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.

Sign up for GitHub

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

Jump to bottom

like benchmark for StringView #5936

Merged
merged 3 commits into from
Jun 26, 2024
Merged

like benchmark for StringView #5936

Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
3c4f7ff
`like` benchmark for StringView
alamb Jun 21, 2024
5a9f134
Merge remote-tracking branch 'apache/master' into alamb/ilike-bench
alamb Jun 24, 2024
8941cbf
Merge remote-tracking branch 'apache/master' into alamb/ilike-bench
alamb Jun 24, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
57 changes: 43 additions & 14 deletions arrow/benches/comparison_kernels.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,6 +33,10 @@ use rand::Rng;

const SIZE: usize = 65536;

fn bench_like_utf8view_scalar(arr_a: &StringViewArray, value_b: &str) {
like(arr_a, &StringViewArray::new_scalar(value_b)).unwrap();
}

fn bench_like_utf8_scalar(arr_a: &StringArray, value_b: &str) {
like(arr_a, &StringArray::new_scalar(value_b)).unwrap();
}
Expand Down Expand Up @@ -77,6 +81,8 @@ fn add_benchmark(c: &mut Criterion) {

let scalar = Float32Array::from(vec![1.0]);

// eq benchmarks

c.bench_function("eq Float32", |b| b.iter(|| eq(&arr_a, &arr_b)));
c.bench_function("eq scalar Float32", |b| {
b.iter(|| eq(&arr_a, &Scalar::new(&scalar)).unwrap())
Expand Down Expand Up @@ -161,24 +167,13 @@ fn add_benchmark(c: &mut Criterion) {
let string_right = StringArray::from_iter(array_gen);
let string_view_right = StringViewArray::from_iter(string_right.iter());

let scalar = StringArray::new_scalar("xxxx");
c.bench_function("eq scalar StringArray", |b| {
b.iter(|| {
eq(
&Scalar::new(StringArray::from_iter_values(["xxxx"])),
&string_left,
)
.unwrap()
})
b.iter(|| eq(&scalar, &string_left).unwrap())
Copy link
Contributor Author

Choose a reason for hiding this comment

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

this is a drive by cleanup to make the code neater

});

c.bench_function("eq scalar StringViewArray", |b| {
b.iter(|| {
eq(
&Scalar::new(StringViewArray::from_iter_values(["xxxx"])),
&string_view_left,
)
.unwrap()
})
b.iter(|| eq(&scalar, &string_view_left).unwrap())
});

c.bench_function("eq StringArray StringArray", |b| {
Expand All @@ -189,6 +184,8 @@ fn add_benchmark(c: &mut Criterion) {
b.iter(|| eq(&string_view_left, &string_view_right).unwrap())
});

// StringArray: LIKE benchmarks

c.bench_function("like_utf8 scalar equals", |b| {
b.iter(|| bench_like_utf8_scalar(&arr_string, "xxxx"))
});
Expand All @@ -209,6 +206,30 @@ fn add_benchmark(c: &mut Criterion) {
b.iter(|| bench_like_utf8_scalar(&arr_string, "%xx_xx%xxx"))
});

// StringViewArray: LIKE benchmarks
// Note: since like/nlike share the same implementation, we only benchmark one
c.bench_function("like_utf8view scalar equals", |b| {
b.iter(|| bench_like_utf8view_scalar(&string_view_left, "xxxx"))
});

c.bench_function("like_utf8view scalar contains", |b| {
b.iter(|| bench_like_utf8view_scalar(&string_view_left, "%xxxx%"))
});

c.bench_function("like_utf8view scalar ends with", |b| {
b.iter(|| bench_like_utf8view_scalar(&string_view_left, "xxxx%"))
});

c.bench_function("like_utf8view scalar starts with", |b| {
b.iter(|| bench_like_utf8view_scalar(&string_view_left, "%xxxx"))
});

c.bench_function("like_utf8view scalar complex", |b| {
b.iter(|| bench_like_utf8view_scalar(&string_view_left, "%xx_xx%xxx"))
});

// StringArray: NOT LIKE benchmarks

c.bench_function("nlike_utf8 scalar equals", |b| {
b.iter(|| bench_nlike_utf8_scalar(&arr_string, "xxxx"))
});
Expand All @@ -229,6 +250,8 @@ fn add_benchmark(c: &mut Criterion) {
b.iter(|| bench_nlike_utf8_scalar(&arr_string, "%xx_xx%xxx"))
});

// StringArray: ILIKE benchmarks

c.bench_function("ilike_utf8 scalar equals", |b| {
b.iter(|| bench_ilike_utf8_scalar(&arr_string, "xxXX"))
});
Expand All @@ -249,6 +272,8 @@ fn add_benchmark(c: &mut Criterion) {
b.iter(|| bench_ilike_utf8_scalar(&arr_string, "%xx_xX%xXX"))
});

// StringArray: NOT ILIKE benchmarks

c.bench_function("nilike_utf8 scalar equals", |b| {
b.iter(|| bench_nilike_utf8_scalar(&arr_string, "xxXX"))
});
Expand All @@ -269,6 +294,8 @@ fn add_benchmark(c: &mut Criterion) {
b.iter(|| bench_nilike_utf8_scalar(&arr_string, "%xx_xX%xXX"))
});

// StringArray: REGEXP_MATCHES benchmarks

c.bench_function("egexp_matches_utf8 scalar starts with", |b| {
b.iter(|| bench_regexp_is_match_utf8_scalar(&arr_string, "^xx"))
});
Expand All @@ -277,6 +304,8 @@ fn add_benchmark(c: &mut Criterion) {
b.iter(|| bench_regexp_is_match_utf8_scalar(&arr_string, "xx$"))
});

// DictionaryArray benchmarks

let strings = create_string_array::<i32>(20, 0.);
let dict_arr_a = create_dict_from_values::<Int32Type>(SIZE, 0., &strings);
let scalar = StringArray::from(vec!["test"]);
Expand Down
Loading