cluster:usage_tips:startslurm [Sleep Revolution Wiki]

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
===== Getting started with the Sleep Revolution slurm cluster =====

**Slurm** is a workload manager,scheduler system. It's build from computer nodes, 1 node is the Login Node, a master node and worker node's\\
The login node is called **cheetah.sleep.ru.is** ''130.208.209.30''.\\
The cluster is a linux environment, connect with SSH\\
\\
Open a terminal to log-in '' ssh <myUser>@cheetah.sleep.ru.is''\\
Those who use Rivendell ''130.208.209.30'' (and other sleep servers) can also connect with any editor which supports SSH like [[https://code.visualstudio.com/|VScode]]\\
==== About the user environment : ====
Each user has his home folder, the path is [/mount/home/<username>]\\
The home folder should have 2 files, a readMe.lst file and template (slurm) script file (exampleJob.sh)\\

  - The first step for users is to make sure their working environment fits their needs.\\ For those who are using Python, we recommend, for ease of use, using a Python virtual environment\\ User can then install different version of Python and all necessary modules for their work.\\
  - However user can setup and install their Python modules locally without a Python virtual environment, by using Pip3 install <Module>\\
  - The cluster gives users access to memory, cpu and GPU. The script file has instructing the slurm cluster what to do\\
  - Users can change the slurm instructions in the script file, but it's not necessary for it to work, just append your command to it.
  - A example of how to execute your job on the cluster; first create a file __prufa.py__ in your home directory, append the line ''python3 prufa.py'' to exampleJob.sh \\ Then submit your job to the queue with command **sbatch exampleJob.sh**

\\
Your template script file (exampleJob.sh) looks like this\\
''#!/bin/bash\\
#SBATCH <nowiki>--</nowiki>account=staff\\
#SBATCH %%--%%job-name=sleepJob\\
#SBATCH %%--%%gpus-per-node=1\\
#SBATCH %%--%%mem-per-cpu=2G\\
#SBATCH <nowiki>--</nowiki>output=Slurm.log\\
''
\\
===Useful commands===
^cmd^Descr^
|sbatch <exampleJob.sh>|Submit job|
|sacct|My jobs|
|squeue|Show all jobs on the queue|
|sinfo|Information about the cluster|
|srun|Run job interactively|
|scancel <id>|kill job with <id>|

===Example of Python virtual environment===
Python venv\\
Miniconda\\
Aaconda\\

[[cluster:usage_tips:advancedSlurm|Advanced tips]]\\