Difference between revisions of "SLURM Partition Limits"

From UFRC
Jump to navigation Jump to search
 
(9 intermediate revisions by 4 users not shown)
Line 1: Line 1:
Different sets of hardware resources presented as SLURM partitions have individual time limits.
+
[[Category:Scheduler]][[Category:Infrastructure]]
 +
{|align=right
 +
  |__TOC__
 +
  |}
 +
Jobs can run on specific HiPerGator servers or nodes based on hardware requirements. Different sets of hardware configurations are presented as '''SLURM partitions'''. See also: [[Available Node Features]]
 +
 
 +
==Time Limits==
 +
Partitions have individual time limits. Example of time limit configuration:
 +
 
 +
#SBATCH --time=4-00:00:00      # Walltime in hh:mm:ss or d-hh:mm:ss
 +
 
 
==Interactive Work==
 
==Interactive Work==
 
Partitions: hpg-dev, gpu, hpg-ai
 
Partitions: hpg-dev, gpu, hpg-ai
Line 12: Line 22:
 
==Jupyter==
 
==Jupyter==
 
* JupyterHub: Sessions are preset with individual limits shown in the menu
 
* JupyterHub: Sessions are preset with individual limits shown in the menu
* JupyterLab in Open OnDemand Maximum: 72 hours
+
* JupyterLab in Open OnDemand Maximum: 72 hours for the GPU partition, other partitions follow standard partition limits
 +
 
 
==GPU/HPG-AI Partitions==
 
==GPU/HPG-AI Partitions==
 
* Default: 10 min
 
* Default: 10 min
* Maximum: 7 days
+
* Maximum: 14 days
 +
 
 +
'''Note:''' There is '''no''' burst QOS for the gpu partitions.
 +
 
 
==Compute Partitions==
 
==Compute Partitions==
 
;Partitions: hpg-default, hpg2-compute, bigmem
 
;Partitions: hpg-default, hpg2-compute, bigmem
 +
Both the hpg-default and the hpg2-compute partitions are selected by default if no partition is specified for a job.
 
===Investment QOS===
 
===Investment QOS===
 
* Default: 10 min
 
* Default: 10 min

Latest revision as of 19:55, 22 April 2024

Jobs can run on specific HiPerGator servers or nodes based on hardware requirements. Different sets of hardware configurations are presented as SLURM partitions. See also: Available Node Features

Time Limits

Partitions have individual time limits. Example of time limit configuration:

  1. SBATCH --time=4-00:00:00 # Walltime in hh:mm:ss or d-hh:mm:ss

Interactive Work

Partitions: hpg-dev, gpu, hpg-ai

  • Default time limit if not specified (Default): 10 min
  • hpg-dev Maximum: 12 hours
  • gpu
    • Maximum: 12 hours for srun .... --pty bash -i sessions
    • Maximum: 72 hours for Jupyter sessions in Open OnDemand.
  • hpg-ai
    • Maximum: 12 hours for srun .... --pty bash -i sessions

Jupyter

  • JupyterHub: Sessions are preset with individual limits shown in the menu
  • JupyterLab in Open OnDemand Maximum: 72 hours for the GPU partition, other partitions follow standard partition limits

GPU/HPG-AI Partitions

  • Default: 10 min
  • Maximum: 14 days

Note: There is no burst QOS for the gpu partitions.

Compute Partitions

Partitions
hpg-default, hpg2-compute, bigmem

Both the hpg-default and the hpg2-compute partitions are selected by default if no partition is specified for a job.

Investment QOS

  • Default: 10 min
  • Maximum: 31 days (744 hours)

Burst QOS

  • Default: 10 min
  • Maximum: 4 days (96 hours)

Hardware Accelerated GUI

Partition
hwgui
  • Default: 10 min
  • Maximum: 4 days (96 hours)