Adjust tuples estimate for appendrels

Richard Guo · Richard Guo · commit fbc0fe9a2efe · 2025-02-17T11:13:15.000+09:00
In set_append_rel_size(), we currently set rel->tuples to rel->rows for an appendrel. Generally, rel->tuples is the raw number of tuples in the relation and rel->rows is the estimated number of tuples after the relation's restriction clauses have been applied. Although an appendrel itself doesn't directly enforce any quals today, its child relations may. Therefore, setting rel->tuples equal to rel->rows for an appendrel isn't always appropriate. Doing so can lead to issues in cost estimates in some cases. For instance, when estimating the number of distinct values from an appendrel, we would not be able to adjust the estimate based on the restriction selectivity. This patch addresses this by setting an appendrel's tuples to the total number of tuples accumulated from each live child, which better aligns with reality. This is arguably a bug, but nobody has complained about that until now, so no back-patch. Author: Richard Guo <guofenglinux@gmail.com> Reviewed-by: Tender Wang <tndrwang@gmail.com> Reviewed-by: Alena Rybakina <a.rybakina@postgrespro.ru> Discussion: https://postgr.es/m/CAMbWs4_TG_+kVn6fjG-5GYzzukrNK57=g9eUo4gsrUG26OFawg@mail.gmail.com
diff --git a/src/backend/optimizer/path/allpaths.c b/src/backend/optimizer/path/allpaths.c
@@ -958,6 +958,7 @@ set_append_rel_size(PlannerInfo *root, RelOptInfo *rel,
 {
 	int			parentRTindex = rti;
 	bool		has_live_children;
+	double		parent_tuples;
 	double		parent_rows;
 	double		parent_size;
 	double	   *parent_attrsizes;
@@ -983,6 +984,15 @@ set_append_rel_size(PlannerInfo *root, RelOptInfo *rel,
 	/*
 	 * Initialize to compute size estimates for whole append relation.
 	 *
+	 * We handle tuples estimates by setting "tuples" to the total number of
+	 * tuples accumulated from each live child, rather than using "rows".
+	 * Although an appendrel itself doesn't directly enforce any quals, its
+	 * child relations may.  Therefore, setting "tuples" equal to "rows" for
+	 * an appendrel isn't always appropriate, and can lead to inaccurate cost
+	 * estimates.  For example, when estimating the number of distinct values
+	 * from an appendrel, we would be unable to adjust the estimate based on
+	 * the restriction selectivity (see estimate_num_groups).
+	 *
 	 * We handle width estimates by weighting the widths of different child
 	 * rels proportionally to their number of rows.  This is sensible because
 	 * the use of width estimates is mainly to compute the total relation
@@ -995,6 +1005,7 @@ set_append_rel_size(PlannerInfo *root, RelOptInfo *rel,
 	 * have zero rows and/or width, if they were excluded by constraints.
 	 */
 	has_live_children = false;
+	parent_tuples = 0;
 	parent_rows = 0;
 	parent_size = 0;
 	nattrs = rel->max_attr - rel->min_attr + 1;
@@ -1161,6 +1172,7 @@ set_append_rel_size(PlannerInfo *root, RelOptInfo *rel,
 		 */
 		Assert(childrel->rows > 0);
 
+		parent_tuples += childrel->tuples;
 		parent_rows += childrel->rows;
 		parent_size += childrel->reltarget->width * childrel->rows;
 
@@ -1207,17 +1219,12 @@ set_append_rel_size(PlannerInfo *root, RelOptInfo *rel,
 		int			i;
 
 		Assert(parent_rows > 0);
+		rel->tuples = parent_tuples;
 		rel->rows = parent_rows;
 		rel->reltarget->width = rint(parent_size / parent_rows);
 		for (i = 0; i < nattrs; i++)
 			rel->attr_widths[i] = rint(parent_attrsizes[i] / parent_rows);
 
-		/*
-		 * Set "raw tuples" count equal to "rows" for the appendrel; needed
-		 * because some places assume rel->tuples is valid for any baserel.
-		 */
-		rel->tuples = parent_rows;
-
 		/*
 		 * Note that we leave rel->pages as zero; this is important to avoid
 		 * double-counting the appendrel tree in total_table_pages.
diff --git a/src/test/regress/expected/inherit.out b/src/test/regress/expected/inherit.out
@@ -3666,3 +3666,33 @@ UPDATE errtst_parent SET partid = 30, data = data + 10 WHERE partid = 20;
 ERROR:  no partition of relation "errtst_parent" found for row
 DETAIL:  Partition key of the failing row contains (partid) = (30).
 DROP TABLE errtst_parent;
+-- Check that we have the correct tuples estimate for an appendrel
+create table tuplesest_parted (a int, b int, c float) partition by range(a);
+create table tuplesest_parted1 partition of tuplesest_parted for values from (0) to (100);
+create table tuplesest_parted2 partition of tuplesest_parted for values from (100) to (200);
+create table tuplesest_tab (a int, b int);
+insert into tuplesest_parted select i%200, i%300, i%400 from generate_series(1, 1000)i;
+insert into tuplesest_tab select i, i from generate_series(1, 100)i;
+analyze tuplesest_parted;
+analyze tuplesest_tab;
+explain (costs off)
+select * from tuplesest_tab join
+  (select b from tuplesest_parted where c < 100 group by b) sub
+  on tuplesest_tab.a = sub.b;
+                             QUERY PLAN                             
+--------------------------------------------------------------------
+ Hash Join
+   Hash Cond: (tuplesest_parted.b = tuplesest_tab.a)
+   ->  HashAggregate
+         Group Key: tuplesest_parted.b
+         ->  Append
+               ->  Seq Scan on tuplesest_parted1 tuplesest_parted_1
+                     Filter: (c < '100'::double precision)
+               ->  Seq Scan on tuplesest_parted2 tuplesest_parted_2
+                     Filter: (c < '100'::double precision)
+   ->  Hash
+         ->  Seq Scan on tuplesest_tab
+(11 rows)
+
+drop table tuplesest_parted;
+drop table tuplesest_tab;
diff --git a/src/test/regress/sql/inherit.sql b/src/test/regress/sql/inherit.sql
@@ -1483,3 +1483,24 @@ UPDATE errtst_parent SET partid = 0, data = data + 10 WHERE partid = 20;
 UPDATE errtst_parent SET partid = 30, data = data + 10 WHERE partid = 20;
 
 DROP TABLE errtst_parent;
+
+-- Check that we have the correct tuples estimate for an appendrel
+create table tuplesest_parted (a int, b int, c float) partition by range(a);
+create table tuplesest_parted1 partition of tuplesest_parted for values from (0) to (100);
+create table tuplesest_parted2 partition of tuplesest_parted for values from (100) to (200);
+
+create table tuplesest_tab (a int, b int);
+
+insert into tuplesest_parted select i%200, i%300, i%400 from generate_series(1, 1000)i;
+insert into tuplesest_tab select i, i from generate_series(1, 100)i;
+
+analyze tuplesest_parted;
+analyze tuplesest_tab;
+
+explain (costs off)
+select * from tuplesest_tab join
+  (select b from tuplesest_parted where c < 100 group by b) sub
+  on tuplesest_tab.a = sub.b;
+
+drop table tuplesest_parted;
+drop table tuplesest_tab;
