Help Center/
GaussDB/
Feature Guide(Distributed_8.x)/
Partitioned Table/
Partitioned Table Query Optimization/
Collecting Statistics on Partitioned Tables/
Collecting Statistics in Cascading Mode
Updated on 2024-06-07 GMT+08:00
Collecting Statistics in Cascading Mode
When ANALYZE | ANALYSE is used to analyze a partitioned table, the system automatically collects all partition-level statistics that comply with semantics in the partitioned table based on the specified or default PARTITION_MODE. For details about PARTITION_MODE, see the PARTITION_MODE parameter in "SQL Reference > SQL Syntax > ANALYZE | ANALYSE" in Developer Guide.
- When statistics about replication, and hash bucket partitioned tables are collected in cascading mode and PARTITION_MODE is set to ALL, the behavior is converted to the ALL COMPLETE mode.
- Cascading collection of partition-level statistics does not support the scenario where the value of default_statistics_target is a negative number.
Examples
- Create a partitioned table and insert data.
gaussdb=# CREATE TABLE t1_range_int ( c1 INT, c2 INT, c3 INT, c4 INT ) PARTITION BY RANGE(c1) ( PARTITION range_p00 VALUES LESS THAN(10), PARTITION range_p01 VALUES LESS THAN(20), PARTITION range_p02 VALUES LESS THAN(30), PARTITION range_p03 VALUES LESS THAN(40), PARTITION range_p04 VALUES LESS THAN(50) ); gaussdb=# INSERT INTO t1_range_int SELECT v,v,v,v FROM generate_series(0, 49) AS v;
- Collect statistics in cascading mode.
gaussdb=# ANALYZE t1_range_int WITH ALL; - View partition-level statistics.
gaussdb=# SELECT relname, parttype, relpages, reltuples FROM pg_partition WHERE parentid=(SELECT oid FROM pg_class WHERE relname='t1_range_int') ORDER BY relname; relname | parttype | relpages | reltuples --------------+----------+----------+----------- range_p00 | p | 4 | 9 range_p01 | p | 7 | 17 range_p02 | p | 6 | 13 range_p03 | p | 2 | 5 range_p04 | p | 4 | 9 t1_range_int | r | 0 | 0 (6 rows) gaussdb=# SELECT schemaname,tablename,partitionname,subpartitionname,attname,inherited,null_frac,avg_width,n_distinct,n_dndistinct,most_common_vals,most_common_freqs,histogram_bounds FROM pg_stats WHERE tablename='t1_range_int' ORDER BY tablename, partitionname, attname; schemaname | tablename | partitionname | subpartitionname | attname | inherited | null_frac | avg_width | n_distinct | n_dndistinct | most_common_vals | most_common_freqs | histogram_bounds ------------+--------------+---------------+------------------+---------+-----------+-----------+-----------+------------+--------------+------------------+-------------------+---------------------------------------------------------------------------------------------- ------------------------------------------------- public | t1_range_int | range_p00 | | c1 | f | 0 | 4 | -1 | -1 | | | {0,1,2,3,4,5,6,7,8,9} public | t1_range_int | range_p00 | | c2 | f | 0 | 4 | -1 | -1 | | | {0,1,2,3,4,5,6,7,8,9} public | t1_range_int | range_p00 | | c3 | f | 0 | 4 | -1 | -1 | | | {0,1,2,3,4,5,6,7,8,9} public | t1_range_int | range_p00 | | c4 | f | 0 | 4 | -1 | -1 | | | {0,1,2,3,4,5,6,7,8,9} public | t1_range_int | range_p01 | | c1 | f | 0 | 4 | -1 | -1 | | | {10,11,12,13,14,15,16,17,18,19} public | t1_range_int | range_p01 | | c2 | f | 0 | 4 | -1 | -1 | | | {10,11,12,13,14,15,16,17,18,19} public | t1_range_int | range_p01 | | c3 | f | 0 | 4 | -1 | -1 | | | {10,11,12,13,14,15,16,17,18,19} public | t1_range_int | range_p01 | | c4 | f | 0 | 4 | -1 | -1 | | | {10,11,12,13,14,15,16,17,18,19} public | t1_range_int | range_p02 | | c1 | f | 0 | 4 | -1 | -1 | | | {20,21,22,23,24,25,26,27,28,29} public | t1_range_int | range_p02 | | c2 | f | 0 | 4 | -1 | -1 | | | {20,21,22,23,24,25,26,27,28,29} public | t1_range_int | range_p02 | | c3 | f | 0 | 4 | -1 | -1 | | | {20,21,22,23,24,25,26,27,28,29} public | t1_range_int | range_p02 | | c4 | f | 0 | 4 | -1 | -1 | | | {20,21,22,23,24,25,26,27,28,29} public | t1_range_int | range_p03 | | c1 | f | 0 | 4 | -1 | -1 | | | {30,31,32,33,34,35,36,37,38,39} public | t1_range_int | range_p03 | | c2 | f | 0 | 4 | -1 | -1 | | | {30,31,32,33,34,35,36,37,38,39} public | t1_range_int | range_p03 | | c3 | f | 0 | 4 | -1 | -1 | | | {30,31,32,33,34,35,36,37,38,39} public | t1_range_int | range_p03 | | c4 | f | 0 | 4 | -1 | -1 | | | {30,31,32,33,34,35,36,37,38,39} public | t1_range_int | range_p04 | | c1 | f | 0 | 4 | -1 | -1 | | | {40,41,42,43,44,45,46,47,48,49} public | t1_range_int | range_p04 | | c2 | f | 0 | 4 | -1 | -1 | | | {40,41,42,43,44,45,46,47,48,49} public | t1_range_int | range_p04 | | c3 | f | 0 | 4 | -1 | -1 | | | {40,41,42,43,44,45,46,47,48,49} public | t1_range_int | range_p04 | | c4 | f | 0 | 4 | -1 | -1 | | | {40,41,42,43,44,45,46,47,48,49} public | t1_range_int | | | c1 | f | 0 | 4 | -1 | -1 | | | {0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33, 34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49} public | t1_range_int | | | c2 | f | 0 | 4 | -1 | -1 | | | {0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33, 34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49} public | t1_range_int | | | c3 | f | 0 | 4 | -1 | -1 | | | {0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33, 34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49} public | t1_range_int | | | c4 | f | 0 | 4 | -1 | -1 | | | {0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33, 34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49} (24 rows)
- Generate partition-level statistics of data in multiple columns.
gaussdb=# ALTER TABLE t1_range_int ADD STATISTICS ((c2, c3)); gaussdb=# ANALYZE t1_range_int WITH ALL;
- View partition-level statistics of data in multiple columns.
gaussdb=# SELECT schemaname,tablename,partitionname,subpartitionname,attname,inherited,null_frac,avg_width,n_distinct,n_dndistinct,most_common_vals,most_common_freqs,histogram_bounds FROM pg_ext_stats WHERE tablename='t1_range_int' ORDER BY tablename,partitionname,attname; schemaname | tablename | partitionname | subpartitionname | attname | inherited | null_frac | avg_width | n_distinct | n_dndistinct | most_common_vals | most_common_freqs | histogram_bounds ------------+--------------+---------------+------------------+---------+-----------+-----------+-----------+------------+--------------+------------------+-------------------+------------------ public | t1_range_int | range_p00 | | 2 3 | f | 0 | 8 | -1 | -1 | | | public | t1_range_int | range_p01 | | 2 3 | f | 0 | 8 | -1 | -1 | | | public | t1_range_int | range_p02 | | 2 3 | f | 0 | 8 | -1 | -1 | | | public | t1_range_int | range_p03 | | 2 3 | f | 0 | 8 | -1 | -1 | | | public | t1_range_int | range_p04 | | 2 3 | f | 0 | 8 | -1 | -1 | | | public | t1_range_int | | | 2 3 | f | 0 | 8 | -1 | -1 | | | (6 rows) - Create an expression index and generate partition-level statistics.
gaussdb=# CREATE INDEX t1_range_int_index ON t1_range_int(text(c1)) LOCAL; gaussdb=# ANALYZE t1_range_int WITH ALL;
- View the partition-level statistics of the expression index.
gaussdb=# SELECT schemaname,tablename,partitionname,subpartitionname,attname,inherited,null_frac,avg_width,n_distinct,n_dndistinct,most_common_vals,most_common_freqs,histogram_bounds FROM pg_stats WHERE tablename='t1_range_int_index' ORDER BY tablename,partitionname,attname; schemaname | tablename | partitionname | subpartitionname | attname | inherited | null_frac | avg_width | n_distinct | n_dndistinct | most_common_vals | most_common_freqs | histogram_bounds ------------+--------------------+--------------------+------------------+---------+-----------+-----------+-----------+------------+--------------+------------------+-------------------+----------------------------------------------------------------------------------- ------------------------------------------------------------ public | t1_range_int_index | range_p00_text_idx | | text | f | 0 | 5 | -1 | 0 | | | {0,1,2,3,4,5,6,7,8,9} public | t1_range_int_index | range_p01_text_idx | | text | f | 0 | 6 | -1 | 0 | | | {10,11,12,13,14,15,16,17,18,19} public | t1_range_int_index | range_p02_text_idx | | text | f | 0 | 6 | -1 | 0 | | | {20,21,22,23,24,25,26,27,28,29} public | t1_range_int_index | range_p03_text_idx | | text | f | 0 | 6 | -1 | 0 | | | {30,31,32,33,34,35,36,37,38,39} public | t1_range_int_index | range_p04_text_idx | | text | f | 0 | 6 | -1 | 0 | | | {40,41,42,43,44,45,46,47,48,49} public | t1_range_int_index | | | text | f | 0 | 5 | -1 | 0 | | | {0,1,10,11,12,13,14,15,16,17,18,19,2,20,21,22,23,24,25,26,27,28,29,3,30,31,32,33,3 4,35,36,37,38,39,4,40,41,42,43,44,45,46,47,48,49,5,6,7,8,9} (6 rows) - Delete the partitioned table.
gaussdb=# DROP TABLE t1_range_int;
Parent topic: Collecting Statistics on Partitioned Tables
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
The system is busy. Please try again later.
For any further questions, feel free to contact us through the chatbot.
Chatbot
