Dr. Alex Morgan

All writings

Five Habits for Reproducible Simulations

· 1 min read

Simulations are experiments, and experiments should be reproducible. These habits have saved my group countless hours:

  • Version-control input files alongside the code.
  • Record the exact compiler, library versions and hardware.
  • Store a small “smoke test” case that runs in under a minute.
  • Script every post-processing step — no hand-edited plots.
  • Archive raw data with a DOI when publishing.
# Example: capture the environment next to each run
module list 2>&1 | tee run/env.txt
git rev-parse HEAD > run/commit.txt

Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim.

Ask the assistant

Powered by AI, may be inaccurate.

Hi! I can answer questions about this researcher's work, explain their papers and projects in plain language, and point you to the right page. What would you like to know?