Thanks so much Greg! That looks like the solution we want, but like John
I'm also unfamiliar with spank plugins. I guess that will have to change.
Mark
On Wed, Apr 6, 2022 at 7:54 AM John Hanks wrote:
> Thanks, Greg! This looks like the right way to do this. I will have to
> stop putting off le
Use the formatting commands.
From the manpage for sinfo:
The format of each field is "%[[.]size]type[suffix]"
size
Minimum field size. If no size is specified, whatever is needed
to print the information will be used.
.
Indicates the output should be right justified
Hi,
I am spinning up an MPI/Slurm cluster on AWS
I am attempting to script the node names discovery for any given cluster
I tried `sinfo --Node --Format=NodeHost`
sinfo truncates the hostname to the first 20 characters but AWS creates
hosts with longer names.
Is there some workaround
Basically, it appears using --mem-per-gpu instead of just --mem gives
you unlimited memory for your job.
$ srun --account=sysadm -p rtx8000 -N 1 --time=1-10:00:00
--ntasks-per-node=1 --cpus-per-task=1 --gpus=1 --mem-per-gpu=8G
--mail-type=FAIL --pty /bin/bash
rtx-07[0]:~$ find /sys/fs/cgroup/me
