-
-
Notifications
You must be signed in to change notification settings - Fork 2k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
perf: Add ArrayChunks
to optimize codegen of BatchDecoder
#17632
Merged
Conversation
This file contains 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
github-actions
bot
added
internal
An internal refactor or improvement
python
Related to Python Polars
rust
Related to Rust Polars
labels
Jul 15, 2024
coastalwhite
force-pushed
the
refactor-parquet-ops
branch
2 times, most recently
from
July 15, 2024 11:54
c1163ae
to
082de50
Compare
coastalwhite
changed the title
refactor: Extract parquet
perf: Extract parquet Jul 15, 2024
op
to traitop
to trait
coastalwhite
changed the title
perf: Extract parquet
perf: Jul 15, 2024
op
to traitArrayChunks
and Mmap
in parquet
coastalwhite
changed the title
perf:
perf: Add Jul 15, 2024
ArrayChunks
and Mmap
in parquetArrayChunks
and Mmap
in parquet
coastalwhite
force-pushed
the
refactor-parquet-ops
branch
3 times, most recently
from
July 15, 2024 13:23
1a8624b
to
17321ee
Compare
This increases visibility and allows for better control of inlining. We also changed `ChunksExact` to `ArrayChunks`. And this seems to help a lot with the code generation. As a result, we get the following benchmark: ``` Benchmark 1: After Optimization Time (mean ± σ): 18.095 s ± 0.135 s [User: 10.459 s, System: 7.564 s] Range (min … max): 17.976 s … 18.412 s 10 runs Benchmark 2: Before Optimization Time (mean ± σ): 20.863 s ± 0.041 s [User: 13.129 s, System: 7.648 s] Range (min … max): 20.794 s … 20.918 s 10 runs Summary After Optimization ran 1.15 ± 0.01 times faster than Before Optimization ```
coastalwhite
force-pushed
the
refactor-parquet-ops
branch
from
July 15, 2024 13:59
17321ee
to
457fc94
Compare
coastalwhite
requested review from
ritchie46,
stinodego,
orlp and
c-peters
as code owners
July 15, 2024 13:59
coastalwhite
changed the title
perf: Add
perf: Add Jul 15, 2024
ArrayChunks
and Mmap
in parquetArrayChunks
to optimize codegen of BatchDecoder
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #17632 +/- ##
==========================================
+ Coverage 80.65% 80.70% +0.05%
==========================================
Files 1484 1485 +1
Lines 195453 195485 +32
Branches 2780 2782 +2
==========================================
+ Hits 157636 157767 +131
+ Misses 37307 37206 -101
- Partials 510 512 +2 ☔ View full report in Codecov by Sentry. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Labels
accepted
Ready for implementation
internal
An internal refactor or improvement
performance
Performance issues or improvements
python
Related to Python Polars
rust
Related to Rust Polars
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.
This increases visibility and allows for better control of inlining.
From the first commit we get the following benchmark.