hello,
I am having a hard time trying to make fairshare work the way I need. I am
now willing to make things simple and just ensure nodes get evenly
distributed between accounts in queue. what's the easiest way to do this?
thanks
hello ryan
thank you for your reply. it appears that you are right, i had created
associations for every project, user and partition and thats how `sshare`
looks like now:
1 account user part3b 10.003623
0 0.00 0.50
2 account
Fairshare is calculated based on an "association". If you look in the
manpage for sacctmgr under ENTITIES, you will see:
association
The entity used to group information consisting of
four parameters: account, cluster, partition (optional), and user.
Users can have en
Makes sense but in case I can guarantee no job will ever request more than
one partition isn’t there any work around to get fairshare calculated per
partition ?
Em ter, 29 de out de 2019 às 18:34, Christopher Samuel
escreveu:
> On 10/29/19 12:42 PM, Igor Feghali wrote:
>
> > fairshare is been ca
On 10/29/19 12:42 PM, Igor Feghali wrote:
fairshare is been calculated for the entire cluster and not per partition.
That's correct - jobs can request multiple partitions (and will run in
the first one available to service it).
All the best,
Chris
--
Chris Samuel : http://www.csamuel.or
hi there
i'm pretty new to slurm and trying to learn my way through it's many
configurations. my config looks like:
SchedulerType=sched/backfill
PriorityType=priority/multifactor
PriorityWeightAge=1
PriorityWeightFairshare=1
PriorityWeightJobSize=5000
PriorityWeightPartition=1
Priorit
