On Tuesday, 12 December 2017 7:42:51 AM AEDT Mike Cammilleri wrote:
> I guess that when this happens, the load average in 'top' can show an
> extremely elevated number due to the fact that lots of processes are
> waiting to run - but in fact the overall availability of the node is still
> quite op
Thanks for the responses. I think I didn't investigate deep enough - it appears
that although I saw many processes running and a load average of something very
high, the cgroups are indeed allocating the correct number of cores to the
jobs, and threads are simply going to wait to run on the same
Hi Darby,
thanks for your help, it is working as expected.
Regards, Götz
On Mon, Dec 11, 2017 at 3:44 PM, Vicker, Darby (JSC-EG311)
wrote:
> Use “Weight” to have slurm assign the non-IO nodes first.
>
> NodeName=pax11[00-03] Weight=2 Feature=10g
> NodeName=pax11[04-31] Weight=1
>
> See the Weig
Use “Weight” to have slurm assign the non-IO nodes first.
NodeName=pax11[00-03] Weight=2 Feature=10g
NodeName=pax11[04-31] Weight=1
See the Weight section in “man slurm.conf” but this is the key for you:
All things being equal, jobs will be allocated the nodes
with the lowest weight w
Hi everyone,
I was wondering how slurm is handling constraints.
I have a setup with 32 nodes, pax11[00-31]. The first four machines
are for I/O and have 10G network interfaces. I have marked them in the
slurm.conf node configuration with
Feature=10g
Now I'm running test jobs with -N 8 --const
