Differences

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

--- doku:quick3save [2014/11/25 10:28] – created ir
+++ doku:quick3save [2015/05/28 12:19] (current) – [A simple job submission script] markus
@@ Line 1: / Line 1: @@
- (%s = %N).
+==== Brief Introduction ====
-It is likely that your copy of e2fsck is old and/or doesn't support this @j format.
-It is also possible the @j @S is corrupt.
-Clearing fields beyond the V1 @j @S...
+=== Login ===
+<code>
+ssh <username>@vsc3.vsc.ac.at
+</code>
+In the following you will be asked to type //first// your password and //then// your **o**ne **t**ime **p**assword (OTP; sms token).
-is %N; @s zero.  	now = %T) is in the future.
+[[doku:win2vsc|How to connect from Windows?]]
-	now = %T) is in the future.
-	(by less than a
+=== VSC 3 ===
+Once you have logged into VSC-3, type:
+  * **module avail**    to get a basic idea of what is around in terms of installed software and available standard tools
+  * **module list**    to see what is currently loaded into your session
+  * **module unload //xyz//**    to unload a particular package **//xyz//** from your session
+  * **module load //xyz//**    to load a particular package **//xyz//** into your session \\
+== Note: ==
+  - **//xyz//** format corresponds exactly to the output of **module avail**. Thus, in order to load or unload a selected module, copy and paste exactly the name listed by **module avail**.\\
+  - a list of **module load/unload** directives may also be included in the top part of a job submission script\\
+When all required/intended modules have been loaded, user packages may be compiled as usual.
+=== SLURM (Simple Linux Utility for Resource Management) ===
+Contrary to previous VSC* times, the scheduler on VSC-3 is [[http://slurm.schedmd.com|SLURM]].
+For basic information type:
+  * **sinfo**   to find out which 'queues'='partitions' are available for job submission. Note: the in SGE times termed 'queue', is now under SLURM called a 'partition'.
+  * **scontrol show //partition//**   more or less the same as the previous command except that with **scontrol** much more information may be obtained and basic settings be modified/reset/abandoned.
+  * **squeue**    to see the current list of submitted jobs, their state and resource allocation.
+==== A simple job submission script====
+vi check.slrm\\
+<code>
+#!/bin/bash
+#
+#SBATCH -J chk
+#SBATCH -N 2
+#SBATCH --ntasks-per-node=16
+#SBATCH --ntasks-per-core=1
+mpirun -np 32 a.out
+</code>
+  * **-J**     some name for the job\\
+  * **-N**     number of nodes requested (16 cores per node available)\\
+  * **--ntasks-per-node**     number of processes run in parallel on a single node \\
+  * **--ntasks-per-core**     number of tasks a single core should work on\\
+  * **mpirun -np 32 a.out**    standard invocation of some parallel program (a.out) running 32 processes in parallel. **Note**,
+  * in SLURM **srun** is preferred over **mpirun**, so an equivalent call to the one on the final line above could have been **srun -l -N2 -n32 a.out** where the **-l** just adds task-specific labels to the beginning of all output lines.
+==Job submission===
+<code>
+[username@l31 ~]$ sbatch check.slrm    # to submit the job
+[username@l31 ~]$ squeue               # to check the status
+[username@l31 ~]$ scancel  JOBID       # for premature removal, where JOBID
+                                       # is obtained from the previous command
+</code>
+===Another simple job submission script===
+This example is for using a set of 4 nodes to compute a series of jobs in two stages, each of them split into two separate subjobs. \\
+vi check.slrm\\
+<code>
+#!/bin/bash
+#
+#SBATCH -J chk
+#SBATCH -N 4
+#SBATCH --ntasks-per-node=16
+#SBATCH --ntasks-per-core=1
+export I_MPI_PMI_LIBRARY=/cm/shared/apps/slurm/current/lib64/libpmi.so
+scontrol show hostnames $SLURM_NODELIST  > ./nodelist
+srun -l -N2 -r0 -n32 job1.scrpt &
+srun -l -N2 -r2 -n32 job2.scrpt &
+wait
+srun -l -N2 -r2 -n32 job3.scrpt &
+srun -l -N2 -r0 -n32 job4.scrpt &
+wait
+</code>
+== Note: ==
+the file 'nodelist' has been written for information only; \\
+it is important to send the jobs into the background (&) and insert the 'wait' at each synchronization point; \\
+with **-r2** one can define an offset in the node list, in particular the **-r2** means taking nodes number 2 and 3 from the set of four (where the list starts with node number 0), hence a combination of -N -r -n allows full control over all involved cores and the tasks they are going to be used for;