SLURM job submission on Gilbreth Cluster

gilbreth.job

#!/bin/bash -l
#SBATCH --nodes=1 #Number of nodes
#SBATCH --ntasks-per-node=24 #Number of CPUs per node. 96 CPUs in 1 node
#SBATCH --gres=gpu:1 #Number of GPUs. We have 4 H100s in 1 node.
#SBATCH --partition=araghu #Always set to araghu
#SBATCH --mem=240G #Amount of memory required. Maximum is 2TB
#SBATCH --time=2-00:00:00 #Maximum Time Limit
#SBATCH -A araghu #Queue name: araghu or araghu-scale, whichever you have access to.
#SBATCH -J job_name
#SBATCH --output=job_output.out #Job output file

echo "=== Job $SLURM_JOB_ID starting at $(date +'%Y-%m-%d %H:%M:%S') ==="
module load conda
conda activate /depot/araghu/data/selvams/CondaEnvs/Colbert-RAG
python xx_yy.py
echo "=== Job $SLURM_JOB_ID starting at $(date +'%Y-%m-%d %H:%M:%S') ==="