SMU’s VPN is required to remotely access the Server. 

CPU: 64 cores, 128 threads, Xeon Gold 6438Y Processors
GPU: Dual NVIDIA L4 (x2), 24 GB each
RAM: 128 GB

SSH:

login: username@qtinslab.smu.edu

Disk/Volume

Home directories: 400 GB, HDD-SAS 2.0, /home
Large data storage: 2 TB, HDD-SAS 2.0, /data/tank
Scratch workspace: 1.7 TB, SSD-M.2, /data/scratch (Please use this dir to run tasks on GPUs)

SLURM

Job submission: sbatch run.shell

#!/bin/bash

#SBATCH --job-name=test_two_gpus

#SBATCH --output=test_two_gpus_%j.out

#SBATCH --error=test_two_gpus_%j.err

#SBATCH --partition=main

##SBATCH --gres=gpu:2

#SBATCH --cpus-per-task=8

#SBATCH --mem=32G

#SBATCH --time=01:00:00

echo "Running on host: $(hostname)"

echo "GPUs allocated:"

nvidia-smi

# Activate your virtual environment

source /opt/conda/etc/profile.d/conda.sh

#conda activate torch_env_3.12.4

conda activate tf_env

python test_gpu.py

Documentations: Slurm Workload Manager - Quick Start User Guide 

How to create a conda environment?

Add conda on your PATH
Add HTTP_PROXY in .bashrc.