Hello Franco, Thank you. I was trying to understand the need for a job_submit plugin in the example that is presented on the gres page. So if I add the lines you indicate to my config files I should be able to request a certain count of “bandwidth” per job and slurm will ensure that there is only a certain number of jobs running requiring less than the count listed in gres correct? Is it possible to do this on a per node basis as well as a per partition basis?
Regards, Brian
smime.p7s
Description: S/MIME cryptographic signature
