Skip to content
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

Support bitwise and operation in the kernel #2703

Merged
merged 2 commits into from Sep 13, 2022

Conversation

liukun4515
Copy link
Contributor

Which issue does this PR close?

part of #2702

Rationale for this change

What changes are included in this PR?

Are there any user-facing changes?

@github-actions github-actions bot added the arrow Changes to the arrow crate label Sep 12, 2022
T: ArrowNumericType,
F: Fn(T::Native, T::Native) -> T::Native,
{
if left.len() != right.len() {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We also check the length in binary, so I am afraid the checking here is somewhat redundant.

Maybe we could let binary to return Err when the 2 arrays have different length.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think it is fine, the compiler should eliminate it, and even if it doesn't the cost will be amortised over the much more expensive code below

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We also check the length in binary, so I am afraid the checking here is somewhat redundant.

Maybe we could let binary to return Err when the 2 arrays have different length.

If the binary return the ERR with checking the len, I can remove this.

"Cannot perform bitwise operation on arrays of different length".to_string(),
));
}
Ok(binary(left, right, op))
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could we directly operate on the value buffer,
as the bitwise operation is irrelevant to the data type ?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You would need to take into account both endianess, and the size of the types, but theoretically yes you could do this. I suspect it would perform worse, however, as you'll have a variable length type now

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could we directly operate on the value buffer, as the bitwise operation is irrelevant to the data type ?

@HaoYang670 we can track a issue for this optimization, and can try it after finish the bitwise kernel

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Agreed @liukun4515. And we should add some benchmarks to see whether we ccoul get performance improvement by doing this.

@alamb
Copy link
Contributor

alamb commented Sep 12, 2022

This is great -- thank you @liukun4515

Copy link
Contributor

@tustvold tustvold left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Minor test nit

// unsigned value
let left = UInt64Array::from(vec![Some(1), Some(2), None, Some(4)]);
let scalar = 7;
let expected = UInt64Array::from(vec![Some(1), Some(2), None, Some(4)]);
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could we get a test where the result is changed 😄

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Done

use std::ops::BitAnd;

// The helper function for bitwise operation with two array
fn bitwise_op<T, F>(
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

What is the difference between this function and math_op (https://github.com/apache/arrow-rs/blob/master/arrow/src/compute/kernels/arithmetic.rs#L60)?

I find the 2 have same types and function body. Could we directly reuse math_op, such as

// The helper function for bitwise operation with two array
fn bitwise_op<T, F>(
    left: &PrimitiveArray<T>,
    right: &PrimitiveArray<T>,
    op: F,
) -> Result<PrimitiveArray<T>>
where
    T: ArrowNumericType,
    F: Fn(T::Native, T::Native) -> T::Native,
{
    math_op(left, right, op)
}

?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

two difference about the math_op

  1. in the bit wise, the left and right has the same data type of T, and the output data type is also T
  2. Minor diff is the Error message.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We can use the math_op to replace the code body, but I concern the error message which can point out what operation is error

@liukun4515 liukun4515 merged commit 259a302 into apache:master Sep 13, 2022
@liukun4515 liukun4515 deleted the bitwise_op branch September 13, 2022 03:12
@ursabot
Copy link

ursabot commented Sep 13, 2022

Benchmark runs are scheduled for baseline = 0ba5c5b and contender = 259a302. 259a302 is a master commit associated with this PR. Results will be available as each benchmark for each run completes.
Conbench compare runs links:
[Skipped ⚠️ Benchmarking of arrow-rs-commits is not supported on ec2-t3-xlarge-us-east-2] ec2-t3-xlarge-us-east-2
[Skipped ⚠️ Benchmarking of arrow-rs-commits is not supported on test-mac-arm] test-mac-arm
[Skipped ⚠️ Benchmarking of arrow-rs-commits is not supported on ursa-i9-9960x] ursa-i9-9960x
[Skipped ⚠️ Benchmarking of arrow-rs-commits is not supported on ursa-thinkcentre-m75q] ursa-thinkcentre-m75q
Buildkite builds:
Supported benchmarks:
ec2-t3-xlarge-us-east-2: Supported benchmark langs: Python, R. Runs only benchmarks with cloud = True
test-mac-arm: Supported benchmark langs: C++, Python, R
ursa-i9-9960x: Supported benchmark langs: Python, R, JavaScript
ursa-thinkcentre-m75q: Supported benchmark langs: C++, Java

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
arrow Changes to the arrow crate
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

5 participants