Apache DataFusion SQL Query Engine
bench: add predicate_eval SQL micro-benchmark suite for conjunctive filter evaluation (#22704)
Adrian Garcia Badaracco0f8a12123f4b
diff --git a/benchmarks/bench.sh b/benchmarks/bench.shindex abd0187b81..16b96da477 100755--- a/benchmarks/bench.sh+++ b/benchmarks/bench.sh@@ -103,6 +103,10 @@ push_down_topk: Benchmark of ORDER BY ... LIMIT over outer joins on TPC-external_aggr: External aggregation benchmark on TPC-H dataset (SF=1)wide_schema: Small-projection queries on a wide synthetic dataset (1024 cols × 256 files) — measures per-file metadata overhead(runs both 'wide' and 'narrow' subgroups: narrow is an internal baseline; the wide-vs-narrow ratio is the signal)+predicate_eval: Conjunctive (AND) filter-evaluation micro-benchmarks; each subgroup is a different predicate pattern, to test how an+ adaptive predicate-ordering system behaves across them (see https://github.com/apache/datafusion/issues/11262)+ (subgroups via BENCH_SUBGROUP: costsel, cost, selectivity, cardinality, width, scale, neutral, correlation, drift)+ (toggle a system under test with its native DATAFUSION_* env var; size data with PRED_ROWS, string width with PRED_FILL)# ClickBench Benchmarksclickbench_1: ClickBench queries against a single parquet file@@ -246,6 +250,10 @@ main() {wide_schema)data_wide_schema;;+ predicate_eval)+ # Data is generated inline by the suite's load SQL.+ echo "predicate_eval: no external data to generate"+ ;;tpcds)data_tpcds;;@@ -463,6 +471,9 @@ main() {wide_schema)run_wide_schema;;+ predicate_eval)+ run_predicate_eval+ ;;tpcds)run_tpcds;;@@ -800,6 +811,33 @@ run_push_down_topk() {bash -c "$SQL_CARGO_COMMAND"}+# Runs the predicate_eval benchmark suite: conjunctive (AND) filter-evaluation+# micro-benchmarks where each subgroup is a different predicate pattern, used to+# test how an adaptive predicate-ordering system behaves across them (see+# https://github.com/apache/datafusion/issues/11262). Data is generated inline+# by the suite's load SQL, so there is no data step.+#+# By default the suite measures DataFusion's built-in left-deep AND short-circuit+# and sets no engine config of its own. To evaluate a system under test, export+# its native DATAFUSION_* config before invoking bench.sh -- the harness reads+# SessionConfig::from_env, and that environment is inherited here, e.g.+# DATAFUSION_EXECUTION_ADAPTIVE_FILTER_REORDERING=true ./bench.sh run predicate_eval+# Suite-specific knobs (string-substituted into the load SQL, not engine config):+# BENCH_SUBGROUP run one subgroup (costsel, cost, selectivity, cardinality,+# width, scale, neutral, correlation, drift)+# PRED_ROWS synthetic row count (default 1_000_000; the scale subgroup+# overrides this per query)+# PRED_FILL filler chars per marker = string-column width knob+run_predicate_eval() {+ echo "Running predicate_eval benchmark (subgroup=${BENCH_SUBGROUP:-all}, rows=${PRED_ROWS:-1000000})..."+ debug_run env BENCH_NAME=predicate_eval \+ ${BENCH_SUBGROUP:+BENCH_SUBGROUP="${BENCH_SUBGROUP}"} \+ PRED_ROWS="${PRED_ROWS:-1000000}" \+ ${PRED_FILL:+PRED_FILL="${PRED_FILL}"} \+ ${QUERY:+BENCH_QUERY="${QUERY}"} \+ bash -c "$SQL_CARGO_COMMAND"+}+# Runs the tpch in memory (needs tpch parquet data)run_tpch_mem() {SCALE_FACTOR=$1diff --git a/benchmarks/sql_benchmarks/README.md b/benchmarks/sql_benchmarks/README.mdindex 38aa3ffbac..f92baf6e73 100644--- a/benchmarks/sql_benchmarks/README.md+++ b/benchmarks/sql_benchmarks/README.md@@ -43,6 +43,7 @@ in the community:| `tpcds` | TPC‑DS queries || `tpch` | TPC‑H queries || `wide_schema` | Small-projection queries on a wide (1024-col, 256-file) synthetic dataset; runs `wide` + `narrow` subgroups for comparison |+| `predicate_eval` | Conjunctive (AND) filter-evaluation micro-benchmarks; each subgroup is a different predicate pattern, to test how an adaptive predicate-ordering system behaves across them ([#11262](https://github.com/apache/datafusion/issues/11262)). Subgroups (`BENCH_SUBGROUP`): `costsel`, `cost`, `selectivity`, `cardinality`, `width`, `scale`, `neutral`, `correlation`, `drift`. Toggle a system under test with its native `DATAFUSION_*` env var |# Running Benchmarks@@ -95,6 +96,8 @@ Some benchmarks use custom environment variables as outlined below:| BENCH_SORTED | Used in the sort_tpch benchmark to indicate whether the lineitem table should be sorted. | false || SORTED_BY | Used in the clickbench_sorted benchmark to indicate the column to sort by. | `EventTime` || SORTED_ORDER | Used in the clickbench_sorted benchmark to indicate the sort order of the column. | `ASC` |+| PRED_ROWS | Used in the predicate_eval benchmark to size the synthetic table (the `scale` subgroup overrides this per query). | `1000000` |+| PRED_FILL | Used in the predicate_eval benchmark as the string-column width knob (filler chars per marker). | `30` |## How it worksdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cardinality/q30.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cardinality/q30.benchmarknew file mode 100644index 0000000000..760ea2ca90--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cardinality/q30.benchmark@@ -0,0 +1,7 @@+subgroup cardinality++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=cardinality+QPAD=30+DATASET=ints+NAME=cardinality_q30_k2diff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cardinality/q31.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cardinality/q31.benchmarknew file mode 100644index 0000000000..74f22715d1--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cardinality/q31.benchmark@@ -0,0 +1,7 @@+subgroup cardinality++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=cardinality+QPAD=31+DATASET=ints+NAME=cardinality_q31_k4diff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cardinality/q32.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cardinality/q32.benchmarknew file mode 100644index 0000000000..b6b69c3852--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cardinality/q32.benchmark@@ -0,0 +1,7 @@+subgroup cardinality++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=cardinality+QPAD=32+DATASET=ints+NAME=cardinality_q32_k8diff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cardinality/q33.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cardinality/q33.benchmarknew file mode 100644index 0000000000..1260e68137--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cardinality/q33.benchmark@@ -0,0 +1,7 @@+subgroup cardinality++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=cardinality+QPAD=33+DATASET=ints+NAME=cardinality_q33_k16diff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/correlation/q70.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/correlation/q70.benchmarknew file mode 100644index 0000000000..ef20f7dc49--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/correlation/q70.benchmark@@ -0,0 +1,7 @@+subgroup correlation++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=correlation+QPAD=70+DATASET=corr+NAME=correlation_q70_independentdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/correlation/q71.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/correlation/q71.benchmarknew file mode 100644index 0000000000..8875f6c44e--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/correlation/q71.benchmark@@ -0,0 +1,7 @@+subgroup correlation++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=correlation+QPAD=71+DATASET=corr+NAME=correlation_q71_positivediff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/correlation/q72.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/correlation/q72.benchmarknew file mode 100644index 0000000000..8109f1439a--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/correlation/q72.benchmark@@ -0,0 +1,7 @@+subgroup correlation++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=correlation+QPAD=72+DATASET=corr+NAME=correlation_q72_antidiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cost/q10.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cost/q10.benchmarknew file mode 100644index 0000000000..9b864b8594--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cost/q10.benchmark@@ -0,0 +1,7 @@+subgroup cost++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=cost+QPAD=10+DATASET=mixed+NAME=cost_q10_expensive_firstdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cost/q11.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cost/q11.benchmarknew file mode 100644index 0000000000..296ea443b3--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/cost/q11.benchmark@@ -0,0 +1,7 @@+subgroup cost++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=cost+QPAD=11+DATASET=mixed+NAME=cost_q11_cheap_firstdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/costsel/q01.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/costsel/q01.benchmarknew file mode 100644index 0000000000..abedd1d580--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/costsel/q01.benchmark@@ -0,0 +1,7 @@+subgroup costsel++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=costsel+QPAD=01+DATASET=markers+NAME=costsel_q01_regexp_selective_lastdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/costsel/q02.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/costsel/q02.benchmarknew file mode 100644index 0000000000..f50aab6642--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/costsel/q02.benchmark@@ -0,0 +1,7 @@+subgroup costsel++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=costsel+QPAD=02+DATASET=markers+NAME=costsel_q02_regexp_selective_firstdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/costsel/q03.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/costsel/q03.benchmarknew file mode 100644index 0000000000..10c4ce184e--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/costsel/q03.benchmark@@ -0,0 +1,7 @@+subgroup costsel++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=costsel+QPAD=03+DATASET=mixed+NAME=costsel_q03_cheap_unselective_then_expensive_selectivediff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/drift/q80.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/drift/q80.benchmarknew file mode 100644index 0000000000..970adc53f8--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/drift/q80.benchmark@@ -0,0 +1,7 @@+subgroup drift++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=drift+QPAD=80+DATASET=drift+NAME=drift_q80_a_then_bdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/drift/q81.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/drift/q81.benchmarknew file mode 100644index 0000000000..93cde75ffe--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/drift/q81.benchmark@@ -0,0 +1,7 @@+subgroup drift++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=drift+QPAD=81+DATASET=drift+NAME=drift_q81_b_then_adiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/neutral/q60.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/neutral/q60.benchmarknew file mode 100644index 0000000000..039fee622b--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/neutral/q60.benchmark@@ -0,0 +1,7 @@+subgroup neutral++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=neutral+QPAD=60+DATASET=ints+NAME=neutral_q60_cheap_uniformdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/neutral/q61.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/neutral/q61.benchmarknew file mode 100644index 0000000000..edaf89b471--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/neutral/q61.benchmark@@ -0,0 +1,7 @@+subgroup neutral++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=neutral+QPAD=61+DATASET=markers+NAME=neutral_q61_expensive_uniformdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/scale/q50.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/scale/q50.benchmarknew file mode 100644index 0000000000..0bef31e14f--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/scale/q50.benchmark@@ -0,0 +1,8 @@+subgroup scale++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=scale+QPAD=50+DATASET=mixed+PRED_ROWS=5000+NAME=scale_q50_5kdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/scale/q51.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/scale/q51.benchmarknew file mode 100644index 0000000000..8f1315fb11--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/scale/q51.benchmark@@ -0,0 +1,8 @@+subgroup scale++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=scale+QPAD=51+DATASET=mixed+PRED_ROWS=100000+NAME=scale_q51_100kdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/scale/q52.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/scale/q52.benchmarknew file mode 100644index 0000000000..7ddbfc19b4--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/scale/q52.benchmark@@ -0,0 +1,8 @@+subgroup scale++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=scale+QPAD=52+DATASET=mixed+PRED_ROWS=5000000+NAME=scale_q52_5mdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/scale/q53.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/scale/q53.benchmarknew file mode 100644index 0000000000..6cea5c44a1--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/scale/q53.benchmark@@ -0,0 +1,8 @@+subgroup scale++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=scale+QPAD=53+DATASET=mixed+PRED_ROWS=50000000+NAME=scale_q53_50mdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/selectivity/q20.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/selectivity/q20.benchmarknew file mode 100644index 0000000000..077a62650d--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/selectivity/q20.benchmark@@ -0,0 +1,7 @@+subgroup selectivity++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=selectivity+QPAD=20+DATASET=ints+NAME=selectivity_q20_unselective_firstdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/selectivity/q21.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/selectivity/q21.benchmarknew file mode 100644index 0000000000..24fc6ef4cd--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/selectivity/q21.benchmark@@ -0,0 +1,7 @@+subgroup selectivity++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=selectivity+QPAD=21+DATASET=ints+NAME=selectivity_q21_selective_firstdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/width/q40.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/width/q40.benchmarknew file mode 100644index 0000000000..df66cf16a3--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/width/q40.benchmark@@ -0,0 +1,8 @@+subgroup width++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=width+QPAD=40+DATASET=markers+PRED_FILL=2+NAME=width_q40_narrowdiff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/width/q41.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/width/q41.benchmarknew file mode 100644index 0000000000..c260dc9985--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/width/q41.benchmark@@ -0,0 +1,8 @@+subgroup width++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=width+QPAD=41+DATASET=markers+PRED_FILL=30+NAME=width_q41_widediff --git a/benchmarks/sql_benchmarks/predicate_eval/benchmarks/width/q42.benchmark b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/width/q42.benchmarknew file mode 100644index 0000000000..988ff59c70--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/benchmarks/width/q42.benchmark@@ -0,0 +1,8 @@+subgroup width++template sql_benchmarks/predicate_eval/predicate_eval.benchmark.template+SUBGROUP=width+QPAD=42+DATASET=markers+PRED_FILL=170+NAME=width_q42_xwidediff --git a/benchmarks/sql_benchmarks/predicate_eval/init/cleanup.sql b/benchmarks/sql_benchmarks/predicate_eval/init/cleanup.sqlnew file mode 100644index 0000000000..48f076a9fa--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/init/cleanup.sql@@ -0,0 +1 @@+DROP TABLE IF EXISTS t;diff --git a/benchmarks/sql_benchmarks/predicate_eval/load/corr.sql b/benchmarks/sql_benchmarks/predicate_eval/load/corr.sqlnew file mode 100644index 0000000000..2d7ceb73e6--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/load/corr.sql@@ -0,0 +1,19 @@+-- Correlation dataset: a base column plus derived columns that control the+-- *conditional* selectivity of one predicate given another (its selectivity+-- among the rows that already passed the other).+--+-- x uniform [0,100)+-- x_pos = x (perfectly positively correlated: `x<k AND x_pos<k`+-- passes ~k%, not the ~k%^2 an independence assumption+-- predicts)+-- x_anti = 99 - x (anti-correlated: `x<k AND x_anti<k` is empty for k<=50)+-- ind independent control column, uniform [0,100)+--+-- PRED_ROWS sizes the table.+CREATE TABLE t AS+SELECT+ (value * 7) % 100 AS x,+ (value * 7) % 100 AS x_pos,+ 99 - ((value * 7) % 100) AS x_anti,+ (value * 13) % 100 AS ind+FROM generate_series(1, ${PRED_ROWS:-1000000});diff --git a/benchmarks/sql_benchmarks/predicate_eval/load/drift.sql b/benchmarks/sql_benchmarks/predicate_eval/load/drift.sqlnew file mode 100644index 0000000000..d5635d91dd--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/load/drift.sql@@ -0,0 +1,16 @@+-- Drift dataset: two predicates whose *relative* selectivity flips partway+-- through the scan, so whole-table selectivity differs from per-batch+-- selectivity. Rows are emitted in `seq` order, so batches observe the drift in+-- order.+--+-- a_sel = 0 is selective (~0.1%) in the first 10% of rows, unselective+-- (~50%) afterwards.+-- b_sel = 0 is the mirror: unselective early, selective late.+--+-- PRED_ROWS sizes the table.+CREATE TABLE t AS+SELECT+ value AS seq,+ CASE WHEN value < ${PRED_ROWS:-1000000} / 10 THEN value % 1000 ELSE value % 2 END AS a_sel,+ CASE WHEN value < ${PRED_ROWS:-1000000} / 10 THEN value % 2 ELSE value % 1000 END AS b_sel+FROM generate_series(1, ${PRED_ROWS:-1000000});diff --git a/benchmarks/sql_benchmarks/predicate_eval/load/ints.sql b/benchmarks/sql_benchmarks/predicate_eval/load/ints.sqlnew file mode 100644index 0000000000..1d3f030915--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/load/ints.sql@@ -0,0 +1,24 @@+-- Sixteen independent integer columns, each uniform on [0,100). The predicate+-- `cN < k` therefore has selectivity ~k%. All columns are equally cheap to+-- evaluate, so only selectivity (not cost) distinguishes orderings here. The+-- multipliers are all coprime to 100, which keeps the residues uniform and the+-- columns mutually decorrelated. PRED_ROWS sizes the table.+CREATE TABLE t AS+SELECT+ (value * 1) % 100 AS c0,+ (value * 3) % 100 AS c1,+ (value * 7) % 100 AS c2,+ (value * 9) % 100 AS c3,+ (value * 11) % 100 AS c4,+ (value * 13) % 100 AS c5,+ (value * 17) % 100 AS c6,+ (value * 19) % 100 AS c7,+ (value * 21) % 100 AS c8,+ (value * 23) % 100 AS c9,+ (value * 27) % 100 AS c10,+ (value * 29) % 100 AS c11,+ (value * 31) % 100 AS c12,+ (value * 33) % 100 AS c13,+ (value * 37) % 100 AS c14,+ (value * 39) % 100 AS c15+FROM generate_series(1, ${PRED_ROWS:-1000000});diff --git a/benchmarks/sql_benchmarks/predicate_eval/load/markers.sql b/benchmarks/sql_benchmarks/predicate_eval/load/markers.sqlnew file mode 100644index 0000000000..6b8fff9385--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/load/markers.sql@@ -0,0 +1,26 @@+-- Wide-string dataset: five markers embedded in `PRED_FILL`-wide filler so that a+-- non-matching `regexp_like` must scan the whole value (every string predicate+-- is "expensive"). Selectivities are coprime so the predicates are independent:+--+-- 'aaa' present in ~90% of rows (value % 10 <> 0)+-- 'bbb' present in ~86% of rows (value % 7 <> 0)+-- 'ccc' present in ~80% of rows (value % 5 <> 0)+-- 'ddd' present in ~75% of rows (value % 4 <> 0)+-- 'rare' present in ~0.1% of rows (value % 1009 = 5) <- the selective one+--+-- PRED_FILL sets the filler width per marker (the string-column width knob: ~6*PRED_FILL+-- chars per row), and PRED_ROWS sizes the table.+CREATE TABLE t AS+SELECT+ repeat('q', ${PRED_FILL:-30})+ || CASE WHEN value % 10 <> 0 THEN 'aaa' ELSE 'zzz' END+ || repeat('q', ${PRED_FILL:-30})+ || CASE WHEN value % 7 <> 0 THEN 'bbb' ELSE 'zzz' END+ || repeat('q', ${PRED_FILL:-30})+ || CASE WHEN value % 5 <> 0 THEN 'ccc' ELSE 'zzz' END+ || repeat('q', ${PRED_FILL:-30})+ || CASE WHEN value % 4 <> 0 THEN 'ddd' ELSE 'zzz' END+ || repeat('q', ${PRED_FILL:-30})+ || CASE WHEN value % 1009 = 5 THEN 'rare' ELSE 'zzzz' END+ || repeat('q', ${PRED_FILL:-30}) AS s+FROM generate_series(1, ${PRED_ROWS:-1000000});diff --git a/benchmarks/sql_benchmarks/predicate_eval/load/mixed.sql b/benchmarks/sql_benchmarks/predicate_eval/load/mixed.sqlnew file mode 100644index 0000000000..a51c1040da--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/load/mixed.sql@@ -0,0 +1,26 @@+-- Mixed-cost dataset: cheap integer columns (`cN < k` ~ k% selectivity)+-- alongside one wide string column carrying three markers matched by expensive+-- `regexp_like`:+--+-- 'rare' present in ~0.1% of rows (value % 1009 = 5)+-- 'ten' present in ~10% of rows (value % 10 = 0)+-- 'aaa' present in ~90% of rows (value % 10 <> 0)+--+-- This lets a single table mix cheap integer compares with expensive regexp+-- scans at independently chosen selectivities (e.g. a cheap, unselective compare+-- next to an expensive, selective regexp). PRED_FILL is the string-width knob;+-- PRED_ROWS sizes the table.+CREATE TABLE t AS+SELECT+ (value * 1) % 100 AS c0,+ (value * 3) % 100 AS c1,+ (value * 7) % 100 AS c2,+ (value * 9) % 100 AS c3,+ repeat('q', ${PRED_FILL:-30})+ || CASE WHEN value % 1009 = 5 THEN 'rare' ELSE 'zzzz' END+ || repeat('q', ${PRED_FILL:-30})+ || CASE WHEN value % 10 = 0 THEN 'ten' ELSE 'zzz' END+ || repeat('q', ${PRED_FILL:-30})+ || CASE WHEN value % 10 <> 0 THEN 'aaa' ELSE 'zzz' END+ || repeat('q', ${PRED_FILL:-30}) AS s+FROM generate_series(1, ${PRED_ROWS:-1000000});diff --git a/benchmarks/sql_benchmarks/predicate_eval/predicate_eval.benchmark.template b/benchmarks/sql_benchmarks/predicate_eval/predicate_eval.benchmark.templatenew file mode 100644index 0000000000..0030a7e946--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/predicate_eval.benchmark.template@@ -0,0 +1,34 @@+# Shared template for every predicate_eval benchmark. Each qNN.benchmark sets+# its `subgroup` directive and then includes this template with parameters:+# SUBGROUP subgroup name, also the query sub-directory (e.g. costsel)+# QPAD zero-padded query id, also the query file stem (e.g. 01)+# DATASET load script stem under load/ (e.g. markers)+# NAME criterion display name (e.g. costsel_q01_regexp_selective_last)+# Optional (consumed by the load scripts via ${...:-default}):+# PRED_ROWS synthetic row count (default 1_000_000)+# PRED_FILL filler chars per marker = string-column width knob (default 30)+#+# The run SQL lives in queries/${SUBGROUP}/q${QPAD}.sql so the WHERE clause is+# readable on its own. The table is always named `t`, so the assert and cleanup+# are uniform across datasets.+#+# The suite is implementation-agnostic and sets no engine config of its own: it+# measures DataFusion's built-in left-deep `AND` short-circuit by default. To+# evaluate a predicate-ordering system under test, set its native config via the+# environment (the bench harness builds its SessionContext with+# SessionConfig::from_env), e.g.+# DATAFUSION_EXECUTION_ADAPTIVE_FILTER_REORDERING=true++load sql_benchmarks/predicate_eval/load/${DATASET}.sql++name ${NAME}+group predicate_eval++assert I+SELECT count(*) > 0 FROM t;+----+true++run sql_benchmarks/predicate_eval/queries/${SUBGROUP}/q${QPAD}.sql++cleanup sql_benchmarks/predicate_eval/init/cleanup.sqldiff --git a/benchmarks/sql_benchmarks/predicate_eval/predicate_eval.suite b/benchmarks/sql_benchmarks/predicate_eval/predicate_eval.suitenew file mode 100644index 0000000000..aba11e06ff--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/predicate_eval.suite@@ -0,0 +1,2 @@+name = "predicate_eval"+description = "Micro-benchmarks for conjunctive (AND) filter evaluation. Each subgroup exercises a different predicate pattern (per-predicate cost, selectivity, conjunct count, string-column width, row count, correlation, selectivity drift, plus an order-neutral control) so the suite can show how an adaptive predicate-ordering system behaves across them -- the kind of change these benchmarks are meant to help drive, e.g. https://github.com/apache/datafusion/issues/11262. By default it measures DataFusion's built-in left-deep AND short-circuit and sets no engine config of its own; toggle a system under test with its native DATAFUSION_* env var (the harness reads SessionConfig::from_env), e.g. DATAFUSION_EXECUTION_ADAPTIVE_FILTER_REORDERING=true. Subgroups (BENCH_SUBGROUP): costsel, cost, selectivity, cardinality, width, scale, neutral, correlation, drift. Size synthetic data with PRED_ROWS and string-column width with PRED_FILL."diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/cardinality/q30.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/cardinality/q30.sqlnew file mode 100644index 0000000000..3be840e917--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/cardinality/q30.sql@@ -0,0 +1,6 @@+-- Hidden: cheap integer compares; `c1 < 5` matches ~5%, the `c0 < 90` family+-- ~90%. k = 2 here. q30..q33 sweep k = 2/4/8/16 with one ~5% predicate written+-- last among ~90% ones.+SELECT count(*) FROM t+WHERE c0 < 90+ AND c1 < 5;diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/cardinality/q31.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/cardinality/q31.sqlnew file mode 100644index 0000000000..4ba84f8124--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/cardinality/q31.sql@@ -0,0 +1,6 @@+-- k = 4: three ~90% compares followed by one ~5% compare. See q30.+SELECT count(*) FROM t+WHERE c0 < 90+ AND c1 < 90+ AND c2 < 90+ AND c3 < 5;diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/cardinality/q32.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/cardinality/q32.sqlnew file mode 100644index 0000000000..d9e920cc62--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/cardinality/q32.sql@@ -0,0 +1,10 @@+-- k = 8: seven ~90% compares followed by one ~5% compare. See q30.+SELECT count(*) FROM t+WHERE c0 < 90+ AND c1 < 90+ AND c2 < 90+ AND c3 < 90+ AND c4 < 90+ AND c5 < 90+ AND c6 < 90+ AND c7 < 5;diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/cardinality/q33.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/cardinality/q33.sqlnew file mode 100644index 0000000000..2408427ab7--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/cardinality/q33.sql@@ -0,0 +1,18 @@+-- k = 16: fifteen ~90% compares followed by one ~5% compare. See q30.+SELECT count(*) FROM t+WHERE c0 < 90+ AND c1 < 90+ AND c2 < 90+ AND c3 < 90+ AND c4 < 90+ AND c5 < 90+ AND c6 < 90+ AND c7 < 90+ AND c8 < 90+ AND c9 < 90+ AND c10 < 90+ AND c11 < 90+ AND c12 < 90+ AND c13 < 90+ AND c14 < 90+ AND c15 < 5;diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/correlation/q70.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/correlation/q70.sqlnew file mode 100644index 0000000000..86e33534c7--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/correlation/q70.sql@@ -0,0 +1,6 @@+-- Hidden: `x` and `ind` are independent, each ~20%, so the conjunction matches+-- ~4% and the second predicate is just as selective among the first's survivors+-- as on its own. Baseline for the correlation sweep. cf. q71, q72.+SELECT count(*) FROM t+WHERE x < 20+ AND ind < 20;diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/correlation/q71.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/correlation/q71.sqlnew file mode 100644index 0000000000..eda61cc289--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/correlation/q71.sql@@ -0,0 +1,6 @@+-- Hidden: `x_pos` is a copy of `x`, so `x < 20 AND x_pos < 20` still matches+-- ~20% (not the ~4% independence would imply) -- the second predicate removes+-- none of the first's survivors. cf. q70.+SELECT count(*) FROM t+WHERE x < 20+ AND x_pos < 20;diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/correlation/q72.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/correlation/q72.sqlnew file mode 100644index 0000000000..ff987524da--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/correlation/q72.sql@@ -0,0 +1,6 @@+-- Hidden: `x_anti` is `99 - x`, so `x < 50 AND x_anti < 50` is empty -- the+-- second predicate removes all of the first's survivors, though each matches+-- ~50% alone. cf. q70.+SELECT count(*) FROM t+WHERE x < 50+ AND x_anti < 50;diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/cost/q10.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/cost/q10.sqlnew file mode 100644index 0000000000..b089ebc7a1--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/cost/q10.sql@@ -0,0 +1,6 @@+-- Hidden: both predicates match ~10%, but `regexp_like(s, 'ten')` scans the+-- string (expensive) while `c0 < 10` is a cheap compare. Equal selectivity,+-- unequal cost; expensive one written first. cf. q11 (opposite order).+SELECT count(*) FROM t+WHERE regexp_like(s, 'ten')+ AND c0 < 10;diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/cost/q11.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/cost/q11.sqlnew file mode 100644index 0000000000..82d748c93b--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/cost/q11.sql@@ -0,0 +1,5 @@+-- Same two predicates as q10 (both ~10%; regexp expensive, compare cheap),+-- opposite written order. cf. q10.+SELECT count(*) FROM t+WHERE c0 < 10+ AND regexp_like(s, 'ten');diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/costsel/q01.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/costsel/q01.sqlnew file mode 100644index 0000000000..bc029ed5d8--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/costsel/q01.sql@@ -0,0 +1,10 @@+-- Hidden in the data: the five markers have very different selectivities --+-- 'aaa' ~90%, 'bbb' ~86%, 'ccc' ~80%, 'ddd' ~75%, 'rare' ~0.1% -- while every+-- regexp_like costs about the same. 'rare' (most selective) is written last.+-- cf. q02 (most selective written first).+SELECT count(*) FROM t+WHERE regexp_like(s, 'aaa')+ AND regexp_like(s, 'bbb')+ AND regexp_like(s, 'ccc')+ AND regexp_like(s, 'ddd')+ AND regexp_like(s, 'rare');diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/costsel/q02.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/costsel/q02.sqlnew file mode 100644index 0000000000..7f7fc61831--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/costsel/q02.sql@@ -0,0 +1,8 @@+-- Same predicates and hidden selectivities as q01 ('rare' ~0.1% is the+-- selective one, the rest 75-90%), but with 'rare' written first. cf. q01.+SELECT count(*) FROM t+WHERE regexp_like(s, 'rare')+ AND regexp_like(s, 'aaa')+ AND regexp_like(s, 'bbb')+ AND regexp_like(s, 'ccc')+ AND regexp_like(s, 'ddd');diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/costsel/q03.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/costsel/q03.sqlnew file mode 100644index 0000000000..a583a498b2--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/costsel/q03.sql@@ -0,0 +1,6 @@+-- Hidden: `c0 < 90` matches ~90% (cheap integer compare); `regexp_like(s,+-- 'rare')` matches ~0.1% (scans the wide string). The cheaper predicate is the+-- less selective one.+SELECT count(*) FROM t+WHERE c0 < 90+ AND regexp_like(s, 'rare');diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/drift/q80.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/drift/q80.sqlnew file mode 100644index 0000000000..b8cb61e85a--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/drift/q80.sql@@ -0,0 +1,7 @@+-- The non-obvious property: selectivity changes across the scan. Rows arrive in+-- `seq` order; `a_sel = 0` matches ~0.1% in the first 10% of rows and ~50%+-- after, `b_sel = 0` is the mirror -- so which predicate is more selective flips+-- partway through. cf. q81 (opposite order).+SELECT count(*) FROM t+WHERE a_sel = 0+ AND b_sel = 0;diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/drift/q81.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/drift/q81.sqlnew file mode 100644index 0000000000..d65ef475cc--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/drift/q81.sql@@ -0,0 +1,5 @@+-- Same drifting predicates as q80 (a_sel/b_sel flip which is more selective+-- partway through the scan), opposite written order. cf. q80.+SELECT count(*) FROM t+WHERE b_sel = 0+ AND a_sel = 0;diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/neutral/q60.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/neutral/q60.sqlnew file mode 100644index 0000000000..b217f56953--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/neutral/q60.sql@@ -0,0 +1,7 @@+-- Hidden: four integer compares of equal cost, each ~50% selective. Nothing is+-- selective and the costs are equal, so the predicates are interchangeable.+SELECT count(*) FROM t+WHERE c0 < 50+ AND c1 < 50+ AND c2 < 50+ AND c3 < 50;diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/neutral/q61.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/neutral/q61.sqlnew file mode 100644index 0000000000..7029a3d9f8--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/neutral/q61.sql@@ -0,0 +1,8 @@+-- Hidden: four regexp scans of about equal cost, all unselective ('aaa' ~90%,+-- 'bbb' ~86%, 'ccc' ~80%, 'ddd' ~75%). Like q60 the predicates are+-- interchangeable, but here each one is expensive.+SELECT count(*) FROM t+WHERE regexp_like(s, 'aaa')+ AND regexp_like(s, 'bbb')+ AND regexp_like(s, 'ccc')+ AND regexp_like(s, 'ddd');diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/scale/q50.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/scale/q50.sqlnew file mode 100644index 0000000000..03a0f1c0db--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/scale/q50.sql@@ -0,0 +1,6 @@+-- Same predicates as costsel/q03 (`c0 < 90` ~90% cheap, `regexp_like(s, 'rare')`+-- ~0.1% expensive). q50..q53 sweep table size; here PRED_ROWS=5_000, roughly a+-- single batch.+SELECT count(*) FROM t+WHERE c0 < 90+ AND regexp_like(s, 'rare');diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/scale/q51.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/scale/q51.sqlnew file mode 100644index 0000000000..28174a5df4--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/scale/q51.sql@@ -0,0 +1,4 @@+-- q50 at PRED_ROWS=100_000 (~12 batches). See q50.+SELECT count(*) FROM t+WHERE c0 < 90+ AND regexp_like(s, 'rare');diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/scale/q52.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/scale/q52.sqlnew file mode 100644index 0000000000..74938c4634--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/scale/q52.sql@@ -0,0 +1,4 @@+-- q50 at PRED_ROWS=5_000_000 (~610 batches). See q50.+SELECT count(*) FROM t+WHERE c0 < 90+ AND regexp_like(s, 'rare');diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/scale/q53.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/scale/q53.sqlnew file mode 100644index 0000000000..8edb4d4a05--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/scale/q53.sql@@ -0,0 +1,4 @@+-- q50 at PRED_ROWS=50_000_000 (~6100 batches); builds a ~9 GB table. See q50.+SELECT count(*) FROM t+WHERE c0 < 90+ AND regexp_like(s, 'rare');diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/selectivity/q20.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/selectivity/q20.sqlnew file mode 100644index 0000000000..3638f757a7--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/selectivity/q20.sql@@ -0,0 +1,6 @@+-- Hidden: two equally cheap integer compares of unequal selectivity -- `c4 < 95`+-- matches ~95%, `c0 < 5` matches ~5%. Less selective one written first.+-- cf. q21 (opposite order).+SELECT count(*) FROM t+WHERE c4 < 95+ AND c0 < 5;diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/selectivity/q21.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/selectivity/q21.sqlnew file mode 100644index 0000000000..5181faf387--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/selectivity/q21.sql@@ -0,0 +1,5 @@+-- Same two equally-cheap compares as q20 (`c4 < 95` ~95%, `c0 < 5` ~5%),+-- opposite written order. cf. q20.+SELECT count(*) FROM t+WHERE c0 < 5+ AND c4 < 95;diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/width/q40.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/width/q40.sqlnew file mode 100644index 0000000000..1b3df3e937--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/width/q40.sql@@ -0,0 +1,9 @@+-- Same predicate set and hidden selectivities as costsel/q01 ('rare' ~0.1%, the+-- rest 75-90%); only the string-column width differs across q40/q41/q42. Narrow:+-- PRED_FILL=2, ~12 chars/row.+SELECT count(*) FROM t+WHERE regexp_like(s, 'aaa')+ AND regexp_like(s, 'bbb')+ AND regexp_like(s, 'ccc')+ AND regexp_like(s, 'ddd')+ AND regexp_like(s, 'rare');diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/width/q41.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/width/q41.sqlnew file mode 100644index 0000000000..a03b576d9c--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/width/q41.sql@@ -0,0 +1,7 @@+-- q40 with wide strings: PRED_FILL=30, ~186 chars/row. See q40.+SELECT count(*) FROM t+WHERE regexp_like(s, 'aaa')+ AND regexp_like(s, 'bbb')+ AND regexp_like(s, 'ccc')+ AND regexp_like(s, 'ddd')+ AND regexp_like(s, 'rare');diff --git a/benchmarks/sql_benchmarks/predicate_eval/queries/width/q42.sql b/benchmarks/sql_benchmarks/predicate_eval/queries/width/q42.sqlnew file mode 100644index 0000000000..cb55d828e3--- /dev/null+++ b/benchmarks/sql_benchmarks/predicate_eval/queries/width/q42.sql@@ -0,0 +1,7 @@+-- q40 with extra-wide strings: PRED_FILL=170, ~1KB/row. See q40.+SELECT count(*) FROM t+WHERE regexp_like(s, 'aaa')+ AND regexp_like(s, 'bbb')+ AND regexp_like(s, 'ccc')+ AND regexp_like(s, 'ddd')+ AND regexp_like(s, 'rare');