compute_ru

This is an old revision of the document!

What's next ?

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

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

When you job need GPU add this line to you slurm cmd file
#SBATCH --gpus-per-node=1

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