[slurm-users] Serial jobs on multi-core nodes using whole compute node

Marcus Wagner wagner at itc.rwth-aachen.de
Thu Jan 2 07:50:45 UTC 2020


Hi Nicholas,

first of all, "Happy new Year".

I'm not sure, what you want to achieve. If you want to
a) schedule whole nodes all the time (e.g. also for serial jobs), your 
config seems on first sight ok.
b) schedule multiple serial jobs on the same node, you will have to replace
SelectType=select/linear
with
SelectType=select/cons_res or SelectType=select/cons_tres
You will also have to select one of the SelectTypeParameters, depending 
on your needs.


Best
Marcus

On 1/1/20 8:49 PM, Nicholas Yue wrote:
> Hi,
>
>   I am new to SLURM
>
>   I am experimenting with a small cluster, 1 head node (with GPU for 
> visualization, not for CUDA jobs yet) and 2 compute nodes
>
>   Each machine have quad core processors but no hyperthreading.
>
>   I have the following test job:
>
> ---------------------------------------------------------------------
> #!/bin/bash
> #SBATCH --job-name=producer               # Job name
> #SBATCH --output=producer.%j.log          # Standard output and error log
> #SBATCH --cpus-per-task=1
>
> /mnt/nfs/projects/workflow-languages/build/producer 
> output.$SLURM_JOB_ID.txt
> ---------------------------------------------------------------------
>
> My slurm.conf is this, is the configuration wrong ?
> ---------------------------------------------------------------------
> # slurm.conf file generated by configurator easy.html.
> # Put this file on all nodes of your cluster.
> # See the slurm.conf man page for more information.
> #
> ControlMachine=head0
> ControlAddr=192.168.0.10
> #
> #MailProg=/bin/mail
> MpiDefault=none
> #MpiParams=ports=#-#
> ProctrackType=proctrack/pgid
> ReturnToService=1
> SlurmctldPidFile=/var/run/slurm-llnl/slurmctld.pid
> #SlurmctldPort=6817
> SlurmdPidFile=/var/run/slurm-llnl/slurmd.pid
> #SlurmdPort=6818
> SlurmdSpoolDir=/var/lib/slurm-llnl/slurmd
> SlurmUser=slurm
> #SlurmdUser=root
> StateSaveLocation=/var/lib/slurm-llnl/slurmctld
> SwitchType=switch/none
> TaskPlugin=task/none
> #
> #
> # TIMERS
> #KillWait=30
> #MinJobAge=300
> #SlurmctldTimeout=120
> #SlurmdTimeout=300
> #
> #
> # SCHEDULING
> FastSchedule=1
> SchedulerType=sched/backfill
> #SchedulerPort=7321
> SelectType=select/linear
> #
> #
> # LOGGING AND ACCOUNTING
> AccountingStorageType=accounting_storage/none
> ClusterName=cluster
> #JobAcctGatherFrequency=30
> JobAcctGatherType=jobacct_gather/none
> #SlurmctldDebug=3
> SlurmctldLogFile=/var/log/slurm-llnl/slurmctld.log
> #SlurmdDebug=3
> SlurmdLogFile=/var/log/slurm-llnl/slurmd.log
> #
> #
> # COMPUTE NODES
> NodeName=head0 CoresPerSocket=4 ThreadsPerCore=1 RealMemory=32768 
> State=UNKNOWN
> NodeName=compute[0-1] CoresPerSocket=4 ThreadsPerCore=1 
> RealMemory=16384 State=UNKNOWN
> PartitionName=cfd Nodes=head0,compute[0-1] Default=YES 
> MaxTime=INFINITE State=UP
> ---------------------------------------------------------------------
>
> Cheers
> -- 
> Nicholas Yue
> Graphics - Arnold, Alembic, RenderMan, OpenGL, HDF5
> Custom Dev - C++ porting, OSX, Linux, Windows
> http://au.linkedin.com/in/nicholasyue
> https://vimeo.com/channels/naiadtools

-- 
Marcus Wagner, Dipl.-Inf.

IT Center
Abteilung: Systeme und Betrieb
RWTH Aachen University
Seffenter Weg 23
52074 Aachen
Tel: +49 241 80-24383
Fax: +49 241 80-624383
wagner at itc.rwth-aachen.de
www.itc.rwth-aachen.de

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20200102/5a8043d8/attachment.htm>


More information about the slurm-users mailing list