Differences

This shows you the differences between two versions of the page.

--- lvl_cluster:info [2023/01/02 12:55] – bjorns
+++ lvl_cluster:info [2023/01/04 09:09] (current) – removed bjorns
@@ Line 1: / Line 1: @@
-=====Language and Voice lab computing cluster=====
- **Smallvoice**, uses [[https://slurm.schedmd.com|Slurm]] workload manager to create a computing cluster\\
-The Cluster has 6 nodes :\\
-^Node name ^ Role(s) ^
-|**atlas**|management node, worker|
-|**freedom**|**login** node, worker|
-|hercules|worker node|
-|samson|worker node|
-|goliath|worker node|
-|obelix|worker node|
-\\
-When logged on to the cluster user is always on the login node, **freedom** and does all his work there\\ but home (& work) are the **same** "physical" disk on all nodes\\
-To use slurm workload manager for your job, you first create a executable batch file with info about the job\\
-and the run your job(submit) with sbatch myBatch.sh\\
-Example batch file\\
-''#!/bin/bash\\
-#SBATCH <nowiki>--</nowiki>account=staff\\
-#SBATCH %%--%%job-name=MyJob\\
-#SBATCH %%--%%gpus-per-node=1\\
-#SBATCH %%--%%mem-per-cpu=2G\\
-#SBATCH <nowiki>--</nowiki>output=myBatch.log\\
-''
-\\
-Create with : ''vi myJob.sh'' + [save] + ''chmod +x myJob.sh''\\ but this example file is available from user home folder with ''cat myBatch.sh''\\
-\\
-== The computing (slurm) environment ==
-There are 3 queue's (partitions) \\
-allWork	(for staff only)\\
-doTrain	(for staff only)\\
-beQuick	(for students)\\
-\\
-The default queue is doTrain so when using slurm it's not necessery to choose a queue, but it's possible to specify another one.\\
-\\
-===Installed software and drivers===
- * **NVIDIA** A100 GPU drivers\\
- * **CuDA toolkit** [version 11.7]\\
- * Intel oneAPI Math Kernel Library\\
- * Python 3.9.2\\
- * pip 20.3.4\\