[slurm-users] cpu limit issue

[Mahmood Naderan]

Hi,
I see that although I have specified cpu limit of 12 for a user, his job
only utilizes 8 cores.



[root at rocks7 ~]# sacctmgr list association
format=partition,account,user,grptres,maxwall
 Partition    Account       User       GrpTRES     MaxWall
---------- ---------- ---------- ------------- -----------
   emerald         z3 noor cpu=12,mem=1+ 30-00:00:00



[noor at rocks7 ~]$ grep nprocshared trimer.gjf
%nprocshared=12
[noor at rocks7 ~]$ cat trimer.sh
#!/bin/bash
#SBATCH --output=trimer.out
#SBATCH --job-name=trimer
#SBATCH --ntasks=12
#SBATCH --mem=18GB
#SBATCH --partition=EMERALD
g09 trimer.gjf
[noor at rocks7 ~]$ rocks run host compute-0-3 "ps aux | grep l401.exe"
noor 30520  796  1.3 21078896 855696 ?     Rl   23:27  16:11
/usr/local/chem/...........




Any idea?


Regards,
Mahmood
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20180710/5000c0a7/attachment.html>