Hi Jessie,
Jessie Poquérusse writes:
> Hello,
>
> In trying to modularize and genericize my Bash scripts as much as
> possible, I was wondering if there was a way to turn #SBATCH options
> (mainly walltime, mem, and output and error file directories) into
> externally defined script parameters
I don’t think you can turn them into command-line arguments to your script
since Bash treats the #SBATCH lines as comments and can’t see “into” them, but
you can, as I’m sure you know, override any #SBATCH options by putting them on
the sbatch command-line before specifying the command to run. I
Hello,
In trying to modularize and genericize my Bash scripts as much as possible,
I was wondering if there was a way to turn #SBATCH options (mainly
walltime, mem, and output and error file directories) into externally
defined script parameters (e.g. in the same command line as during sbatch
job
Hey all,
We used to numbers from the following commands:
sinfo -o %G(as suggested above) - gives total gpu in cluster
squeue -o %b - gives amount of gpu in use for each running job
sum all the numbers under %b gives you gpu in use in cluster
pestat was sug
On Friday, 16 March 2018 2:44:09 AM AEDT Mahmood Naderan wrote:
> That is not what I want. I want a total upper limit for time usage of
> a user.
I'm not sure Slurm can do that for you by setting limits on a user I'm afraid!
You can set GrpWall on an association or QoS but that will govern the m
On Sunday, 18 March 2018 4:30:34 PM AEDT Mahmood Naderan wrote:
> Thanks for the exaplnation Chris. I will read cgroup.
My pleasure! The Slurm docs on it are here:
https://slurm.schedmd.com/cgroups.html
I've been using cgroups for all three abilities (process tracking, task
management and ac
