GIPDEEP – How to launch jobs

Slurm has two commands for launching jobs: srun and sbatch.
srun is straightforward: srun <arguments> <program>
You can use various arguments to control the resource allocation of your job, as well as other settings your job requires.
You can find the full list of available arguments here. Some of the most common arguments are:
-c # – allocate # CPUs per task.
‑‑gres=gpu:# – allocate # GPUs.
‑‑gres=gpu:<type>:# – allocate # GPUs of type <type>.
‑‑mpi=<mpi_type> – define the type of mpi to use. The default (when isn’t explicitly specified) is pmi2.
‑‑pty – run job in pseudo terminal mode. Useful when running bash.
-p <partition> – run job on the selected partition instead of the default one.
-w <node> – run job on a specific node.

Examples:
1. To get a shell with two GPUs, run:
srun ‑‑gres=gpu:2 ‑‑pty /bin/bash
run ‘nvidia-smi’ to verify the job received two GPUs.
2. Run the script ‘script.py’ using python3:
srun python3 script.py

sbatch lets you use a Slurm script to run jobs.
Exemples:

#!/bin/bash
#SBATCH -c 8                      # number of cores (treats)
#SBATCH --gres=gpu:2080ti:1          # Request 1 gpu type 2080ti
#SBATCH --mail-user=[user]@cs.technion.ac.il
#SBATCH --mail-type=ALL           # Valid values are NONE, BEGIN, END, FAIL, REQUEUE, ALL
#SBATCH --job-name="JobName"
#SBATCH -o ./out_job%j.txt        # stdout goes to out_job.txt
#SBATCH -e ./err_job%j.txt        # stderr goes to err_job.txt module purge # clean active modules list module load matlab/R2023a # activate matlab 2023 module conda activate [your miniconda environment] python main.py # run your script

• For several types you can mark it in “–gres” flag

#SBATCH --gres={gpu:A40:1,gpu:L40:1}