perf: Remove bool_val from Parquet Thrift FieldIdentifier - #9945
Merged
Conversation
Contributor
Author
|
run benchmark metadata |
|
🤖 Arrow criterion benchmark running (GKE) | trigger CPU Details (lscpu)Comparing thrift_bool (92eb8a7) to c2021f1 (merge-base) diff File an issue against this benchmark runner |
|
🤖 Arrow criterion benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagebase (merge-base)
branch
File an issue against this benchmark runner |
etseidl
changed the title
bool_val from Parquet Thrift FieldIdentifierbool_val from Parquet Thrift FieldIdentifier
etseidl
marked this pull request as ready for review
May 7, 2026 16:55
Contributor
Author
|
run benchmark metadata |
|
🤖 Arrow criterion benchmark running (GKE) | trigger CPU Details (lscpu)Comparing thrift_bool (1081e7a) to 6ce4bc8 (merge-base) diff File an issue against this benchmark runner |
|
🤖 Arrow criterion benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagebase (merge-base)
branch
File an issue against this benchmark runner |
Contributor
Author
|
bench results look weird. On x86 WS got |
Contributor
Author
|
run benchmark metadata |
Contributor
|
run benchmarks metadata |
alamb
reviewed
May 8, 2026
alamb
left a comment
alamb
left a comment
Contributor
There was a problem hiding this comment.
The code certainly looks nicer in my mind ❤️
I agree we should sort out the benchmarks
Merged
4 tasks
Contributor
|
Benchmarks seem not to be working. Maybe related to this |
Contributor
Author
|
run benchmark metadata |
|
🤖 Arrow criterion benchmark running (GKE) | trigger CPU Details (lscpu)Comparing thrift_bool (f58ff9b) to 7abb225 (merge-base) diff File an issue against this benchmark runner |
Contributor
Author
|
adriangb/datafusion-benchmarking#7 seems to have fixed benchmarks |
|
🤖 Arrow criterion benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagebase (merge-base)
branch
File an issue against this benchmark runner |
Contributor
Author
|
run benchmark metadata |
|
🤖 Arrow criterion benchmark running (GKE) | trigger CPU Details (lscpu)Comparing thrift_bool (00326ea) to 7abb225 (merge-base) diff File an issue against this benchmark runner |
|
🤖 Arrow criterion benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagebase (merge-base)
branch
File an issue against this benchmark runner |
Contributor
Author
|
run benchmark metadata |
|
🤖 Arrow criterion benchmark running (GKE) | trigger CPU Details (lscpu)Comparing thrift_bool (99de981) to 7abb225 (merge-base) diff File an issue against this benchmark runner |
|
🤖 Arrow criterion benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagebase (merge-base)
branch
File an issue against this benchmark runner |
Contributor
|
I ran the benchmarks locally on my machine (Intel Tigerlake), and this branch seems to be around 5% faster on all metadata benchmarks, compared to |
Contributor
Author
Same, but all on intel as well. I've been playing around on godbolt and honestly do not understand why there's any difference at all. However I write let field_type = self.read_byte()?;
if field_type == 0 {
// exit early
} else if field_type < 16 { // unlikely
// read full i16 field_id and return
} else {
// compute field_id from delta and return
}Lacking an aarch64 machine to profile on, it's hard to see why the benchmarks here are so different. It's possible that the difference comes from either the new |
Contributor
Author
|
run benchmark metadata |
|
🤖 Arrow criterion benchmark running (GKE) | trigger CPU Details (lscpu)Comparing thrift_bool (876fc37) to 1ffd202 (merge-base) diff File an issue against this benchmark runner |
|
🤖 Arrow criterion benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagebase (merge-base)
branch
File an issue against this benchmark runner |
alamb
approved these changes
May 14, 2026
alamb
left a comment
alamb
left a comment
Contributor
There was a problem hiding this comment.
I think this is fine and we have done enough diligence
I can turn an agent loose to try and test on an aarch machine and figure out what is going on. However, this seems perhaps unecessary
Contributor
Author
|
Thanks @alamb. I'll see if I can hunt up some AWS time and see if this is really a thing on ARM. |
Contributor
I can provide access to an AWS machine -- DM or email me a public SSH key and i'll setup one for you to test on |
Contributor
Author
Ooh, awesome! Thanks! Just sent you one via email. |
Contributor
Author
|
Running manually on an ARM instance picked the best run of the current main branch vs 2 runs of this branch. I think we can conclude it's just noise, and if there is a difference it's not really showing up strongly (although on x86 there's a definite advantage to this branch). Thanks @alamb! |
Contributor
|
Thanks for the diligence everyone. This is great |
Rich-T-kid
pushed a commit
to Rich-T-kid/arrow-rs
that referenced
this pull request
Jun 2, 2026
…#9945) # Which issue does this PR close? - Closes apache#9946. # Rationale for this change Remove some unnecessary branching from a hot path in the Thrift parser, improving performance and making for easier to read code. # What changes are included in this PR? Removes the `bool_val` field from `FieldIdentifier`. # Are these changes tested? Covered by existing tests # Are there any user-facing changes? No, this is an internal-only API
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
4 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?
Rationale for this change
Remove some unnecessary branching from a hot path in the Thrift parser, improving performance and making for easier to read code.
What changes are included in this PR?
Removes the
bool_valfield fromFieldIdentifier.Are these changes tested?
Covered by existing tests
Are there any user-facing changes?
No, this is an internal-only API