[feat] Improve state selection criteria in the --distribute
and the --flex-alloc-nodes
option
#3140
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The key problem with those options currently is that there is no way to select a node that is exclusively in the
idle
state. As a result, nodes inidle+drain
etc. states will be selected and jobs will be blocked forever. This PR fixes this by introducing exclusive state selection. More specifically, the following changes are made:STATE
argument implies exclusive state selection.STATE*
, meaning that a node to be selected must be at least in stateSTATE
.avail
is added which will selected all the nodes available for running a job. For Slurm, these are nodes inIDLE
,ALLOCATED
orCOMPLETING
states.Closes #2258.