resource: round timestamp of drained ranks #5866
Merged
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Outdated suggestions cannot be applied.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.
This PR rounds the drain timestamp in the resource module while building the drain object for a
resource.status
response. The idea here is that this will allow more compression in thedrain
object in the response and also result in less lines of output influx resource drain
(when sub-second precision in the output is used)A kind of best case scenario is show by separately draining 16000 nodes, such that each rank has a different timestamp (but many are actually within the same second)
before
Note the RPC time was reduced from 0.766s to 0.292s -- probably from massively reducing the payload size.
There's only a slight difference in
flux resource status
because the command is now bound by performance of librlist operations (mostlyrlist_diff
).