Spark: Document broadcast size-map scaling in rewrite table action#17023
Closed
szehon-ho wants to merge 1 commit into
Closed
Spark: Document broadcast size-map scaling in rewrite table action#17023szehon-ho wants to merge 1 commit into
szehon-ho wants to merge 1 commit into
Conversation
Follow-up to apache#15470. Add a comment explaining that the delete-file size map collected to the driver and broadcast to the manifest-rewrite tasks scales with the number of distinct delete files being rewritten, not the table's total file count. Generated-by: Cursor (Claude Opus 4.8)
Member
Author
|
sorry spurious pr |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What changes are included in this PR?
Follow-up to #15470. That PR introduced a per-delete-file size map that is collected to the driver and broadcast to the manifest-rewrite tasks of
RewriteTablePathSparkAction. During review I asked whether the broadcast map's memory footprint had been considered for delete-heavy tables (review comment).This adds a short comment near the broadcast site documenting the scaling assumption: the map holds one entry per distinct rewritten delete-file location, so its footprint scales with the number of distinct delete files being rewritten rather than the table's total file count. No behavior change.
The comment is added identically to the v3.5, v4.0, and v4.1 Spark modules.
How are these changes tested?
Comment-only change; no functional change, so no new tests.
AI Disclosure