[slurm-users] Re: Setting QoS with slurm 24.05.7

WHoops, my mistake, sorry. Is this closer to what you want: MaxTRESRunMinsPU MaxTRESRunMinsPerUser Maximum number of TRES minutes each user is able to use. This takes into consideration the time limit of running jobs. If the limit is reached, no new jobs are started until other jobs finish to all

[slurm-users] Re: Setting QoS with slurm 24.05.7

Hi I think you want one of the "MaxTRESMins*" options: MaxTRESMins=TRES=[,TRES=,...] MaxTRESMinsPJ=TRES=[,TRES=,...] MaxTRESMinsPerJob=TRES=[,TRES=,...] Maximum number of TRES minutes each job is able to use in this association. This is overridden if set directly on a user. Default is the cluster

[slurm-users] Re: [EXTERNAL] avoid using same GPU by the interactive job

Well that's kind of the core issue- without cgroups _any_ process in the job will have access to all of the GPUs on the system and there's not much more that Slurm can do about it at that point. I would have a look at the environment variable CUDA_VISIBLE_DEVICES