[slurm-users] associations, limits,qos

[Diego Zuccato]

Il 29/01/21 08:47, Diego Zuccato ha scritto:

>> Jobs submitted with sbatch cannot run on multiple partitions. The job
>> will be submitted to the partition where it can start first. (from
>> sbatch reference)
> Did I misunderstand or heterogeneous jobs can workaround this limitation?
My quick test seems to confirm that it works in 18.08 (as packaged in
Debian Buster).

My test jobscript:
-8<--
#!/bin/bash
#SBATCH --time 1
#SBATCH --cpus-per-task=1 --mem-per-cpu=1g --ntasks=2 --constraint=blade
#SBATCH packjob
#SBATCH --cpus-per-task=1 --mem-per-cpu=1g  --ntasks=4 --constraint=matrix
srun --label : hostname
-8<--
Its output:
-8<--
1: str957-bl0-01
0: str957-bl0-01
3: str957-mtx-01
5: str957-mtx-01
4: str957-mtx-01
2: str957-mtx-01
-8<--

bl-* and mtx-* nodes are in disjoint partitions:
-8<--
$ scontrol show partitions
PartitionName=b1
[...]
   Nodes=str957-bl0-[01-02]
[...]
PartitionName=m1
[...]
   Nodes=str957-mtx-[00-15]
-8<--

-- 
Diego Zuccato
DIFA - Dip. di Fisica e Astronomia
Servizi Informatici
Alma Mater Studiorum - Università di Bologna
V.le Berti-Pichat 6/2 - 40127 Bologna - Italy
tel.: +39 051 20 95786