bench(parquet): add short and large string arrow_writer benchmarks#10021
Merged
etseidl merged 1 commit intoMay 26, 2026
Conversation
Adds two BYTE_ARRAY write benchmarks to `arrow_writer`: - `short_string_non_null`: 1M fixed-width 8-byte strings, the small-value hot path where mini-batch sizing should resolve to the full chunk. - `large_string_non_null`: 1024 × 256 KiB strings, the large-value case where a single value far exceeds the page byte limit. Split out of apache#9972 so the page-size fix can be measured against a stable baseline (in particular the large-string case). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Contributor
|
Thanks @adriangb 🚀 |
etseidl
approved these changes
May 26, 2026
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Which issue does this PR close?
Split out of #9972 per this review comment.
Rationale for this change
#9972 makes the parquet writer's mini-batch sizing byte-budget aware so large variable-width values don't produce oversized data pages. To measure that change against a stable baseline — and in particular to see the difference in the large-string case — these benchmarks belong on
mainfirst.What changes are included in this PR?
Adds two BYTE_ARRAY write cases to the
arrow_writercriterion bench:short_string_non_null— 1M fixed-width 8-byte strings. The small-value hot path, where byte-budget-based sub-batch sizing should always resolve to the full chunk (no granular splitting, no regression).large_string_non_null— 1024 × 256 KiB strings (256 MiB total). The large-value case: with the default 1 MiB page byte limit each value needs its own page, and awrite_batch_sizeof 1024 would otherwise buffer all 256 MiB before the post-write size check runs.No library code changes — benchmarks only.
Are there any user-facing changes?
No.
🤖 Generated with Claude Code