Submitting a job to a slurm queue - Revision history

Dvory: /* Accessing the System */

2026-05-14T12:40:23Z

Accessing the System

2026-05-13T08:47:02Z

2026-05-13T08:44:52Z

2026-04-29T05:57:48Z

Using special resources

2026-04-26T05:42:10Z

Script for 1 GPU

2026-04-26T05:41:46Z

Basic Script

2026-01-19T14:47:30Z

Feature

2026-01-19T14:47:02Z

Feature

2026-01-19T14:45:26Z

Feature

2026-01-19T14:39:40Z

Using special resources

@@ Line 3: / Line 3: @@
 * '''We have chatgpt page for the new qos configuration, please look in [https://hpctoolkit.tau.ac.il HPC Helper Toolkit For TAU]'''
-To submit jobs to SLURM at Tel Aviv University, you need to access the system through one of the following login nodes:
+To submit jobs to SLURM at Tel Aviv University, you need to access the system through one of the following login node:
 * slurmlogin.tau.ac.il

@@ Line 1: / Line 1: @@
 == Accessing the System ==
-* '''We have chatgpt page for the new qos configuration, please look in [https://hpctoolkit.tau.ac.il]'''
+* '''We have chatgpt page for the new qos configuration, please look in [https://hpctoolkit.tau.ac.il HPC Helper Toolkit For TAU]'''
 To submit jobs to SLURM at Tel Aviv University, you need to access the system through one of the following login nodes:

@@ Line 1: / Line 1: @@
 == Accessing the System ==
-* '''We have chatgpt page for the new qos configuration, please look in [https://chatgpt.com/g/g-68be7f9acfb88191978615c1693e2cff-hpc-helper-toolkit HPC-helper-toolkit]'''
+* '''We have chatgpt page for the new qos configuration, please look in [https://hpctoolkit.tau.ac.il]'''
 To submit jobs to SLURM at Tel Aviv University, you need to access the system through one of the following login nodes:

@@ Line 285: / Line 285: @@
 There are several defined resources, which can be used in addition to the regular resources (e.g. memory, cpus)
 Thoses resources are:
-GresTypes=gpu,amd,af3,intel,disk700g
+GresTypes=gpu,amd,af3,intel,localscratch
 When your job needs to use one or more of the above resource, please add "--constraint=<resources list with commas>"
-So for example, one may submit a job using parameter --constraint=disk700g,amd
+So for example, one may submit a job using parameter --constraint=localscratch,amd
 And this would mean that the user's job requires amd processors family and up to 700 gb cache disk space during the job's run.
@@ Line 296: / Line 296: @@
 <pre>
-export CACHEDIR=/tmp/dvory_${SLURM_JOB_ID}
+export CACHEDIR=/localscratch/dvory_${SLURM_JOB_ID}
 mkdir -p $CACHEDIR
@@ Line 310: / Line 310: @@
 You may see all the features, which are defined resources (a.k.a. constraints), when typing the command 'features' on the login node
 The following features are available:
-Af3,AMD avx,disk700g ib_bh,ib_dj Intel
+Af3,AMD avx,localscratch,ib_bh,ib_dj,Intel
 Using them is by adding parameter --constraint=<feature>
@@ Line 322: / Line 322: @@
 . avx - Nodes with avx cpu capabilities
-. disk700g - nodes which have at least 700 gb in their local /tmp partition
+. localscratch - nodes which have at least 700 gb in their local /localscratch partition
 . ib_bh - Nodes with infiniband network #1

@@ Line 213: / Line 213: @@
 #SBATCH --ntasks=1                     # Number of tasks
 #SBATCH --nodes=1                      # Number of nodes
-#SBATCH --cpus-per-task=1              # CPUs per task
+#SBATCH --cpus-per-task=10             # CPUs per task
 #SBATCH --gres=gpu:1                   # Number of GPUs
 #SBATCH --mem-per-cpu=4G               # Memory per CPU

@@ Line 329: / Line 329: @@
 . Intel - Nodes with cpu of intel family
-<pre>
+</pre>
 To see all the features you may type '''features'''

@@ Line 331: / Line 331: @@
 <pre>
 To see all the features you may type '''features'''
 ==AlphaFold==

@@ Line 315: / Line 315: @@
 The meaning of the features:
+<pre>
 . Af3 - nodes which may run alphafold3
@@ Line 329: / Line 329: @@
 . Intel - Nodes with cpu of intel family
 ==Running matlab example==

@@ Line 306: / Line 306: @@
 trap cleanup EXIT INT TERM HUP
 </pre>
 ==Running matlab example==