Hi Chris
Yes, we're using constructions like this in our job_submit.lua:
-- For --gres=gpu:N requests
if job_desc.tres_per_node ~= nil then
slurm.user_msg("Tres_per_node => " .. job_desc.tres_per_node)
else
slurm.user_msg("Tres_per_node => is nil")
end
-- For --gpus=N requests
if job_desc.tres_per_job ~= nil then
slurm.user_msg("Tres_per_job => " .. job_desc.tres_per_job)
else
slurm.user_msg("Tres_per_job => is nil")
end
Best
Niels Carl Hansen
University of Aarhus
Denmark
On 25/11/2020 18.35, Christopher Benjamin Coffey wrote:
Hi Niels,
Have you found a solution? I just noticed this recently as well. We've
traditionally told our users to use --gres:gpu:tesla:# for requesting gpus.
Then, our job submit plugin would detect the gres ask, specifically gpu, and
set a a qos, and partition accordingly. Unforutnately I started pushing folks
to use -G1, or --gpus=1 for simplicity and just realized our plugin does not
pick up gpu stuff anymore. Looking at the docs here:
https://slurm.schedmd.com/job_submit_plugins.html
The lua portion says that the function: " _get_job_req_field()" should
highlight the attributes available. Yet, the gpu request specifics don't appear to be
there in the code:
https://github.com/SchedMD/slurm/blob/master/src/plugins/job_submit/lua/job_submit_lua.c
Here is hoping slurm devs can add them, or point to the correct attributes to use. I did
try "gpus_per_task" but that didn't work.
Best,
Chris
