Argonne National Laboratory

Article | Mathematics and Computer Science

MCS staff play key role in Argonne high-performance training program
Researchers in the Mathematics and Computer Science (MCS) division at Argonne National Laboratory played a major role in the 2022 Argonne Training Program on Extreme-Scale Computing (ATPESC).

The program offers early career computational scientists two weeks of in-depth instruction on high-performance computing.

In this year’s program, held July 31 to August 12, 2022, MCS division staff gave numerous presentations and led hands-on sessions on key skills and tools for designing, implementing and executing applications on high-end computing systems.

Following is a list of the activities MCS researchers participated in at this year’s training program. For the full ATPESC 2022 agenda, see the website: https://​extreme​com​put​ing​train​ing​.anl​.gov/​a​g​e​n​d​a​-​2022/.

Track 2 B – Programming Models and Languages – MPI

  • Derived datatypes and RMA -- Ken Raffenetti
  • Hybrid programming with threads and GPUs and what’s new in MPI-4 – Yanfei Guo

Track 3 – I/O

  • Introduction; principles of HPC – Phil Carns
  • Darshan introduction; understanding and tuning performance – Shane Snyder
  • MPI/IO and Parallel netCDF – Rob Latham

Track 5 – Numerical Algorithms and Software for Extreme-Scale Science

  • Nonlinear solvers (with PETSc) – Richard Tran Mills
  • Optimization (with TAO) – Todd Munson
  • Panel Discussion:  Extreme-scale numerical algorithms and software
  • Moderator – Richard Tran Mills; panelist – Todd Munson

Track 7 – Software Productivity and Sustainability

  • Scientific software design – Anshu Dubey
  • Lab notebooks for computational mathematics, sciences, and engineering – Jared O’Neal
  • Managing computational experiments – Anshu Dubey and Jared O’Neal

Track 8 – Machine Learning

  • Deep learning methods – Tanwi Mallick
  • Scientific machine learning for dynamical systems – Romit Maulik