(arrow-rs) branch main updated: Add benchmark for casting to RunEndEncoded (REE) (#8710)

[alamb]

This is an automated email from the ASF dual-hosted git repository.

alamb pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/arrow-rs.git


The following commit(s) were added to refs/heads/main by this push:
     new 62df32e29d Add benchmark for casting to RunEndEncoded (REE) (#8710)
62df32e29d is described below

commit 62df32e29d03b69c3eeeb12268fdd539ebd00098
Author: Vegard Stikbakke <vegard.stikba...@gmail.com>
AuthorDate: Mon Oct 27 20:24:02 2025 +0100

    Add benchmark for casting to RunEndEncoded (REE) (#8710)
    
    Closes #8709.
    
    Adds bench `cast_ree` which can be run with `cargo bench --bench
    cast_ree`.
    
    ---------
    
    Co-authored-by: Andrew Lamb <and...@nerdnetworks.org>
---
 arrow-cast/Cargo.toml         |  1 +
 arrow/benches/cast_kernels.rs | 40 ++++++++++++++++++++++++++++++++++++++++
 2 files changed, 41 insertions(+)

diff --git a/arrow-cast/Cargo.toml b/arrow-cast/Cargo.toml
index f3309783fb..fb5ad1af3d 100644
--- a/arrow-cast/Cargo.toml
+++ b/arrow-cast/Cargo.toml
@@ -75,3 +75,4 @@ harness = false
 [[bench]]
 name = "parse_decimal"
 harness = false
+
diff --git a/arrow/benches/cast_kernels.rs b/arrow/benches/cast_kernels.rs
index a54529c8d1..040c118a1e 100644
--- a/arrow/benches/cast_kernels.rs
+++ b/arrow/benches/cast_kernels.rs
@@ -359,6 +359,46 @@ fn add_benchmark(c: &mut Criterion) {
     c.bench_function("cast binary view to string view", |b| {
         b.iter(|| cast_array(&binary_view_array, DataType::Utf8View))
     });
+
+    c.bench_function("cast string single run to ree<int32>", |b| {
+        let source_array = StringArray::from(vec!["a"; 8192]);
+        let array_ref = Arc::new(source_array) as ArrayRef;
+        let target_type = DataType::RunEndEncoded(
+            Arc::new(Field::new("run_ends", DataType::Int32, false)),
+            Arc::new(Field::new("values", DataType::Utf8, true)),
+        );
+        b.iter(|| cast(&array_ref, &target_type).unwrap());
+    });
+
+    c.bench_function("cast runs of 10 string to ree<int32>", |b| {
+        let source_array: Int32Array = (0..8192).map(|i| i / 10).collect();
+        let array_ref = Arc::new(source_array) as ArrayRef;
+        let target_type = DataType::RunEndEncoded(
+            Arc::new(Field::new("run_ends", DataType::Int32, false)),
+            Arc::new(Field::new("values", DataType::Int32, true)),
+        );
+        b.iter(|| cast(&array_ref, &target_type).unwrap());
+    });
+
+    c.bench_function("cast runs of 1000 int32s to ree<int32>", |b| {
+        let source_array: Int32Array = (0..8192).map(|i| i / 1000).collect();
+        let array_ref = Arc::new(source_array) as ArrayRef;
+        let target_type = DataType::RunEndEncoded(
+            Arc::new(Field::new("run_ends", DataType::Int32, false)),
+            Arc::new(Field::new("values", DataType::Int32, true)),
+        );
+        b.iter(|| cast(&array_ref, &target_type).unwrap());
+    });
+
+    c.bench_function("cast no runs of int32s to ree<int32>", |b| {
+        let source_array: Int32Array = (0..8192).collect();
+        let array_ref = Arc::new(source_array) as ArrayRef;
+        let target_type = DataType::RunEndEncoded(
+            Arc::new(Field::new("run_ends", DataType::Int32, false)),
+            Arc::new(Field::new("values", DataType::Int32, true)),
+        );
+        b.iter(|| cast(&array_ref, &target_type).unwrap());
+    });
 }
 
 criterion_group!(benches, add_benchmark);