openGauss每日一练第19天｜学习openGauss收集统计信息、打印执行计划、垃圾收集和checkpoint

坚持学习openGauss数据库，坚持每天打卡。第十九天学习openGauss收集统计信息、打印执行计划、垃圾收集和checkpoint。

连接openGauss

root@modb:~# su - omm
omm@modb:~$ gsql -r
gsql ((opengauss 2.0.0 build 78689da9) compiled at 2021-03-31 21:03:52 commit 0 last mr  )
non-ssl connection (ssl connection is recommended when requiring high-security)
type "help" for help.
omm=#
复制

1.创建分区表，并用generate_series(1,N)函数对表插入数据

omm=# create table tab(id int not null,name varchar(20))
omm-# partition by range(id)
omm-# (
omm(# partition store_p0 values less than (50),
omm(# partition store_p1 values less than (100),
omm(# partition store_p2 values less than (150),
omm(# partition store_p3 values less than (200)
omm(# );
CREATE TABLE
omm=# insert into tab values(1,'a'),(51,'b'),(101,'c'),(151,'d');
INSERT 0 4
omm=# insert into tab select generate_series(1,199),'test';
INSERT 0 199
复制

2.收集表统计信息

--查看系统表中表的统计信息
omm=# select relname, relpages, reltuples from pg_class where relname = 'tab';
 relname | relpages | reltuples 
---------+----------+-----------
 tab     |        0 |         0
(1 row)
--使用ANALYZE VERBOSE语句更新统计信息，并输出表的相关信息
omm=# analyze verbose tab;
INFO:  analyzing "public.tab"(gaussdb pid=1)
INFO:  ANALYZE INFO : "tab": scanned 1 of 1 pages, containing 50 live rows and 0 dead rows; 50 rows in sample, 50 estimated total rows(gaussdb pid=1)
INFO:  ANALYZE INFO : "tab": scanned 1 of 1 pages, containing 51 live rows and 0 dead rows; 51 rows in sample, 51 estimated total rows(gaussdb pid=1)
INFO:  ANALYZE INFO : "tab": scanned 1 of 1 pages, containing 51 live rows and 0 dead rows; 51 rows in sample, 51 estimated total rows(gaussdb pid=1)
INFO:  ANALYZE INFO : "tab": scanned 1 of 1 pages, containing 51 live rows and 0 dead rows; 51 rows in sample, 51 estimated total rows(gaussdb pid=1)
ANALYZE
--查看系统表中表的统计信息
omm=# select relname, relpages, reltuples from pg_class where relname = 'tab';
 relname | relpages | reltuples 
---------+----------+-----------
 tab     |        4 |       203
(1 row)
复制

3.显示简单查询的执行计划；建立索引并显示有索引条件的执行计划

--使用默认的打印格式
SET explain_perf_mode=normal;
--显示表简单查询的执行计划
omm=# explain select * from tab;
                              QUERY PLAN                               
-----------------------------------------------------------------------
 Partition Iterator  (cost=0.00..6.03 rows=203 width=8)
   Iterations: 4
   ->  Partitioned Seq Scan on tab  (cost=0.00..6.03 rows=203 width=8)
         Selected Partitions:  1..4
(4 rows)
--有索引条件的执行计划
omm=# create index store_index1 on tab(id) local;
CREATE INDEX
omm=# explain select * from tab where id<100;
                              QUERY PLAN                               
-----------------------------------------------------------------------
 Partition Iterator  (cost=0.00..4.54 rows=102 width=8)
   Iterations: 2
   ->  Partitioned Seq Scan on tab  (cost=0.00..4.54 rows=102 width=8)
         Filter: (id < 100)
         Selected Partitions:  1..2
(5 rows)
复制

4.更新表数据，并做垃圾收集

omm=# update tab set id = id + 1 where id < 100;
UPDATE 101
omm=# VACUUM (VERBOSE, ANALYZE) tab;
INFO:  vacuuming "public.tab"(gaussdb pid=1)
DETAIL:  50 dead row versions cannot be removed yet.
There were 0 unused item pointers.
0 pages are entirely empty.
CPU 0.00s/0.00u sec elapsed 0.00 sec.
INFO:  index "store_index1" now contains 99 row versions in 2 pages(gaussdb pid=1)
DETAIL:  0 index row versions were removed.
0 index pages have been deleted, 0 are currently reusable.
CPU 0.00s/0.00u sec elapsed 0.00 sec.
INFO:  "tab": found 0 removable, 99 nonremovable row versions in 1 out of 1 pages(gaussdb pid=1)INFO:  vacuuming "public.tab"(gaussdb pid=1)
INFO:  index "store_index1" now contains 102 row versions in 2 pages(gaussdb pid=1)
DETAIL:  0 index row versions were removed.
0 index pages have been deleted, 0 are currently reusable.
CPU 0.00s/0.00u sec elapsed 0.00 sec.
INFO:  "tab": found 0 removable, 102 nonremovable row versions in 1 out of 1 pages(gaussdb pid=1)
DETAIL:  51 dead row versions cannot be removed yet.
There were 0 unused item pointers.
0 pages are entirely empty.
CPU 0.00s/0.00u sec elapsed 0.00 sec.
INFO:  vacuuming "public.tab"(gaussdb pid=1)
INFO:  index "store_index1" now contains 52 row versions in 2 pages(gaussdb pid=1)
DETAIL:  0 index row versions were removed.
0 index pages have been deleted, 0 are currently reusable.
CPU 0.00s/0.00u sec elapsed 0.00 sec.
INFO:  "tab": found 0 removable, 52 nonremovable row versions in 1 out of 1 pages(gaussdb pid=1)
DETAIL:  0 dead row versions cannot be removed yet.
There were 0 unused item pointers.
0 pages are entirely empty.
CPU 0.00s/0.00u sec elapsed 0.00 sec.
INFO:  vacuuming "public.tab"(gaussdb pid=1)
INFO:  index "store_index1" now contains 51 row versions in 2 pages(gaussdb pid=1)
DETAIL:  0 index row versions were removed.
0 index pages have been deleted, 0 are currently reusable.
CPU 0.00s/0.00u sec elapsed 0.00 sec.
INFO:  "tab": found 0 removable, 51 nonremovable row versions in 1 out of 1 pages(gaussdb pid=1)
DETAIL:  0 dead row versions cannot be removed yet.
There were 0 unused item pointers.
0 pages are entirely empty.
CPU 0.00s/0.00u sec elapsed 0.00 sec.
INFO:  analyzing "public.tab"(gaussdb pid=1)
INFO:  ANALYZE INFO : "tab": scanned 1 of 1 pages, containing 49 live rows and 50 dead rows; 49 rows in sample, 49 estimated total rows(gaussdb pid=1)
INFO:  ANALYZE INFO : "tab": scanned 1 of 1 pages, containing 51 live rows and 51 dead rows; 51 rows in sample, 51 estimated total rows(gaussdb pid=1)
INFO:  ANALYZE INFO : "tab": scanned 1 of 1 pages, containing 52 live rows and 0 dead rows; 52 rows in sample, 52 estimated total rows(gaussdb pid=1)
INFO:  ANALYZE INFO : "tab": scanned 1 of 1 pages, containing 51 live rows and 0 dead rows; 51 rows in sample, 51 estimated total rows(gaussdb pid=1)
VACUUM
复制

5.清理数据

drop table tab;
复制

通过学习openGauss的收集统计信息、打印执行计划、垃圾收集和checkpoint。了解到VACUUM回收表或B-Tree索引中已经删除的行所占据的存储空间。
检查点（CHECKPOINT）是一个事务日志中的点，所有数据文件都在该点被更新以反映日志中的信息，所有数据文件都将被刷新到磁盘。