld, Germany. 19-21 April 2017.
    • Keynote speaker: Direct solution of sparse linear equations on parallel computers. Iain Duff, STFC, UK.
  • Swedish e-Science Academy 2017, 11-12 October 2017, Umeå University, Sweden
    • Chair of first session and welcome presentation, Bo Kågström, Umeå University, Sweden
    • Poster presentations:
      • A Task-Based Algorithm for Reordering the Eigenvalues of a Matrix in Real Schur Form, Mirko Myllykoski, Umeå University
      • Blocked and Robust solution of triangular systems, Carl Christian K. Mikkelsen and Lars Karlsson, Umeå University
      • An Auto-tuning Framework for a NUMA-Aware Hessenberg Reduction Algorithm, Mahmoud Eljammaly, Lars Karlsson and Bo Kågström, Umeå University
  • PPAM 2017, September 10-13, 2017, Lublin, Poland.
    • Keynote speaker: An Overview of High Performance and a Look at Energy Saving on the Intel Knights Landing for Linear Algebra Computations, Jack Dongarra, Univ. of Manchester
    • Presentations:
      • On the Tunability of a New Hessenberg Reduction Algorithm Using Parallel Cache Assignment – Mahmoud Eljammaly, Lars Karlsson and Bo Kågström, Umeå University
      • Blocked Algorithms for Robust Solution of Triangular Linear Systems – Carl Christian Kjelgaard Mikkelsen and Lars Karlsson, Umeå University
      • A Task-Based Algorithm for Reordering the Eigenvalues of a Matrix in Real Schur Form – Mirko Myllykoski, Umeå University
      • Experiments with sparse Cholesky using a parametrized task graph implementation – Florent Lopez and Iain Duff, Rutherford Appleton Laboratory, STFC.

2018

  • SIAM PP18, March 7-10, 2018, Waseda University, Tokyo, Japan.
    • Minisymposia MS43 organization: Parallel Numerical Linear Algebra for Future Extreme-Scale Systems – Part I of II: Iain Duff, Science & Technology Facilities Council, United Kingdom & CERFACS, Toulouse, France and Bo T. Kågström, Umeå University, Sweden
    • Minisymposia MS43 presentations:
      • Dense Linear Systems for Extreme Scale, Jack J. Dongarra, University of Tennessee, ORNL, and University of Manchester, USA
      • The Batched Blas, Sven J. Hammarling and Mawussi Zounon, The University of Manchester, UK
      • Task Based Robust Computation of Eigenvectors, Carl Christian Kjelgaard Mikkelsen, Lars Karlsson, Mirko Myllykoski, and Angelika Schwarz, Umeå University, Sweden
    • Minisymposia MS55 organization: Parallel Numerical Linear Algebra for Future Extreme-Scale Systems – Part II of II: Iain Duff, Science & Technology Facilities Council, United Kingdom & CERFACS, Toulouse, France and Bo T. Kågström, Umeå University, Sweden
    • Minisymposia MS55 presentations:
      • Sparse Solvers on Extreme-Scale Systems, Iain Duff, Science & Technology Facilities Council, United Kingdom and CERFACS, Toulouse, France
      • Task-Based Sparse Direct Solver for Symmetric Indefinite Systems, Florent Lopez, Rutherford Appleton Laboratory, United Kingdom; Iain Duff, Science & Technology Facilities Council, United Kingdom and CERFACS, Toulouse, France
      • Robust Algebraic Preconditioners for Large Scale Applications, Laura Grigori, Simplice Donfack, and Olivier Tissot, Inria, France
      • Enlarged Conjugate Gradient Method for Reducing Communication, Olivier Tissot and Laura Grigori, Inria, France
  • Posters:
    • Efficient Robust Multi-Shift Triangular Solves, Angelika Schwarz, Mirko Myllykoski, Carl Christian Kjelgaard Mikkelsen, and Lars Karlsson, Umeå University, Sweden

Leave a Reply