It would be cool if CPUEfctv was available as an environment variable?
Maybe worth opening an issue?

I am of course assuming that the job gets a set of nodes with a consistent
CPUEfctv

which leads me to an idiot question - how do jobs get node information if
you request N nodes with --exclusive but the nodes have differing CPU core
counts

On Thu, 21 May 2026 at 03:05, Dan Healy via slurm-users <
[email protected]> wrote:

> I think I've found my answer. When running `scontrol show node <node>` I
> can use the variable called `CPUEfctv` to represent exactly how many
> effective schedulable CPUs are available. I'll probably wrap this up into a
> helper script too for our users plus add to our documentation.
>
> On Tue, May 19, 2026 at 1:45 PM Dan Healy <[email protected]>
> wrote:
>
>> I'm curious how to report to our users that the number of available CPU
>> cores on a given node are *(CoresPerSocket * SocketsPerBoard *
>> ThreadsPerCore) - CpuSpecCount*. In a basic example, if the node has 32
>> cores and CpuSpecCount=1, then there's actually 31 cores available. If a
>> user runs `srun -n32 -N1 bash` he/she will be denied for resources. Then,
>> If a user runs `sinfo -O cpus,nodes` shows there's 32 cores available and
>> the user is confused.
>>
>> How can the user clearly identify that in the above case, the exact
>> number of available CPU cores available are 31 and *not* 32?
>>
>> --
>> Thanks,
>>
>> Daniel Healy
>>
>
>
> --
> Thanks,
>
> Daniel Healy
>
> --
> slurm-users mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
>
-- 
slurm-users mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to