This is an old revision of the document!
What's next ?
Home folder for students is /student/<user>
Most users are using Python for their jobs
Setup your environment, Python virtual environment/Anaconda and install in you home folder the necessary libraries for your project
When you are ready to send your job to the slurm queue, adjust the slurm command script
The slurm command script : sbatch script.sh
| cmd | Description |
|---|---|
#SBATCH --account=student | Your user belongs to this group |
#SBATCH --job-name=MyJob | name of my job |
#SBATCH –ntasks=2 | Number of job tasks per worker |
#SBATCH –mem=4G | Job wants 4gb on memory |
#SBATCH –time=0 | Time limit on my job e.g. time=11::00 (11 hours), 0 means nolimit |
#SBATCH –partition=Lokaverk | Run job on this queue |
#SBATCH --output=myBatch.log | Log file for environment & slurm |
| python3 file.py | Here you put your run command |
When you job need GPU add this line to you slurm cmd file
#SBATCH --gpus-per-node=1
Note : We recommend all students in this 3-week-course run their GPU job for a short time, using “–time=hh:MM” slurm directive, so all jobs get some GPU time
Other slurm directives
#SBATCH --mem-per-cpu=2G
#SBATCH --cpus-per-task=2 : 2 cores per process/task
#SBATCH --ntasks-per-node=4 : 4 procees per node(worker
Tools and libraries installed
CUDA toolkit version 11,7
Intel oneAPI Math Kernel Library
sox
libsndfile1-dev
ffmpeg
python3-venv
python3-pip