https://github.com/lgarrison/dynamic-disbatch

An example of using disBatch's Python interface to submit tasks and dynamically resubmit them based on the return code
https://github.com/lgarrison/dynamic-disbatch

Last synced: 26 days ago
JSON representation

An example of using disBatch's Python interface to submit tasks and dynamically resubmit them based on the return code

• Host: GitHub
• URL: https://github.com/lgarrison/dynamic-disbatch
• Owner: lgarrison
• Created: 2023-07-06T20:36:55.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-07-24T14:41:14.000Z (over 1 year ago)
• Last Synced: 2024-10-16T01:31:08.825Z (2 months ago)
• Language: Python
• Size: 3.91 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# dynamic-disBatch

This is an example of using [disBatch's](https://github.com/flatironinstitute/disBatch) `DisBatcher` Python interface to submit tasks and dynamically resubmit them based on the return code. Each task uses 1 GPU.

## Organization

* `job.sbatch`: the main entry point; submit with `sbatch job.sbatch`

* `run_disbatcher.py`: invoked by `job.sbatch` to launch the tasks dynamically

* `do_work.py`: a dummy "science" script that takes a seed and some parameters, and uses 1 GPU

* `env.sh`: an example environment setup script, used by `job.sbatch`

## Usage

Modify the `OUTDIR` path in `run_disbatcher.py`. Logs from `do_work.py` will go there.

Submit with:

```console

$ sbatch job.sbatch

```

disBatch logs go in a directory named `job$SLURM_JOBID`. Some status information will be written into the `slurm-*.out` log.