Refresh parquet benchmark filter descriptions

Author: hhhizzz

parquet/benches/arrow_reader_materialization_policy.rs

@@ -133,7 +133,7 @@ fn append_utf8_view_value(builder: &mut StringViewBuilder, value: &str) {
 /// * Number of RowSelections = 14054784
 /// * Average run length of each RowSelection: 99997497 / 14054784 = 7.114
 ///
-/// The properties of this array are:
+/// A 100K-row reference generated by this shape has:
 /// * Selectivity is: 15144 / 100000 = 0.15144
 /// * Number of RowSelections = 12904
 /// * Average run length of each RowSelection: 100000 / 12904 = 7.75
@@ -171,7 +171,8 @@ fn create_ts_array(size: usize) -> ArrayRef {
     Arc::new(TimestampMillisecondArray::from(values)) as ArrayRef
 }
 
-/// Creates a RecordBatch with 100K rows and 4 columns: int64, float64, utf8View, and ts.
+/// Creates a RecordBatch with `size` rows and 4 columns: int64, float64,
+/// utf8View, and ts.
 pub(crate) fn create_record_batch(size: usize) -> RecordBatch {
     let fields = vec![
         Field::new("int64", DataType::Int64, false),
@@ -315,9 +316,9 @@ pub(crate) enum FilterType {
     /// │               │    │               │
     /// └───────────────┘    └───────────────┘
     /// ```
-    /// (1000 RowSelection of 10 rows each)
+    /// (fragmented, approx 5K selected rows in 500K)
     SelectiveUnclustered,
-    /// moderately selective (10%) clustered filter
+    /// moderately selective (~9%) unclustered filter
     /// ```text
     /// ┌───────────────┐    ┌───────────────┐
     /// │      ...      │    │               │
@@ -331,7 +332,7 @@ pub(crate) enum FilterType {
     /// │               │    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
     /// └───────────────┘    └───────────────┘
     /// ```
-    /// (10 RowSelections of 10,000 rows each)
+    /// (fragmented, approx 45K selected rows in 500K)
     ModeratelySelectiveUnclustered,
     /// unselective (99%) unclustered filter
     /// ```text
@@ -348,23 +349,23 @@ pub(crate) enum FilterType {
     /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
     /// └───────────────┘    └───────────────┘
     /// ```
-    /// (99,000 RowSelections of 10 rows each)
+    /// (fragmented, approx 495K selected rows in 500K)
     UnselectiveUnclustered,
     /// unselective (90%) clustered filter
     /// ```text
     /// ┌───────────────┐    ┌───────────────┐
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │      ...      │
     /// │               │    │               │
-    /// │               │    │               │
-    /// │               │    │     ...       │
-    /// │               │    │               │
-    /// │     ...       │    │               │
-    /// │               │    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
-    /// │               │    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
-    /// │               │    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
-    /// │               │    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
     /// └───────────────┘    └───────────────┘
     /// ```
-    /// (99 RowSelection of 10,000 rows each)
+    /// (50 selected runs of 9K rows each in 500K)
     UnselectiveClustered,
     /// `utf8View <> ''` modeling [ClickBench] [Q21-Q27]
     ///

parquet/benches/arrow_reader_row_filter.rs

@@ -135,7 +135,7 @@ fn append_utf8_view_value(builder: &mut StringViewBuilder, value: &str) {
 /// * Number of RowSelections = 14054784
 /// * Average run length of each RowSelection: 99997497 / 14054784 = 7.114
 ///
-/// The properties of this array are:
+/// A 100K-row reference generated by this shape has:
 /// * Selectivity is: 15144 / 100000 = 0.15144
 /// * Number of RowSelections = 12904
 /// * Average run length of each RowSelection: 100000 / 12904 = 7.75
@@ -173,7 +173,8 @@ fn create_ts_array(size: usize) -> ArrayRef {
     Arc::new(TimestampMillisecondArray::from(values)) as ArrayRef
 }
 
-/// Creates a RecordBatch with 100K rows and 4 columns: int64, float64, utf8View, and ts.
+/// Creates a RecordBatch with `size` rows and 4 columns: int64, float64,
+/// utf8View, and ts.
 pub(crate) fn create_record_batch(size: usize) -> RecordBatch {
     let fields = vec![
         Field::new("int64", DataType::Int64, false),
@@ -319,13 +320,134 @@ impl std::fmt::Display for AsyncStrategy {
 /// The variants correspond to the different filter patterns.
 #[derive(Clone, Copy, Debug)]
 pub(crate) enum FilterType {
+    /// point lookup: selects a single row in 500K.
+    /// ```text
+    /// ┌───────────────┐    ┌───────────────┐
+    /// │               │    │               │
+    /// │               │    │      ...      │
+    /// │               │    │               │
+    /// │               │    │               │
+    /// │      ...      │    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │               │    │               │
+    /// │               │    │      ...      │
+    /// │               │    │               │
+    /// │               │    │               │
+    /// └───────────────┘    └───────────────┘
+    /// ```
     PointLookup,
+    /// selective (1%) unclustered filter: approx 5K selected rows in 500K.
+    /// ```text
+    /// ┌───────────────┐    ┌───────────────┐
+    /// │      ...      │    │               │
+    /// │               │    │               │
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │               │
+    /// │               │    │      ...      │
+    /// │               │    │               │
+    /// │               │    │               │
+    /// │      ...      │    │               │
+    /// │               │    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │               │    │               │
+    /// └───────────────┘    └───────────────┘
+    /// ```
     SelectiveUnclustered,
+    /// moderately selective (10%) clustered filter: 50 selected runs of 1K
+    /// rows each in 500K.
+    /// ```text
+    /// ┌───────────────┐    ┌───────────────┐
+    /// │               │    │               │
+    /// │               │    │               │
+    /// │      ...      │    │      ...      │
+    /// │               │    │               │
+    /// │               │    │               │
+    /// │               │    │               │
+    /// │               │    │               │
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// └───────────────┘    └───────────────┘
+    /// ```
     ModeratelySelectiveClustered,
+    /// moderately selective (~9%) unclustered filter: approx 45K selected
+    /// rows in 500K.
+    /// ```text
+    /// ┌───────────────┐    ┌───────────────┐
+    /// │      ...      │    │               │
+    /// │               │    │               │
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │               │    │               │
+    /// │               │    │               │
+    /// │               │    │      ...      │
+    /// │      ...      │    │               │
+    /// │               │    │               │
+    /// │               │    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// └───────────────┘    └───────────────┘
+    /// ```
     ModeratelySelectiveUnclustered,
+    /// unselective (99%) unclustered filter: approx 495K selected rows in
+    /// 500K.
+    /// ```text
+    /// ┌───────────────┐    ┌───────────────┐
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │               │
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// └───────────────┘    └───────────────┘
+    /// ```
     UnselectiveUnclustered,
+    /// unselective (90%) clustered filter: 50 selected runs of 9K rows each
+    /// in 500K.
+    /// ```text
+    /// ┌───────────────┐    ┌───────────────┐
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │      ...      │
+    /// │               │    │               │
+    /// └───────────────┘    └───────────────┘
+    /// ```
     UnselectiveClustered,
+    /// composite sparse filter: `SelectiveUnclustered` AND
+    /// `ModeratelySelectiveClustered`, approx 0.1% selected rows in 500K.
+    /// ```text
+    /// ┌───────────────┐    ┌───────────────┐
+    /// │               │    │               │
+    /// │               │    │      ...      │
+    /// │               │    │               │
+    /// │               │    │               │
+    /// │      ...      │    │               │
+    /// │               │    │               │
+    /// │               │    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │               │    │               │
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │               │
+    /// └───────────────┘    └───────────────┘
+    /// ```
     Composite,
+    /// `utf8View <> ''` modeling [ClickBench] [Q21-Q27] with fragmented
+    /// short string runs and sentinel values every 1K rows.
+    /// ```text
+    /// ┌───────────────┐    ┌───────────────┐
+    /// │               │    │               │
+    /// │      ...      │    │      ...      │
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │               │
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │               │
+    /// │               │    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │               │    │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│
+    /// │      ...      │    │      ...      │
+    /// │               │    │               │
+    /// │▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒│    │               │
+    /// └───────────────┘    └───────────────┘
+    /// ```
+    ///
+    /// [ClickBench]: https://github.com/ClickHouse/ClickBench
+    /// [Q21-Q27]: https://github.com/apache/datafusion/blob/b7177234e65cbbb2dcc04c252f6acd80bb026362/benchmarks/queries/clickbench/queries.sql#L22-L28
     Utf8ViewNonEmpty,
 }