Dear all,
are experiencing problems with job backfilling having this cluster setup:
- two QOS defined within the same partition of size 1000 nodes
- QOS A has a high priority (e.g. 10000) and a node limit of 10
- QOS B has a low priority (e.g. 100) and a node limit of 990
If QOS A has 10 nodes occupied + further jobs queued, no jobs in QOS B
will be started since their priority is too low.
Here Information from slurm.conf:
SLURM_VERSION = 16.05.6
PriorityType = priority/multifactor
SchedulerParameters =
bf_max_job_user=20,bf_interval=120,default_queue_depth=50,bf_window=4320,bf_continue,max_sched_time=6,bf_resolution=1800,defer
SchedulerType = sched/backfill
Are there any suggestion on how to configure the backfilling or the QOS
setup to make jobs run in QOS B?
thanks in advance!
best regards,
Markus
--
=====================================================
Dr. Markus Stöhr
Zentraler Informatikdienst BOKU Wien / TU Wien
Wiedner Hauptstraße 8-10
1040 Wien
Tel. +43-1-58801-420754
Fax +43-1-58801-9420754
Email: [email protected]
=====================================================
