hello,
I want to limit the memory per cpu in my cluster, some settings in
slurm.conf are like this:
NodeName=c[01-10] CPUs=32 RealMemory=127360
PartitionName=C032 Nodes=c[01-10] MaxMemPerCPU=3980 DefMemPerCPU=3980
MaxCPUsPerNode=32
I have 10 nodes, each node has 127360M memory, and 32 CPUs, DefMemPerCPU =
127360M/32 = 3980M
If users don't use --mem or --mem-per-cpu to override the default settings,
it's ok.
But if users use --mem to ask more memory per cpu, for example, in
partition C032, if someone used the command:
srun -p C032 --mem=125G -n1 --pty /bin/bash
It means that he used only one cpu and asked for 125G memoey, so he used
most of the memory on that node, then it will affect other user's job, this
is invalid.
So is there any way to strictly limit the avarage memory per CPU and users
can't override it? or any way to disable --mem and --mem-per-cpu ?
best.
Ashlee