[qob] don’t use unbounded threads reading results #13634
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Cap number of threads allocated when reading results of a cda on the driver.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
CHANGELOG: fix a bug which caused jobs with a large number of partitions (close to 100k) to run out of memory on the driver after all partitions finish.
Cap number of threads allocated when reading results of a cda on the driver.
Still waiting on a test to confirm this fixes the bad scaling with number of partitions.We were reading cda results with a threadpool containing one thread per partition, but kept all but 1000 threads blocked. This changes to use a threadpool of 1000 threads.
We still use a surprising amount of memory when reading results, but this change will likely allow jobs with ~100k partitions to complete.
fixes #13533