A guide to running computational workflows on DesignSafe, from interactive exploration in a Jupyter notebook to production-scale simulations on TACC supercomputers.
- How It Works The DesignSafe portal, three compute environments (JupyterHub, Virtual Machines, HPC), storage, and how to design a workflow around your research
- Submitting a Job Through the Portal Step-by-step walkthrough with screenshots (OpenSees-MP example)
- Running HPC Jobs Job submission with dapi, serial vs parallel (MPI) execution, resource parameters, and file staging
- Debugging Failed Jobs Job states, output files, troubleshooting checklist, and common failure patterns
- Parameter Sweeps Running hundreds of independent simulations with PyLauncher
- DesignSafe Applications Catalog of 45+ tools for simulation, analysis, visualization, GIS, and hazard data
- Advanced Topics Tapis internals, execution strategies, custom app development
This work is licensed under CC BY 4.0. See LICENSE.md.
See AUTHORS.md.