You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@arrow.apache.org by "simonvandel (via GitHub)" <gi...@apache.org> on 2024/07/21 18:10:11 UTC
[PR] Optimize `bool_or` using `max_boolean` [arrow-rs]
simonvandel opened a new pull request, #6100:
URL: https://github.com/apache/arrow-rs/pull/6100
# Which issue does this PR close?
<!--
We generally require a GitHub issue to be filed for all bug fixes and enhancements and this helps us generate change logs for our releases. You can link an issue to this PR using the GitHub syntax. For example `Closes #123` indicates that this PR will close issue #123.
-->
Closes #.
# Rationale for this change
<!--
Why are you proposing this change? If this is already explained clearly in the issue then this section is not needed.
Explaining clearly why changes are proposed helps reviewers understand your changes and offer better suggestions for fixes.
-->
After optimizing `max_boolean` in https://github.com/apache/arrow-rs/pull/6098, I noticed that `bool_or` is AFAICT the same operation. So we might as well re-use that optimized function.
# What changes are included in this PR?
<!--
There is no need to duplicate the description in the issue here but it is sometimes worth providing a summary of the individual changes in this PR.
-->
(Depends on https://github.com/apache/arrow-rs/pull/6098)
- Make `bool_or` an alias for `max_boolean`
- Add `bool_or` benchmarks
`cargo bench --bench aggregate_kernels -- "bool/or"` shows throughput improvements between 68%-22366% on my machine.
<details><summary>Full benchmark results:</summary>
```
cargo bench --bench aggregate_kernels -- "bool/or"
bool/or nonnull mixed time: [5.4610 ns 5.5037 ns 5.5515 ns]
thrpt: [ 11805 Gelem/s 11908 Gelem/s 12001 Gelem/s]
change:
time: [-99.220% -99.189% -99.157%] (p = 0.00 < 0.05)
thrpt: [+11766% +12236% +12723%]
Performance has improved.
Found 14 outliers among 100 measurements (14.00%)
4 (4.00%) high mild
10 (10.00%) high severe
bool/or nonnull false time: [367.84 ns 381.18 ns 394.94 ns]
thrpt: [165.94 Gelem/s 171.93 Gelem/s 178.16 Gelem/s]
change:
time: [-42.777% -40.455% -38.300%] (p = 0.00 < 0.05)
thrpt: [+62.075% +67.939% +74.754%]
Performance has improved.
Found 18 outliers among 100 measurements (18.00%)
12 (12.00%) high mild
6 (6.00%) high severe
bool/or nonnull true time: [5.5665 ns 5.6436 ns 5.7398 ns]
thrpt: [ 11418 Gelem/s 11612 Gelem/s 11773 Gelem/s]
change:
time: [-99.123% -99.088% -99.048%] (p = 0.00 < 0.05)
thrpt: [+10406% +10865% +11297%]
Performance has improved.
Found 13 outliers among 100 measurements (13.00%)
2 (2.00%) high mild
11 (11.00%) high severe
bool/or nullable mixed time: [9.5163 ns 9.5980 ns 9.6975 ns]
thrpt: [6758.1 Gelem/s 6828.1 Gelem/s 6886.7 Gelem/s]
change:
time: [-99.572% -99.555% -99.537%] (p = 0.00 < 0.05)
thrpt: [+21509% +22366% +23247%]
Performance has improved.
Found 14 outliers among 100 measurements (14.00%)
3 (3.00%) high mild
11 (11.00%) high severe
bool/or nullable false time: [1.0237 µs 1.0348 µs 1.0481 µs]
thrpt: [62.528 Gelem/s 63.331 Gelem/s 64.017 Gelem/s]
change:
time: [-50.620% -48.865% -47.217%] (p = 0.00 < 0.05)
thrpt: [+89.454% +95.562% +102.51%]
Performance has improved.
Found 11 outliers among 100 measurements (11.00%)
3 (3.00%) high mild
8 (8.00%) high severe
bool/or nullable true time: [9.7835 ns 9.9299 ns 10.108 ns]
thrpt: [6483.6 Gelem/s 6599.9 Gelem/s 6698.6 Gelem/s]
change:
time: [-99.505% -99.484% -99.463%] (p = 0.00 < 0.05)
thrpt: [+18515% +19298% +20115%]
Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
2 (2.00%) high mild
```
</details>
# Are there any user-facing changes?
Faster `bool_or`.
<!--
If there are user-facing changes then we may require documentation to be updated before approving the PR.
-->
<!---
If there are any breaking changes to public APIs, please add the `breaking change` label.
-->
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] Optimize `bool_or` using `max_boolean` [arrow-rs]
Posted by "Dandandan (via GitHub)" <gi...@apache.org>.
Dandandan merged PR #6100:
URL: https://github.com/apache/arrow-rs/pull/6100
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] Optimize `bool_or` using `max_boolean` [arrow-rs]
Posted by "Dandandan (via GitHub)" <gi...@apache.org>.
Dandandan commented on PR #6100:
URL: https://github.com/apache/arrow-rs/pull/6100#issuecomment-2244585903
Can you rebase/solve the merge conflict? @simonvandel
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] Optimize `bool_or` using `max_boolean` [arrow-rs]
Posted by "simonvandel (via GitHub)" <gi...@apache.org>.
simonvandel commented on PR #6100:
URL: https://github.com/apache/arrow-rs/pull/6100#issuecomment-2244844368
> Can you rebase/solve the merge conflict? @simonvandel
Done
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org