Using Adastra - Revision history

Martin.turbet: /* Getting access to the cluster */

2026-01-19T10:34:57Z

‎Getting access to the cluster

2025-06-27T06:52:01Z

‎Example of a mixed MPI/OpenMP job to launch a simulation

2025-06-27T06:50:28Z

‎Example of a mixed MPI/OpenMP job to launch a simulation

2025-06-27T06:38:21Z

2025-05-13T14:25:14Z

2025-04-30T14:06:57Z

2025-04-22T11:28:01Z

2024-10-14T08:20:41Z

2024-06-24T09:00:54Z

2024-05-15T10:28:15Z

‎Using gdb4hpc

@@ Line 17: / Line 17: @@
 # YOU MUST THEN VALIDATE THE REQUEST: click on the "Valider la saisie des informations"
 # You then receive an automatic mail, but it's only to tell you to go to the next step: You must now download the pre-filled form from e-dari: find "télécharger la demande" and download the pdf. Sign it, and upload it on e-dari "déposer la demande de création de compte".
-# Wait for your application to be preprocessed by the system...
+# Wait for your application to be preprocessed by the system... (if after 1-2 weeks you still don't have news, check with your lab director to make sure he/she granted authorization)
 == A couple of pointers ==

@@ Line 142: / Line 142: @@
 </syntaxhighlight>
 Note however that the ''srun'' instructions above will not yield very efficient results as one actually needs to specify the cpu binding (i.e. how core relate to one another) via dedicated functions :
-<syntaxhighlight>
+<syntaxhighlight  lang="bash">
 srun --ntasks-per-node=${SLURM_NTASKS_PER_NODE} --cpu-bind=none --mem-bind=none --label -- ./adastra_cpu_binding.sh ./gcm_64x48x54_phymars_para.e > gcm.out 2>&1
 </syntaxhighlight>

@@ Line 135: / Line 135: @@
 ### OMP_NUM_THREADS value must match "#SBATCH --cpus-per-task"
-export OMP_NUM_THREADS=4
+export OMP_NUM_THREADS=${SLURM_CPUS_PER_TASK}
 export OMP_STACKSIZE=400M
@@ Line 141: / Line 141: @@
 </syntaxhighlight>
+Note however that the ''srun'' instructions above will not yield very efficient results as one actually needs to specify the cpu binding (i.e. how core relate to one another) via dedicated functions :
 == Using python ==

@@ Line 66: / Line 66: @@
 <pre>
 rsync -avz irenelogin@irene-fr-ccfr-gw.ccc.cea:irene_path_to_your_data adastra_path_to_your_data
 </pre>

@@ Line 193: / Line 193: @@
 module load ddt
 </syntaxhighlight>
 ...TODO: Add some example using DDT ...

@@ Line 7: / Line 7: @@
 # Beware! If you are on 2 lab (LMD and LATMOS for example), you must register with your email address corresponding to your Janus account.
 # Click on "se rattacher à un dossier ayant obtenu des resources" or "Attach yourself to an application file that has obtained resources"
-# "Atmosphères Planétaires" project number to provide:  A0160110391
+# "Atmosphères Planétaires" project number to provide:  A0180110391
 # Ehouarn then receives an email to allow you to join the project. Once he has validated it, you receive a confirmation mail.
 # Once approved, you have to request for an account, click on "CINES: créer une demande d'ouverture de compte"
@@ Line 21: / Line 21: @@
 == A couple of pointers ==
-* Connecting to Adastra: For those who had an account on Occigen, we have retained group and login credentials from then; To connect to Adastra you need first go through the LMD gateway (hakim) or the IPSL (Ciclad/Spirit) gateway and then
+* Connecting to Adastra: For those who had an account on Occigen, we have retained group and login credentials from then; To connect to Adastra you need first go through the LMD gateway (hakim) or the IPSL (Spirit/SpiritX) gateway and then
 <syntaxhighlight lang="bash">
 ssh your_cines_login@adastra.cines.fr

@@ Line 178: / Line 178: @@
 # Launch a request for an allocation on a compute node: <syntaxhighlight lang="bash"> salloc --account=cin0391 --constraint=GENOA --job-name="debug" --nodes=1 --time=1:00:00 --exclusive </syntaxhighlight>
 # Identify which node it is linked to and directly ssh (from login node) to it, e.g. if it is node "c1516" <syntaxhighlight lang="bash"> ssh c1516 </syntaxhighlight>
-# source your usual environment and then the gdb4hpc module <syntaxhighlight> module load gdb4hpc/4.16.0.1 </syntaxhighlight>
+# source your usual environment and then the gdb4hpc module <syntaxhighlight lang="bash"> module load gdb4hpc/4.16.0.1 </syntaxhighlight>
 # Go to your work directory and launch gdb4hpc
-# within gdb4hpc: <syntaxhighlight> dbg all> launch $a{1} --launcher-args="--mpi=cray_shasta -A cin0391 --constraint=GENOA -t 00:30:00 -N 1 --cpu-bind=verbose,cores --exclusive"  ./executable.exe </syntaxhighlight>
+# within gdb4hpc: <syntaxhighlight lang="bash"> dbg all> launch $a{1} --launcher-args="--mpi=cray_shasta -A cin0391 --constraint=GENOA -t 00:30:00 -N 1 --cpu-bind=verbose,cores --exclusive"  ./executable.exe </syntaxhighlight>
 Once everything running, the first thing you have to do is set a breakpoint at the beginning of the program, e.g.:
-<syntaxhighlight>
+<syntaxhighlight lang="bash">
 break icosa_lmdz.f90:1
 </syntaxhighlight>
 And then "continue" to that point
 == Are you being disconnected when inactive? ==

@@ Line 59: / Line 59: @@
 </syntaxhighlight>
 * In a nutshell: we have lots of space on the WORKDIR (250 To) which is "permanent" (unlike the SCRATCHDIR, where files older than 30 days are purged), so use it! And when you want to archive things, make some large tar files and put them on the STOREDIR
 == Submitting jobs ==

← Older revision		Revision as of 09:00, 24 June 2024
Line 49:		Line 49:

	* Link to the Adastra technical documentation: https://dci.dci-gitlab.cines.fr/webextranet/		* Link to the Adastra technical documentation: https://dci.dci-gitlab.cines.fr/webextranet/
		+
		+	* Link to the webpage where you can find out (login and password are those of your Adastra account) how many hour left we have on the project and details about everyone's use of Adastra: https://reser.cines.fr

	== Disks and workspaces ==		== Disks and workspaces ==