count(*)、count(val)和count(1)的解释
一、关于count的一些谣言：
1、count(*)比count(val)更慢！项目组必须用count(val)，不准用count(*)，谁用扣谁钱！
2、count(*)用不到索引，count(val)才能用到。
3、count(*)是统计出全表的记录，是吞吐量的操作，肯定用不到索引。
4、count(1)比count(*)的速度快。
二、验证count(*)和count(val)
1、首先创建一个表，使用count(*)和count(val)查询比较：
----删除echo表----
SQL> drop table echo purge;
drop table echo purge
*
第 1 行出现错误:
ORA-00942: 表或视图不存在
----创建一张echo的测试表----
SQL> create table echo as select * from dba_objects;
表已创建。
SQL> update echo set object_id = rownum;
已更新72509行。
SQL> commit;
提交完成。
SQL> set timing on
SQL> set linesize 100
SQL> set autotrace on
SQL> select count(*) from echo;
COUNT(*)
----------
72509
已用时间: 00: 00: 00.01
执行计划
----------------------------------------------------------
Plan hash value: 99109176
-------------------------------------------------------------------
| Id | Operation | Name | Rows | Cost (%CPU)| Time |
-------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 290 (1)| 00:00:04 |
| 1 | SORT AGGREGATE | | 1 | | |
| 2 | TABLE ACCESS FULL| ECHO | 80064 | 290 (1)| 00:00:04 |
-------------------------------------------------------------------
Note
-----
- dynamic sampling used for this statement (level=2)
统计信息
----------------------------------------------------------
4 recursive calls
0 db block gets
1265 consistent gets
0 physical reads
11060 redo size
425 bytes sent via SQL*Net to client
415 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
1 rows processed
SQL> select count(*) from echo;
COUNT(*)
----------
72509
已用时间: 00: 00: 00.01
执行计划
----------------------------------------------------------
Plan hash value: 99109176
-------------------------------------------------------------------
| Id | Operation | Name | Rows | Cost (%CPU)| Time |
-------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 290 (1)| 00:00:04 |
| 1 | SORT AGGREGATE | | 1 | | |
| 2 | TABLE ACCESS FULL| ECHO | 80064 | 290 (1)| 00:00:04 |
-------------------------------------------------------------------
Note
-----
- dynamic sampling used for this statement (level=2)
统计信息
----------------------------------------------------------
0 recursive calls
0 db block gets
1038 consistent gets
0 physical reads
0 redo size
425 bytes sent via SQL*Net to client
415 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
1 rows processed
SQL> select count(object_id) from echo;
COUNT(OBJECT_ID)
----------------
72509
已用时间: 00: 00: 00.01
执行计划
----------------------------------------------------------
Plan hash value: 99109176
---------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
---------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 13 | 290 (1)| 00:00:04 |
| 1 | SORT AGGREGATE | | 1 | 13 | | |
| 2 | TABLE ACCESS FULL| ECHO | 80064 | 1016K| 290 (1)| 00:00:04 |
---------------------------------------------------------------------------
Note
-----
- dynamic sampling used for this statement (level=2)
统计信息
----------------------------------------------------------
4 recursive calls
0 db block gets
1112 consistent gets
0 physical reads
0 redo size
433 bytes sent via SQL*Net to client
415 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
1 rows processed
SQL> select count(object_id) from echo;
COUNT(OBJECT_ID)
----------------
72509
已用时间: 00: 00: 00.01
执行计划
----------------------------------------------------------
Plan hash value: 99109176
---------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
---------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 13 | 290 (1)| 00:00:04 |
| 1 | SORT AGGREGATE | | 1 | 13 | | |
| 2 | TABLE ACCESS FULL| ECHO | 80064 | 1016K| 290 (1)| 00:00:04 |
---------------------------------------------------------------------------
Note
-----
- dynamic sampling used for this statement (level=2)
统计信息
----------------------------------------------------------
0 recursive calls
0 db block gets
1038 consistent gets
0 physical reads
0 redo size
433 bytes sent via SQL*Net to client
415 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
1 rows processed
从上面的执行计划可以看出count(*)和count(val)是一样快的。
2、建立索引做比较
SQL> create index idx_object_id on echo(object_id);
索引已创建。
已用时间: 00: 00: 05.69
SQL> select count(*) from echo;
COUNT(*)
----------
72509
已用时间: 00: 00: 00.05
执行计划
----------------------------------------------------------
Plan hash value: 99109176
-------------------------------------------------------------------
| Id | Operation | Name | Rows | Cost (%CPU)| Time |
-------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 290 (1)| 00:00:04 |
| 1 | SORT AGGREGATE | | 1 | | |
| 2 | TABLE ACCESS FULL| ECHO | 80064 | 290 (1)| 00:00:04 |
-------------------------------------------------------------------
Note
-----
- dynamic sampling used for this statement (level=2)
统计信息
----------------------------------------------------------
5 recursive calls
0 db block gets
1113 consistent gets
0 physical reads
0 redo size
425 bytes sent via SQL*Net to client
415 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
1 rows processed
SQL> select count(object_id) from echo;
COUNT(OBJECT_ID)
----------------
72509
已用时间: 00: 00: 00.08
执行计划
----------------------------------------------------------
Plan hash value: 1131838604
---------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
---------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 13 | 49 (0)| 00:00:01 |
| 1 | SORT AGGREGATE | | 1 | 13 | | |
| 2 | INDEX FAST FULL SCAN| IDX_OBJECT_ID | 80064 | 1016K| 49 (0)| 00:00:01 |
---------------------------------------------------------------------------------------
Note
-----
- dynamic sampling used for this statement (level=2)
统计信息
----------------------------------------------------------
4 recursive calls
0 db block gets
244 consistent gets
161 physical reads
0 redo size
433 bytes sent via SQL*Net to client
415 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
1 rows processed
哇，原来真的是用count(val)比count(*)要快啊，因为count(*)不能用到索引，而count(val)可以，真相真是如此吗？
3、将object_id设置为非空
SQL> alter table echo modify object_id not null;
表已更改。
已用时间: 00: 00: 01.41
SQL> select count(*) from echo;
COUNT(*)
----------
72509
已用时间: 00: 00: 00.00
执行计划
----------------------------------------------------------
Plan hash value: 1131838604
-------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Cost (%CPU)| Time |
-------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 49 (0)| 00:00:01 |
| 1 | SORT AGGREGATE | | 1 | | |
| 2 | INDEX FAST FULL SCAN| IDX_OBJECT_ID | 80064 | 49 (0)| 00:00:01 |
-------------------------------------------------------------------------------
Note
-----
- dynamic sampling used for this statement (level=2)
统计信息
----------------------------------------------------------
0 recursive calls
0 db block gets
169 consistent gets
0 physical reads
0 redo size
425 bytes sent via SQL*Net to client
415 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
1 rows processed
SQL> select count(object_id) from echo;
COUNT(OBJECT_ID)
----------------
72509
已用时间: 00: 00: 00.01
执行计划
----------------------------------------------------------
Plan hash value: 1131838604
-------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Cost (%CPU)| Time |
-------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 49 (0)| 00:00:01 |
| 1 | SORT AGGREGATE | | 1 | | |
| 2 | INDEX FAST FULL SCAN| IDX_OBJECT_ID | 80064 | 49 (0)| 00:00:01 |
-------------------------------------------------------------------------------
Note
-----
- dynamic sampling used for this statement (level=2)
统计信息
----------------------------------------------------------
0 recursive calls
0 db block gets
169 consistent gets
0 physical reads
0 redo size
433 bytes sent via SQL*Net to client
415 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
1 rows processed
--看来count(val)和count(*)其实一样快，如果索引列是非空的，count(*)可用到索引，此时一样快！真相真是如此吗？
其实两者根本没有可比性，性能比较首先考虑写法等价，这两个语句根本就不等价。
结论：
其实优化器里的算法是这么玩的，列的偏移量决定性能，列越靠后，访问的开销越大。由于count(*)的算法与列偏移量无关，所以count(*)最快，count(最后列val)最慢。
3、用实验验证上面的结论：
SQL> set serveroutput on
SQL> set echo on
SQL> drop table t purge;
表已删除。
----构造出有25个字段的表T----
DECLARE
l_sql VARCHAR2(32767);
BEGIN
l_sql := 'CREATE TABLE t (';
FOR i IN 1..25
LOOP
l_sql := l_sql || 'n' || i || ' NUMBER,';
END LOOP;
l_sql := l_sql || 'pad VARCHAR2(1000)) PCTFREE 10';
EXECUTE IMMEDIATE l_sql;
END;
12 /
PL/SQL 过程已成功完成。
----将记录还有这个表T中填充----
DECLARE
l_sql VARCHAR2(32767);
BEGIN
l_sql := 'INSERT INTO t SELECT ';
FOR i IN 1..25
LOOP
l_sql := l_sql || '0,';
END LOOP;
l_sql := l_sql || 'NULL FROM dual CONNECT BY level <= 10000';
EXECUTE IMMEDIATE l_sql;
COMMIT;
END;
13 /
PL/SQL 过程已成功完成。
execute dbms_stats.gather_table_stats(ownname=>user, tabname=>'t')
SELECT num_rows, blocks FROM user_tables WHERE table_name = 'T';
PL/SQL 过程已成功完成。
SQL>
NUM_ROWS BLOCKS
---------- ----------
10000 80
----以下动作观察执行速度，比较发现count(*)最快，count(最大列)最慢----
SQL> DECLARE
l_dummy PLS_INTEGER;
l_start PLS_INTEGER;
l_stop PLS_INTEGER;
l_sql VARCHAR2(100);
BEGIN
l_start := dbms_utility.get_time;
FOR j IN 1..1000
LOOP
EXECUTE IMMEDIATE 'SELECT count(*) FROM t' INTO l_dummy;
END LOOP;
l_stop := dbms_utility.get_time;
dbms_output.put_line((l_stop-l_start)/100);
14
FOR i IN 1..25
LOOP
l_sql := 'SELECT count(n' || i || ') FROM t';
l_start := dbms_utility.get_time;
FOR j IN 1..1000
LOOP
EXECUTE IMMEDIATE l_sql INTO l_dummy;
END LOOP;
l_stop := dbms_utility.get_time;
dbms_output.put_line((l_stop-l_start)/100);
END LOOP;
END;
27 /
.18
.33
.39
.38
.42
.4
.45
.49
.48
.46
.48
.48
.55
.51
.56
.57
.61
.62
.75
.67
.68
.7
.73
.78
.77
.81
PL/SQL 过程已成功完成。
三、验证count(*)和count(1)
沿用TOM大师的解释：
