Training

Find training resources for using NREL’s high-performance computing (HPC) systems as well as related online tutorials.

Schedule of Office Hours and Training Events

Tutorials and Guides

Please see our GitHub repository for cloneable walkthroughs and examples you can follow along within your shell.

For additional provisional documentation on systems, see our Github.io site.

NREL HPC User GitHub Repository

The GitHub repository features more tips and tricks for developing effective workflows on HPC systems. Users are welcomed and encouraged to contribute information, scripts, and any other tools they think will benefit the whole community.

View WIKI

Past Workshops and Training Events

This workshop covered the following features, which require a more involved awareness of Slurm's functionality and how it is configured in our environment:

  • Job monitoring and forensics: usage examples on sreport, sacct, sinfo, and sview (FastX)
  • Advanced srun and sbatch functions (flags)
  • Parallelizing with SLURM
  • Remote exclusive GPU usage, requesting GPU nodes.

The resources used during this presentation are available here:

Presentation Slides

This workshop covered the following features, which are new to the NREL HPC workflow relative to what was possible on Peregrine and its job scheduler:

  • Basic Slurm core functionality overview
  • Slurm partitions - request by features
  • Job dependencies and job arrays
  • Job steps
  • Job monitoring and basic troubleshooting. 

The resources used during this presentation are available here:

Presentation Slides

This workshop detailed changes, advice, and caveats for acclimating to Eagle relative to what was standard on Peregrine. The resources used during this presentation are available here:

Presentation Slides

Separate instructions for how to use Globus to migrate files quickly and reliably

PBS to Slurm Analogous Command Cheat Sheet


Share