Frequently Asked Questions
This page addresses common questions about using UTEP's High-Performance Computing (HPC) resources, including the Jakar, Paro, and Punakha clusters.
π Access & Accounts
Request access to the HPC clusters
Go to the HPC Account Requests and follow the instructions.
More information?
π¦ Software & Modules
What software is available?
module spider
How do I load a module?
module load <module_name>
More information?
β‘ Performance & Storage
How much storage do I get?
| Storage Type | Quota | Backups | Check | Location | Notes |
|---|---|---|---|---|---|
| Home | 10 GB | β | qta_h |
/home/ |
clean for configs and critical scripts |
| Work | 100 GB | β | qta_w |
/work/ |
active, non-critical data |
| Scratch | Unlimited* | β | /scratch/ |
temporal data 7-day purge policy applies |
More information?
π₯ Researchers Under Partner PIs
Ask your PI contact HPC staff. General Inquiry
π₯οΈ Job Submission
π₯οΈ RAD-HPC Job Submission Guide
π§ Useful Commands
# See your associations (accounts + QoS)
sact_me
# Inspect why a job is pending
scontrol show job -dd <jobid>
sprio -j <jobid>
# Check partition status
sinfo -p DGX
sinfo -p medium
# Why nodes unavailable?
sinfo -R
π€ Standby QoS
Best-Effort Jobs Β· Opportunistic Scheduling Β· Lower Priority Β· Preemptible Workloads
More information?
π€ RAD-HPC Standby QoS Guide
π οΈ Troubleshooting
My job failed. What should I do?
- Check the SLURM output/error logs (e.g.,
slurm-<jobid>.out). - Verify your job script for errors.
- If still unresolved, contact the HPC support team with your job ID.
My module/software isnβt working. What next?
Unload and reload the module, or try a different version:
module purge
module load <module_name>/<version>
Tip
If the issue persists, contact support. General Inquiry
π¬ Getting Help
π Direct Service Desk Links
π Quick Commands Cheat Sheet
Here are some of the most common HPC commands youβll need:
| Purpose | Command Example |
|---|---|
| Submit a batch job | sbatch job.slurm |
| Run an interactive job | srun --pty bash |
| Check running jobs | squeue -u $USER |
| Cancel a job | scancel <jobid> |
| Show job details | scontrol show job <jobid> |
| Check account usage | sacct -u $USER |
| List available modules | module avail |
| Load a module | module load gcc/11.2.0 |
| Unload all modules | module purge |
| Check storage quota (home) | mmlsquota --block-size AUTO home |
| Check storage quota (work) | mmlsquota --block-size AUTO work |
β This page is intended as a living FAQ. If you think a question should be added, please suggest it through the General Inquiry.