You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@arrow.apache.org by "tustvold (via GitHub)" <gi...@apache.org> on 2023/04/11 16:43:15 UTC
[GitHub] [arrow-datafusion] tustvold opened a new pull request, #5964: Specialized Cursor for StringArray and BinaryArray
tustvold opened a new pull request, #5964:
URL: https://github.com/apache/arrow-datafusion/pull/5964
# 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.
-->
Part of #5882
# 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.
-->
```
merge sorted utf8 low cardinality
time: [5.0056 ms 5.0200 ms 5.0368 ms]
change: [-24.330% -24.082% -23.834%] (p = 0.00 < 0.05)
Performance has improved.
Found 11 outliers among 100 measurements (11.00%)
5 (5.00%) high mild
6 (6.00%) high severe
sort merge utf8 low cardinality
time: [5.7681 ms 5.7938 ms 5.8190 ms]
change: [-21.867% -21.399% -20.931%] (p = 0.00 < 0.05)
Performance has improved.
merge sorted utf8 high cardinality
time: [7.8982 ms 7.9206 ms 7.9468 ms]
change: [-14.207% -13.879% -13.546%] (p = 0.00 < 0.05)
Performance has improved.
Found 7 outliers among 100 measurements (7.00%)
2 (2.00%) high mild
5 (5.00%) high severe
sort merge utf8 high cardinality
time: [9.1147 ms 9.1690 ms 9.2234 ms]
change: [-13.848% -13.085% -12.344%] (p = 0.00 < 0.05)
Performance has improved.
````
# 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.
-->
# Are these changes tested?
<!--
We typically require tests for all PRs in order to:
1. Prevent the code from being accidentally broken by subsequent changes
2. Serve as another way to document the expected behavior of the code
If tests are not included in your PR, please explain why (for example, are they covered by existing tests)?
-->
# Are there any user-facing changes?
<!--
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 `api 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
[GitHub] [arrow-datafusion] alamb commented on a diff in pull request #5964: Specialized Cursor for StringArray and BinaryArray
Posted by "alamb (via GitHub)" <gi...@apache.org>.
alamb commented on code in PR #5964:
URL: https://github.com/apache/arrow-datafusion/pull/5964#discussion_r1163293051
##########
datafusion/core/src/physical_plan/sorts/cursor.rs:
##########
@@ -97,31 +99,104 @@ impl Cursor for RowCursor {
}
}
-/// A cursor over sorted, nullable [`ArrowNativeTypeOp`]
+/// An [`Array`] that can be converted into [`FieldValues`]
+pub trait FieldArray: Array + 'static {
+ type Values: FieldValues;
+
+ fn values(&self) -> Self::Values;
+}
+
+/// A comparable set of non-nullable values
+pub trait FieldValues {
+ type Value: ?Sized;
+
+ fn len(&self) -> usize;
+
+ fn compare(a: &Self::Value, b: &Self::Value) -> Ordering;
+
+ fn value(&self, idx: usize) -> &Self::Value;
+}
+
+impl<T: ArrowPrimitiveType> FieldArray for PrimitiveArray<T> {
+ type Values = PrimitiveValues<T::Native>;
+
+ fn values(&self) -> Self::Values {
+ PrimitiveValues(self.values().clone())
+ }
+}
+
+#[derive(Debug)]
+pub struct PrimitiveValues<T: ArrowNativeTypeOp>(ScalarBuffer<T>);
+
+impl<T: ArrowNativeTypeOp> FieldValues for PrimitiveValues<T> {
+ type Value = T;
+
+ fn len(&self) -> usize {
+ self.0.len()
+ }
+
+ #[inline]
+ fn compare(a: &Self::Value, b: &Self::Value) -> Ordering {
+ T::compare(*a, *b)
+ }
+
+ #[inline]
+ fn value(&self, idx: usize) -> &Self::Value {
+ &self.0[idx]
+ }
+}
+
+impl<T: ByteArrayType> FieldArray for GenericByteArray<T> {
+ type Values = Self;
+
+ fn values(&self) -> Self::Values {
+ self.clone()
Review Comment:
Can we at least leave a reference to https://github.com/apache/arrow-rs/pull/4048 as a comment with a hint about how to reduce the codegen once https://github.com/apache/arrow-rs/pull/4048 is releaed?
--
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
[GitHub] [arrow-datafusion] tustvold commented on a diff in pull request #5964: Specialized Cursor for StringArray and BinaryArray
Posted by "tustvold (via GitHub)" <gi...@apache.org>.
tustvold commented on code in PR #5964:
URL: https://github.com/apache/arrow-datafusion/pull/5964#discussion_r1163084143
##########
datafusion/core/src/physical_plan/sorts/cursor.rs:
##########
@@ -97,31 +99,104 @@ impl Cursor for RowCursor {
}
}
-/// A cursor over sorted, nullable [`ArrowNativeTypeOp`]
+/// An [`Array`] that can be converted into [`FieldValues`]
+pub trait FieldArray: Array + 'static {
+ type Values: FieldValues;
+
+ fn values(&self) -> Self::Values;
+}
+
+/// A comparable set of non-nullable values
+pub trait FieldValues {
+ type Value: ?Sized;
+
+ fn len(&self) -> usize;
+
+ fn compare(a: &Self::Value, b: &Self::Value) -> Ordering;
+
+ fn value(&self, idx: usize) -> &Self::Value;
+}
+
+impl<T: ArrowPrimitiveType> FieldArray for PrimitiveArray<T> {
+ type Values = PrimitiveValues<T::Native>;
+
+ fn values(&self) -> Self::Values {
+ PrimitiveValues(self.values().clone())
+ }
+}
+
+#[derive(Debug)]
+pub struct PrimitiveValues<T: ArrowNativeTypeOp>(ScalarBuffer<T>);
+
+impl<T: ArrowNativeTypeOp> FieldValues for PrimitiveValues<T> {
+ type Value = T;
+
+ fn len(&self) -> usize {
+ self.0.len()
+ }
+
+ #[inline]
+ fn compare(a: &Self::Value, b: &Self::Value) -> Ordering {
+ T::compare(*a, *b)
+ }
+
+ #[inline]
+ fn value(&self, idx: usize) -> &Self::Value {
+ &self.0[idx]
+ }
+}
+
+impl<T: ByteArrayType> FieldArray for GenericByteArray<T> {
+ type Values = Self;
+
+ fn values(&self) -> Self::Values {
+ self.clone()
Review Comment:
I originally planned to make it so that this deconstructed to the underlying buffers, in order to avoid redundant codegen for strings and binary arrays. Unfortunately this needed https://github.com/apache/arrow-rs/pull/4048
In practice the additional codegen is not likely to matter, we can always revisit if it becomes a problem
--
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
[GitHub] [arrow-datafusion] tustvold merged pull request #5964: Specialized Cursor for StringArray and BinaryArray
Posted by "tustvold (via GitHub)" <gi...@apache.org>.
tustvold merged PR #5964:
URL: https://github.com/apache/arrow-datafusion/pull/5964
--
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