basic use example
Completion requirements
Accessing the Cluster
-
To access the Faith HPC cluster:
-
Use SSH to connect to the server:
ssh your_username@diufrd200.unifr.ch -
VPN Required: This server is only accessible via the university VPN.
-
Using the Cluster with Slurm
Please find a concrete example to how to use that cluster here. Below, you can find only a brief introduction on how to use learn Slurm.
To run your jobs on the Faith HPC cluster, you'll use Slurm, a job scheduling system. Here are some basic commands to get you started:
-
Submit a job:
sbatch my_script.sh -
Check job status:
squeue -u your_username -
Cancel a job:
scancel job_id -
Example Slurm script:
#!/bin/bash #SBATCH --job-name=my_job #SBATCH --output=output_%j.txt #SBATCH --error=error_%j.txt #SBATCH --ntasks=1 #SBATCH --mail-type=ALL #SBATCH --mail-user=yourmail@unifr.ch source activate my_env python3 my_script.py- For advance usage, you can specify additional options to your scripts to tune it to your needs :
#SBATCH --cpus-per-task=4 #SBATCH --gres=gpu:1 #SBATCH --time=01:00:00 #SBATCH --nodelist=diufrd204
- For advance usage, you can specify additional options to your scripts to tune it to your needs :
Make sure to adjust the resources requested according to your needs. If you’re unsure, start small and scale up!
If you encounter any issues or have questions, please contact the cluster administrator micheal.baechler@unifr.ch. Enjoy your computing experience!
Faith Cluster Team