Story #13760: Provide more information to SLURM to make scheduling decisions on HPC - Arvados

Actions

Copy link

Story #13760

open

Provide more information to SLURM to make scheduling decisions on HPC

Added by Peter Amstutz over 6 years ago. Updated almost 2 years ago.

Status:

New

Priority:

Normal

Assigned To:

Category:

Target version:

Start date:

Due date:

% Done:

Estimated time:

Story points:

Release:

Postponed

Release relationship:

Auto

Description

Make sure we are providing all the information we could be to ensure that SLURM can make the best scheduling decisions on HPC.

For example, the --time parameter is used by the slurm backfill scheduler, and we now have the ability to specify container time limits in #13219. (TBD: how to avoid situations where slurm sends SIGKILL to crunch-run while crunch-run is trying to write logs/outputs from a container after the container succeeded, failed, or got terminated for exceeding its time limit?)

Related issues 1 (0 open — 1 closed)