HPC Job Scheduler Training

Master SLURM.
Without a cluster.

Interactive terminal simulator for SLURM — the world's most widely used HPC workload manager. Practice job submission, monitoring, GPU jobs, arrays, and accounting directly in your browser.

user@login-01 — hpccluster
user@login-01:~$ sinfo
PARTITION AVAIL TIMELIMIT NODES STATE NODELIST
compute* up 7-00:00:00 8 2 mix, 2 alloc node[001-008]
gpu up 2-00:00:00 3 1 alloc, 2 idle gpu[01-03]
user@login-01:~$ sbatch mpi_job.sh
Submitted batch job 4821101
user@login-01:~$ squeue --me
JOBID PARTITION NAME USER ST TIME NODES
4821101 compute mpi_run user R 0:00:03 4
user@login-01:~$
17
SLURM Commands
145
Searchable Commands
15
Simulated Nodes
100%
Browser-based

Everything to run
real cluster jobs

SLURM powers most of the world's supercomputers. Practice the full workflow — from first sbatch to advanced scheduling — without touching a real cluster.

Real Terminal Feel

Inline prompt inside the output, blinking cursor, Ctrl+C, history, Tab completion — exactly like Rocky Linux on a real HPC login node.

📋

Live Job Queue

Jobs transition PD→R→COMPLETED in real time. Sidebar shows live queue state updated as you submit and cancel jobs.

🔍

Command Search

Search 145+ commands — all SLURM commands, #SBATCH directives, Linux, systemd, and modules — by name or description. Click any result to paste it into the terminal.

🔧

All #SBATCH Directives

GPU jobs, job arrays, dependencies, MPI, memory limits, QOS — all sbatch flags parsed and simulated correctly.

📊

Full Accounting

sacct, sreport, sshare, sacctmgr — understand fairshare, QOS, and cluster usage reporting.

🐧

Linux + Modules

Bash, pipes, grep, module load/avail/purge — full environment just like a real cluster login node.

The complete SLURM toolkit

Submit
  • sbatch
  • srun
  • salloc
  • sattach
  • sbcast
Monitor
  • squeue
  • sinfo
  • sstat
  • sprio
  • sdiag
Control
  • scancel
  • scontrol
  • strigger
Accounting
  • sacct
  • sreport
  • sshare
  • sacctmgr