Skip to content

PERUN Slurm Script Generator

Interaktivny generator Slurm skriptov so automatickou podporou scratch, spravnymi particiami a e-mailovymi notifikaciami.

PERUN Slurm Script Generator
Interactive generator with automatic scratch support
Slurm · PERUN · HPC

Job Settings

Fill in the form and the script will be generated automatically.

days
-
hours
:
min
:
sec
Automatic Scratch
Specific Node
Append Output
Set OMP_NUM_THREADS

Email Notifications

Select the events for which you want to receive an email.

Quick Options
BEGIN,END,FAIL
ALL
Clear All
Job States
BEGIN
END
FAIL
REQUEUE
ALL
Time Limits
TIME_LIMIT (100%)
TIME_LIMIT_90
TIME_LIMIT_80
TIME_LIMIT_50
Other
ARRAY_TASKS
STAGE_OUT
INVALID_DEPEND
Automatic Scratch (recommended)
The job runs on fast Lustre scratch with 40x faster I/O. Just add source .activate_scratch; the prolog/epilog handles the rest. Results are synced to ~/results_job_XXXXX/ after completion.
GPU Partitions
Use gpu_short (gpu01–18, max 2 days) or gpu_long (gpu19–26, max 4 days). GPUs are reserved via --gres=gpu:X.
Email Notifications
ALL = BEGIN + END + FAIL + REQUEUE + STAGE_OUT. Options can be combined, e.g. END,TIME_LIMIT_80. ARRAY_TASKS sends a notification for each array task separately.
Pinning a Specific Node
Use only for debugging. In production, let Slurm decide optimally.

Generated Slurm Script

#!/bin/bash · #SBATCH · automatic scratch · prolog/epilog
How It Works
PROLOG (automatic): Creates scratch, copies files
YOUR JOB: Runs on scratch with source .activate_scratch
EPILOG (automatic): Syncs results, cleans up
Submit & Monitor
sbatch job.sh · squeue -u $USER · tail -f ~/results_job_*/output.log
PERUN Slurm Script Generator · cpu_short / cpu_long / gpu_short / gpu_long / viz · TUKE