[Enhancement] Removes reserve() from array_agg(). #56958
Merged
Conversation
This file contains hidden or 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
trueeyu
approved these changes
Mar 15, 2025
[Java-Extensions Incremental Coverage Report]✅ pass : 0 / 0 (0%) |
[FE Incremental Coverage Report]✅ pass : 0 / 0 (0%) |
[BE Incremental Coverage Report]✅ pass : 0 / 0 (0%) |
alvin-celerdata
approved these changes
Mar 17, 2025
|
@jkim650 thank you for this PR, could you sign-off this PR by |
|
@mergify backport branch-3.4 branch3.3 |
❌ No backport have been created
GitHub error: |
|
@mergify backport branch-3.3 |
✅ Backports have been created
|
mergify bot
pushed a commit
that referenced
this pull request
Mar 17, 2025
(cherry picked from commit 3757748)
mergify bot
pushed a commit
that referenced
this pull request
Mar 17, 2025
(cherry picked from commit 3757748)
This was referenced Mar 17, 2025
wanpengfei-git
pushed a commit
that referenced
this pull request
Mar 17, 2025
…57009) Co-authored-by: Jinho Kim <[email protected]>
wanpengfei-git
pushed a commit
that referenced
this pull request
Mar 17, 2025
…57008) Co-authored-by: Jinho Kim <[email protected]>
crossoverJie
pushed a commit
to crossoverJie/starrocks
that referenced
this pull request
Mar 26, 2025
Signed-off-by: crossoverJie <[email protected]>
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Outdated suggestions cannot be applied.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.
Why I'm doing:
While debugging a slow query which took a few minutes, we found that it was due to skewed data.
But it was not clear why the skewed data caused such a long latency. After further investigation, I found that it was due to unnecessary
reserve()calls inarray_agg().After appending a large array to a result column in one aggregated row, following
finalize_to_column()s even with a small number of elements took a few ms likely to allocate new memory and copy existing values. Thousands of aggregated rows with a few ms added a lot of latency.Repeatedly calling
reserve()with small increases is harmful. Withoutreserve(), in the following append calls, std::vector would be able to increase the capacity exponentially efficiently.What I'm doing:
Removes
reserve().In a test on skewed data, a latency of a query with array_agg() decreased from 2m45s to 21s.
What type of PR is this:
Does this PR entail a change in behavior?
If yes, please specify the type of change:
Checklist:
Bugfix cherry-pick branch check: