Skip to content

Getting Started at REPACSS

About this page

This page will guide you through the basics of accessing, connecting to, and beginning work on the REPACSS High-Performance Computing (HPC) cluster.

Welcome to REPACSS — Texas Tech University's research and educational platform for advanced computing support services. This guide is tailored to new users who are preparing to run jobs on the cluster for the first time.

Account Access

To utilize REPACSS resources, users must have a valid TTU account and a TTUnet VPN access.

TTU Users

  • Access must be requested through the system administrator.
  • A valid institutional email address (ending in ttu.edu) is required.
  • Provisioning typically takes 2–3 business days after the request is approved.

Multi-Factor Authentication (MFA) and VPN

All users accessing the system remotely must use TTU’s GlobalProtect VPN and have Microsoft Multi-Factor Authentication (MFA) configured.

MFA and VPN required

Login attempts from outside the TTU network require both MFA and VPN authentication.

Connecting to the Cluster

Once account provisioning and secure connection setup are complete, users may connect via Secure Shell (SSH):

ssh <your-ttu-username>@repacss.ttu.edu

Access via Visual Studio Code is also supported:

First-time login tip

On initial login, run quota -s to review disk usage, and module avail to browse available software modules.

Storage areas

  • Use /home for scripts and long-term storage.
  • Use /work for project-related data under active development.
  • Use /scratch for temporary data during compute jobs.

Environment Setup

REPACSS uses an environment module system to manage software packages.

module avail         # List available modules
module load gcc/12   # Load a specific version

Resources:

Submitting Your First Job

REPACSS employs the SLURM workload manager to schedule jobs.

Tip

Interactive sessions should be used exclusively for development and debugging purposes. Once your job is ready for full execution, please submit it using a SLURM batch job.

Sample SLURM Script

#!/bin/bash
#SBATCH --job-name=test
#SBATCH --output=output.txt
#SBATCH --ntasks=1
#SBATCH --time=00:05:00

srun ./my_program

Submit the job with:

sbatch job.slurm

Monitor your job with:

squeue -u <your-username>

Resources:

Getting Help

If assistance is required: