[slurm-users] More cpus blocked than booked

René Neumaier r.neumaier at lrz.uni-muenchen.de
Fri Sep 13 15:35:22 UTC 2019

Dear Christoph,

On 13.09.19 11:33, Christoph Brüning wrote:
> this seems actually to be related to hyperthreading on Intel CPUs. We
> see the same behaviour on our Xeon Gold 6134 Nodes running Slurm 17.11.
> The Slurm FAQ ( https://slurm.schedmd.com/faq.html#cpu_count ) states:
> "Note that even on systems with hyperthreading enabled, the resources
> will generally be allocated to jobs at the level of a core. Two
> different jobs will not share a core except through the use of a
> partition OverSubscribe configuration parameter. For example, a job
> requesting resources for three tasks on a node with ThreadsPerCore=2
> will be allocated two full cores."

this was very helpful.
Thanks a lot for your feedback and the corresponding information.
Then we just hadn't noticed this so far and most of the time the AMD
Opterons will have been used for this.

Everything is ok.


René Neumaier

LMU München
Department für Geo und
Paläontologie & Geobiologie

Richard-Wagner-Str. 10
D-80333 München
Tel.: +4989-2180-6625
Fax.: +4989-2180-6601
rene.neumaier at lmu.de
r.neumaier at lrz.uni-muenchen.de

EC0E B6F6 B3FF 6324 B0C8 9452 EF6B 4E3C 2E59 F5AA

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20190913/62982bd3/attachment.sig>

More information about the slurm-users mailing list