ATPESC provides intensive, two-week training on the key skills, approaches and tools to design, implement and execute computational science and engineering applications on current and next-generation supercomputers. The program is an annual event organized by Argonne National Laboratory and funded by the US. Department of Energy Exascale Computing Project. This year’s program was, for the first time, conducted virtually.

Following is a list of the MCS researchers and their contributions to this year’s virtual ATPESC conference:

Hardware Architectures

• Pete Beckman – Introduction to Track 1 – Hardware architectures
• Ken Raffenetti – Interconnects

Programming Models and Languages

• Yanfei Guo – Hands-on: Basic stencil example; Derived datatypes and hands-on; Hybrid programming with GPUs and hands-on
• Pavan Balaji – Introduction to RMA; RMA and hands-on; track wrap-up

Data-Intensive Computing and I/O

• Phil Carns and Rob Latham – Welcome and introduction; Discussion forum and Q/A
• Phil Carns – HPC I/O principles; Introduction to Darshan          
• Rob Latham – MPI-I/O; Parallel NetCDF
• Shane Snyder – How to understand and tune performance

Numerical Algorithms and Software for Extreme-Scale Science

• Lois Curfman McInnes – Introduction to numerical libraries and virtual logistics
• Richard Mills – Contributing to the numerical package community (panelist); Nonlinear solvers (with PETSc)
• Alp Dener – Optimization (with TAO); Optimization

Software Productivity and Sustainability

• Anshu Dubey – Software design; Software testing; Refactoring

Machine Learning

• Prasanna Balaprakash – Hyperameter optimization