ARROW-10079: [Rust] Benchmark and improve count bits #8663
Closed
Conversation
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
jhorstmann
commented
Nov 14, 2020
| if self.is_dense() { | ||
| let valid_slots = match self.data.null_buffer() { | ||
| Some(b) => bit_util::count_set_bits_offset(b.data(), 0, index), | ||
| Some(b) => b.count_set_bits_offset(0, index), |
Contributor
Author
There was a problem hiding this comment.
Unrelated to this refactoring, but this code in UnionArray did not match my understanding of the array format documentation. At least I don't see any mention that the offsets array is compressed by omitting null values. Or maybe the documentation needs a better example that includes a null bitmap.
Contributor
Author
There was a problem hiding this comment.
Oh, I got it, the union format actually changed and deprecated the use of a null bitmap: 6df8620
jorgecarleitao
approved these changes
Nov 14, 2020
Member
jorgecarleitao
left a comment
jorgecarleitao
left a comment
There was a problem hiding this comment.
Really cool, @jhorstmann. LGTM
Left some minor comments, all optional.
nevi-me
approved these changes
Nov 15, 2020
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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Outdated suggestions cannot be applied.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.
This refactors the calculation of null counts by using the bitchunk iterator and
count_onesintrinsic. The performance of array creation improves by between 3%-5%. The biggest impact is on getting a slice of an existing array, for a slice length of 2048 the performance in a microbenchmark doubles.Benchmark results on a Ryzen 3700U. LLVM seems to be able to vectorize the
count_onesintrinsic, so performance on machines with better AVX units should be even higher.