到oracle11G 时,已经有三种方式来固定执行计划了
1.10G以前用的是存储大纲
2.10G多了一种,sql_profile
3.11G又多了sql plan management
今天先讲讲sql_profile
1.先创建一张测试表,
SQL> create table emp as select * from scott.emp;
Table created.
2.选择要固定执行计划的sql语句
这里选择 select * from emp where empno=7521;
SQL> select * from emp where empno=7521;可以看到这里没有创建索引,所以走的是全表扫描
EMPNO ENAME JOB MGR HIREDATESAL COMM DEPTNO
---------- ------------------------------ --------------------------- ---------- --------------- ---------- ---------- ----------
7521 WARD SALESMAN 7698 22-FEB-81 1250 500 30
Execution Plan
----------------------------------------------------------
Plan hash value: 3956160932
--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 87 | 2 (0)| 00:00:01 |
|* 1 | TABLE ACCESS FULL| EMP | 1 | 87 | 2 (0)| 00:00:01 |
--------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
1 - filter("EMPNO"=7521)
Note
-----
- dynamic sampling used for this statement (level=2)
3.查询出sql的sql_id
SQL> select sql_id,sql_text from v$sql where sql_text like '%emp where empno%';
SQL_ID SQL_TEXT
-------------------- --------------------------------------------------
<span style="color:#ff6666;">9th5zrrtnxkjy</span> select * from emp where empno=7521
4.根据sql_id获取 outline
---------------------------------------------------------------------------------------------------------------
SQL_ID9th5zrrtnxkjy, child number 0
-------------------------------------
select * from emp where empno=7521
Plan hash value: 3956160932
--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | | | 2 (100)| |
|* 1 | TABLE ACCESS FULL| EMP | 1 | 87 | 2 (0)| 00:00:01 |
--------------------------------------------------------------------------
Outline Data
-------------
/*+
<span style="color:#ff0000;">BEGIN_OUTLINE_DATA
IGNORE_OPTIM_EMBEDDED_HINTS
OPTIMIZER_FEATURES_ENABLE('11.2.0.4')
DB_VERSION('11.2.0.4')
ALL_ROWS
OUTLINE_LEAF(@"SEL$1")
FULL(@"SEL$1" "EMP"@"SEL$1")
END_OUTLINE_DATA</span>
*/
Predicate Information (identified by operation id):
---------------------------------------------------
1 - filter("EMPNO"=7521)
Note
-----
- dynamic sampling used for this statement (level=2)
36 rows selected.
SQL> declare
v_hints sys.sqlprof_attr;
begin
v_hints:=sys.sqlprof_attr(
'BEGIN_OUTLINE_DATA',
'IGNORE_OPTIM_EMBEDDED_HINTS',
'OPTIMIZER_FEATURES_ENABLE(''11.2.0.4'')',
'DB_VERSION(''11.2.0.4'')',
'ALL_ROWS',
'OUTL 2 3 4 5 6 7 8 9 10 INE_LEAF(@"SEL$1")',
'FULL(@"SEL$1" "EMP"@"SEL$1")',
'END_OUTLINE_DATA');
dbms_sqltune.import_sql_profile(
'select * from emp where empno=7521',
v_hints,'SQLPROFILE_001',
force_match=>true,replace=>false);
end;
/ 11 12 13 14 15 16 17 18 19
PL/SQL procedure successfully completed.
5.查看是否使用sql profile
set autot trace explain
select * from emp where empno=7521;
SQL> set autot trace explain可以看到,已经使用了sql_profile
select * from emp where empno=7521;SQL>
Execution Plan
----------------------------------------------------------
Plan hash value: 3956160932
--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 87 | 2 (0)| 00:00:01 |
|* 1 | TABLE ACCESS FULL| EMP | 1 | 87 | 2 (0)| 00:00:01 |
--------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
1 - filter("EMPNO"=7521)
Note
-----
- SQL profile "<span style="color:#ff0000;">SQLPROFILE_001</span>" used for this statement
6.在表字段empno上创建索引,再执行SQL看看执行计划是否有变化
SQL> create index idx_empno on emp(empno);从上图可以看到,执行计划仍然使用的是sql_profile中的执行计划
Index created.
SQL> select * from emp where empno=7521;
Execution Plan
----------------------------------------------------------
Plan hash value: 3956160932
--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 87 | 2 (0)| 00:00:01 |
|* 1 | TABLE ACCESS FULL| EMP | 1 | 87 | 2 (0)| 00:00:01 |
--------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
1 - filter("EMPNO"=7521)
Note
-----
- SQL profile "SQLPROFILE_001" used for this statement
7.删除sql_profile再看看执行计划是否有变化(按照最优计划来说,应该会走index的)
删除sql_profile
SQL> BEGIN
DBMS_SQLTUNE.DROP_SQL_PROFILE(name => 'SQLPROFILE_001');
END;
/ 2 3 4
PL/SQL procedure successfully completed.
SQL> select * from emp where empno=7522;
Execution Plan
----------------------------------------------------------
Plan hash value: 2426388914
-----------------------------------------------------------------------------------------
| Id | Operation | Name| Rows| Bytes | Cost (%CPU)| Time|
-----------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT || 1 | 87 | 1 (0)| 00:00:01 |
| 1 | TABLE ACCESS BY INDEX ROWID| EMP| 1 | 87 | 1 (0)| 00:00:01 |
|* 2 | INDEX RANGE SCAN | IDX_EMPNO | 1 || 1 (0)| 00:00:01 |
-----------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - access("EMPNO"=7522)
Note
-----
- dynamic sampling used for this statement (level=2)
果然,执行计划改变了,这次SQL走的是index了.