Skip to content

Add Parquet arrow_reader benchmarks for {u}int{8,16} columns #7484

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

Merged
merged 16 commits into from
May 8, 2025
Merged

Add Parquet arrow_reader benchmarks for {u}int{8,16} columns #7484

Changes from all commits
Commits
Show all changes
16 commits
Select commit Hold shift + click to select a range
97418a7
allow for reading improperly encode UINT_8 and UINT_16 parquet data
etseidl Jan 31, 2025
6ca4aad
add some benchmarks
etseidl Feb 5, 2025
244e4b4
remove print
etseidl Feb 7, 2025
75adc3d
checkpoint some experimental code
etseidl Feb 13, 2025
01d6c13
Merge remote-tracking branch 'origin/main' into uint8_tests
etseidl Apr 3, 2025
cf713ac
Merge remote-tracking branch 'origin/main' into fix_uint_cast
etseidl Apr 3, 2025
863e40f
Merge remote-tracking branch 'origin/main' into fix_uint_cast
etseidl Apr 4, 2025
21c0bb0
Merge branch 'uint8_tests' into fix_uint_cast
etseidl Apr 4, 2025
3d2b87f
checkpoint
etseidl Apr 7, 2025
023f7bd
Merge remote-tracking branch 'origin/main' into fix_uint_cast
etseidl Apr 8, 2025
db0823a
add a few more types
etseidl Apr 10, 2025
500a631
Merge remote-tracking branch 'origin/main' into fix_uint_cast
etseidl Apr 10, 2025
1df021a
Merge branch 'fix_uint_cast' of github.com:etseidl/arrow-rs into fix_…
etseidl Apr 10, 2025
dd9e262
modify comment
etseidl Apr 10, 2025
6f583c4
another edit
etseidl Apr 10, 2025
facbc76
Revert non benchmark code
alamb May 8, 2025
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
90 changes: 90 additions & 0 deletions parquet/benches/arrow_reader.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -76,6 +76,18 @@ fn build_test_schema() -> SchemaDescPtr {
OPTIONAL FIXED_LEN_BYTE_ARRAY (8) optional_flba8_leaf;
REQUIRED FIXED_LEN_BYTE_ARRAY (16) mandatory_flba16_leaf;
OPTIONAL FIXED_LEN_BYTE_ARRAY (16) optional_flba16_leaf;
REQUIRED INT32 mandatory_uint8_leaf (INTEGER(8, false));
OPTIONAL INT32 optional_uint8_leaf (INTEGER(8, false));
REQUIRED INT32 mandatory_uint16_leaf (INTEGER(16, false));
OPTIONAL INT32 optional_uint16_leaf (INTEGER(16, false));
REQUIRED INT32 mandatory_uint32_leaf (INTEGER(32, false));
OPTIONAL INT32 optional_uint32_leaf (INTEGER(32, false));
REQUIRED INT32 mandatory_int8_leaf (INTEGER(8, true));
OPTIONAL INT32 optional_int8_leaf (INTEGER(8, true));
REQUIRED INT32 mandatory_int16_leaf (INTEGER(16, true));
OPTIONAL INT32 optional_int16_leaf (INTEGER(16, true));
REQUIRED INT64 mandatory_uint64_leaf (INTEGER(64, false));
OPTIONAL INT64 optional_uint64_leaf (INTEGER(64, false));
}
";
parse_message_type(message_type)
Expand Down Expand Up @@ -1280,6 +1292,18 @@ fn add_benches(c: &mut Criterion) {
let string_list_desc = schema.column(14);
let mandatory_binary_column_desc = schema.column(15);
let optional_binary_column_desc = schema.column(16);
let mandatory_uint8_column_desc = schema.column(27);
let optional_uint8_column_desc = schema.column(28);
let mandatory_uint16_column_desc = schema.column(29);
let optional_uint16_column_desc = schema.column(30);
let mandatory_uint32_column_desc = schema.column(31);
let optional_uint32_column_desc = schema.column(32);
let mandatory_int8_column_desc = schema.column(33);
let optional_int8_column_desc = schema.column(34);
let mandatory_int16_column_desc = schema.column(35);
let optional_int16_column_desc = schema.column(36);
let mandatory_uint64_column_desc = schema.column(37);
let optional_uint64_column_desc = schema.column(38);

// primitive / int32 benchmarks
// =============================
Expand All @@ -1294,6 +1318,61 @@ fn add_benches(c: &mut Criterion) {
);
group.finish();

// primitive int32 / logical uint8 benchmarks
let mut group = c.benchmark_group("arrow_array_reader/UInt8Array");
bench_primitive::<Int32Type>(
&mut group,
&mandatory_uint8_column_desc,
&optional_uint8_column_desc,
0,
256,
);
group.finish();

// primitive int32 / logical int8 benchmarks
let mut group = c.benchmark_group("arrow_array_reader/Int8Array");
bench_primitive::<Int32Type>(
&mut group,
&mandatory_int8_column_desc,
&optional_int8_column_desc,
0,
128,
);
group.finish();

// primitive int32 / logical uint16 benchmarks
let mut group = c.benchmark_group("arrow_array_reader/UInt16Array");
bench_primitive::<Int32Type>(
&mut group,
&mandatory_uint16_column_desc,
&optional_uint16_column_desc,
0,
65536,
);
group.finish();

// primitive int32 / logical int16 benchmarks
let mut group = c.benchmark_group("arrow_array_reader/Int16Array");
bench_primitive::<Int32Type>(
&mut group,
&mandatory_int16_column_desc,
&optional_int16_column_desc,
0,
32768,
);
group.finish();

// primitive int32 / logical uint32 benchmarks
let mut group = c.benchmark_group("arrow_array_reader/UInt32Array");
bench_primitive::<Int32Type>(
&mut group,
&mandatory_uint32_column_desc,
&optional_uint32_column_desc,
0,
1000,
);
group.finish();

// primitive / int64 benchmarks
// =============================

Expand All @@ -1307,6 +1386,17 @@ fn add_benches(c: &mut Criterion) {
);
group.finish();

// primitive int64 / logical uint64 benchmarks
let mut group = c.benchmark_group("arrow_array_reader/UInt64Array");
bench_primitive::<Int64Type>(
Copy link
Contributor Author

@alamb alamb May 8, 2025 •
edited
Loading

Choose a reason for hiding this comment

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

I was confused at first -- I thought this is meant to be

Suggested change
bench_primitive::<Int64Type>(
bench_primitive::<UInt64Type>(

However, the type is the parquet type not the arrow / logical type

&mut group,
&mandatory_uint64_column_desc,
&optional_uint64_column_desc,
0,
1000,
);
group.finish();

// string benchmarks
//==============================

Expand Down
Loading