Slurm

From ISOR
Revision as of 13:45, 20 September 2021 by Roland (Talk | contribs)

Jump to: navigation, search

The Slurm job scheduler on the High Performance WiWi Cluster (HPC3)

Contents

1 Introduction

2 Cluster topology & hardware specs

3 Submitting jobs

3.1 Batch jobs

  • Job scripts

Job scripts are standard BASH-scripts with some extra information for the slurm job scheduler. Hence, we start with the first line of the script, telling the OS which kind of UNIX shell to use for interpreting the commands in the script.

#!/bin/bash

Then we add a series of directives for the slurm job scheduler, each starting with a '#SBATCH'. Although the '#' character usually indicates a comment, this specific string gets interpreted by slurm and allows to set various options.

#SBATCH -N 1
#SBATCH --ntasks-per-node=16
#SBATCH --ntasks-per-core=1
#SBATCH --partition=mem_0064
#SBATCH --qos=normal_0064
#SBATCH --mail-type=BEGIN,END
#SBATCH --mail-user=roland.braune@univie.ac.at


3.2 Interactive jobs

Personal tools
Namespaces

Variants
Actions
Navigation
Tools