【SPM】Oracle如何固定执行计划

简介: 【SPM】Oracle如何固定执行计划   1.1  BLOG文档结构图   1.2  前言部分   1.


SPM】Oracle如何固定执行计划

 

1.1  BLOG文档结构图

wps73.tmp 

1.2  前言部分

 

1.2.1  导读和注意事项

各位技术爱好者,看完本文后,你可以掌握如下的技能,也可以学到一些其它你所不知道的知识,~O(∩_∩)O~

固定执行计划的常用方法:outlineSQL ProfileSPM(重点)

coe_xfr_sql_profile.sql脚本的使用

 

  Tips:

       若文章代码格式有错乱,推荐使用QQ、搜狗或360浏览器,也可以下载pdf格式的文档来查看,pdf文档下载地址:http://yunpan.cn/cdEQedhCs2kFz (提取码:ed9b 

       ② 本篇BLOG中命令的输出部分需要特别关注的地方我都用灰色背景和粉红色字体来表示,比如下边的例子中,thread 1的最大归档日志号为33thread 2的最大归档日志号为43是需要特别关注的地方;而命令一般使用黄色背景和红色字体注;对代码或代码输出部分的注释一般采用蓝色字体表示

 

  List of Archived Logs in backup set 11

  Thrd Seq     Low SCN    Low Time            Next SCN   Next Time

  ---- ------- ---------- ------------------- ---------- ---------

  1    32      1621589    2015-05-29 11:09:52 1625242    2015-05-29 11:15:48

  1    33      1625242    2015-05-29 11:15:48 1625293    2015-05-29 11:15:58

  2    42      1613951    2015-05-29 10:41:18 1625245    2015-05-29 11:15:49

  2    43      1625245    2015-05-29 11:15:49 1625253    2015-05-29 11:15:53

 

 

 

 

[ZHLHRDB1:root]:/>lsvg -o

T_XDESK_APP1_vg

rootvg

[ZHLHRDB1:root]:/>

00:27:22 SQL> alter tablespace idxtbs read write;

 

 

====》2097152*512/1024/1024/1024=1G 

 

 

 

 

本文如有错误或不完善的地方请大家多多指正,ITPUB留言或QQ皆可,您的批评指正是我写作的最大动力。

 

 

1.2.2  相关参考文章链接

11.2.0.2SPM的一个bug http://blog.itpub.net/26736162/viewspace-1248506/

10g/11g中如何查看SQL Profiles信息:http://blog.itpub.net/26736162/viewspace-2106743/

OUTLINE】使用Oracle Outline技术暂时锁定SQL的执行计划:http://blog.itpub.net/26736162/viewspace-2102180/

 

 

 

 

1.2.3  本文简介

本文介绍了oracle在固定执行计划的过程中常使用的3种方法,outline,SQL ProfileSPM,其中SQL ProfileSPM是重点需要掌握的内容。

 

 

---------------------------------------------------------------------------------------------------------------------

 

 

第二章 固定执行计划的三种方法介绍

2.1  outline

2.1.1  outline基础知识

在实际项目中,通常在开发环境下一些SQL 执行没有任何问题,而到了生产环境或生产环境的数据量发生较大的变量时,其SQL 的执行效率会异常的慢。此时如果更改SQL ,则可能需要重新修改源程序以及重新编译程序。如果觉得修改源程序的成本比较大,则可以使用OUTLINE在不改变原应用程序的情况下更改特定SQL 的执行计划。

OUTLINE的原理是将调好的SQL 的执行计划(一系列的HINT)存贮起来,然后该执行计划所对应的SQL 用目前系统那个效率低下的SQL 来替代之。从而使得系统每次执行该SQL 时,都会使用已存贮的执行计划来执行。因此可以在不改变已有系统SQL 的情况下达到改变其执行计划的目的。

OUTLINE方式也是通过存贮HINT的方式来达到执行计划的稳定与改变。

当发现低效SQL之后,可以使用hint优化他,对于SQL代码可以修改的情况,直接修改SQL代码加上hint即可,但是对于SQL代码不可修改的情况,Oracle提供了outLine功能来为SQL修改hint,以致执行计划变更!

 

?OutLine机制:

Outline保存了SQLhintoutline的表中。当执行SQL时,Oracle会使用outline中的hint来为SQL生成执行计划。

?使用  OutLine的步骤:

1)生成新SQL和老SQL2Outline

2)交换两个SQL的提示信息

3ON LOGON触发器设定sessionCATEGORY(自定义类别)

SQL命令行为:SQL> alter session set use_stored_outlines=special;

2.1.2  ouline使用演示

测试过程如下:

SYS@test> create user lhr identified by lhr;

 

User created.

 

SYS@test> grant dba to lhr;

 

Grant succeeded.

 

SYS@test> grant create any outline,alter any outline,DROP ANY OUTLINE to lhr;

 

Grant succeeded.

 

SYS@test> grant all on OL$HINTS to lhr;

 

Grant succeeded.

 

SYS@test> conn lhr/lhr

Connected.

LHR@test> select * from v$version;

 

BANNER

--------------------------------------------------------------------------------

Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production

PL/SQL Release 11.2.0.4.0 - Production

CORE    11.2.0.4.0      Production

TNS for IBM/AIX RISC System/6000: Version 11.2.0.4.0 - Production

NLSRTL Version 11.2.0.4.0 - Production

 

LHR@test> create table TB_LHR_20160518 as select * from dba_tables;

 

Table created.

 

LHR@test> create index idx_TB_LHR_20160518 on TB_LHR_20160518(TABLE_NAME);

 

Index created.

 

LHR@test> SET AUTOTRACE ON;

LHR@test> select owner from TB_LHR_20160518 where table_name='TB_LHR_20160518';

 

no rows selected

 

 

Execution Plan

----------------------------------------------------------

Plan hash value: 2186742855

 

---------------------------------------------------------------------------------------------------

| Id  | Operation                   | Name                | Rows  | Bytes | Cost (%CPU)| Time     |

---------------------------------------------------------------------------------------------------

|   0 | SELECT STATEMENT            |                     |     1 |    34 |     1   (0)| 00:00:01 |

|   1 |  TABLE ACCESS BY INDEX ROWID| TB_LHR_20160518     |     1 |    34 |     1   (0)| 00:00:01 |

|*  2 |   INDEX RANGE SCAN          | IDX_TB_LHR_20160518 |     1 |       |     1   (0)| 00:00:01 |

---------------------------------------------------------------------------------------------------

 

Predicate Information (identified by operation id):

---------------------------------------------------

 

   2 - access("TABLE_NAME"='TB_LHR_20160518')

 

Note

-----

   - dynamic sampling used for this statement (level=2)

 

 

Statistics

----------------------------------------------------------

         11  recursive calls

          0  db block gets

         72  consistent gets

          8  physical reads

          0  redo size

        333  bytes sent via SQL*Net to client

        508  bytes received via SQL*Net from client

          1  SQL*Net roundtrips to/from client

          0  sorts (memory)

          0  sorts (disk)

          0  rows processed

 

LHR@test> select /*+full(TB_LHR_20160518)*/ owner from TB_LHR_20160518 where table_name='TB_LHR_20160518';

 

no rows selected

 

 

Execution Plan

----------------------------------------------------------

Plan hash value: 1750418716

 

-------------------------------------------------------------------------------------

| Id  | Operation         | Name            | Rows  | Bytes | Cost (%CPU)| Time     |

-------------------------------------------------------------------------------------

|   0 | SELECT STATEMENT  |                 |     1 |    34 |    31   (0)| 00:00:01 |

|*  1 |  TABLE ACCESS FULL| TB_LHR_20160518 |     1 |    34 |    31   (0)| 00:00:01 |

-------------------------------------------------------------------------------------

 

Predicate Information (identified by operation id):

---------------------------------------------------

 

   1 - filter("TABLE_NAME"='TB_LHR_20160518')

 

Note

-----

   - dynamic sampling used for this statement (level=2)

 

 

Statistics

----------------------------------------------------------

          7  recursive calls

          0  db block gets

        170  consistent gets

          0  physical reads

          0  redo size

        333  bytes sent via SQL*Net to client

        508  bytes received via SQL*Net from client

          1  SQL*Net roundtrips to/from client

          0  sorts (memory)

          0  sorts (disk)

          0  rows processed

 

LHR@test> set autotrace off;

LHR@test> create or replace outline TB_LHR_20160518_1 on select owner from TB_LHR_20160518 where table_name='TB_LHR_20160518';

 

Outline created.

 

LHR@test> create or replace outline TB_LHR_20160518_2 on select /*+full(TB_LHR_20160518)*/ owner from TB_LHR_20160518 where table_name='TB_LHR_20160518';

 

Outline created.

 

LHR@test> select name,USED,sql_text from dba_outlines where name like '%TB_LHR_20160518%';

 

NAME                           USED   SQL_TEXT

------------------------------ ------ --------------------------------------------------------------------------------

TB_LHR_20160518_1              UNUSED select owner from TB_LHR_20160518 where table_name='TB_LHR_20160518'

TB_LHR_20160518_2              UNUSED select /*+full(TB_LHR_20160518)*/ owner from TB_LHR_20160518 where table_name='T

 

LHR@test> select name,HINT from dba_outline_hints where JOIN_POS=1 and name like '%TB_LHR_20160518%';

 

NAME                           HINT

------------------------------ --------------------------------------------------------------------------------

TB_LHR_20160518_1              INDEX_RS_ASC(@"SEL$1" "TB_LHR_20160518"@"SEL$1" ("TB_LHR_20160518"."TABLE_NAME")

TB_LHR_20160518_2              FULL(@"SEL$1" "TB_LHR_20160518"@"SEL$1")

 

LHR@test> UPDATE OUTLN.OL$ SET OL_NAME=DECODE(OL_NAME,'TB_LHR_20160518_2','TB_LHR_20160518_1','TB_LHR_20160518_1','TB_LHR_20160518_2') WHERE OL_NAME  IN ('TB_LHR_20160518_1','TB_LHR_20160518_2');

 

2 rows updated.

 

LHR@test> commit;

 

Commit complete.

 

LHR@test> select name,USED,sql_text from dba_outlines where name like '%TB_LHR_20160518%';

 

NAME                           USED   SQL_TEXT

------------------------------ ------ --------------------------------------------------------------------------------

TB_LHR_20160518_1              UNUSED select /*+full(TB_LHR_20160518)*/ owner from TB_LHR_20160518 where table_name='T

TB_LHR_20160518_2              UNUSED select owner from TB_LHR_20160518 where table_name='TB_LHR_20160518'

 

LHR@test> select name,HINT from dba_outline_hints where JOIN_POS=1 and name like '%TB_LHR_20160518%';

 

NAME                           HINT

------------------------------ --------------------------------------------------------------------------------

TB_LHR_20160518_1              INDEX_RS_ASC(@"SEL$1" "TB_LHR_20160518"@"SEL$1" ("TB_LHR_20160518"."TABLE_NAME")

TB_LHR_20160518_2              FULL(@"SEL$1" "TB_LHR_20160518"@"SEL$1")

 

LHR@test> SET AUTOTRACE ON;

LHR@test> alter system set use_stored_outlines=true;

 

System altered.

 

LHR@test> select owner from TB_LHR_20160518 where table_name='TB_LHR_20160518';

 

no rows selected

 

 

Execution Plan

----------------------------------------------------------

Plan hash value: 1750418716

 

-------------------------------------------------------------------------------------

| Id  | Operation         | Name            | Rows  | Bytes | Cost (%CPU)| Time     |

-------------------------------------------------------------------------------------

|   0 | SELECT STATEMENT  |                 |    89 |  3026 |    31   (0)| 00:00:01 |

|*  1 |  TABLE ACCESS FULL| TB_LHR_20160518 |    89 |  3026 |    31   (0)| 00:00:01 |

-------------------------------------------------------------------------------------

 

Predicate Information (identified by operation id):

---------------------------------------------------

 

   1 - filter("TABLE_NAME"='TB_LHR_20160518')

 

Note

-----

   - outline "TB_LHR_20160518_2" used for this statement

 

 

Statistics

----------------------------------------------------------

         34  recursive calls

        147  db block gets

        125  consistent gets

          0  physical reads

        624  redo size

        333  bytes sent via SQL*Net to client

        508  bytes received via SQL*Net from client

          1  SQL*Net roundtrips to/from client

          2  sorts (memory)

          0  sorts (disk)

          0  rows processed

 

LHR@test> select /*+full(TB_LHR_20160518)*/ owner from TB_LHR_20160518 where table_name='TB_LHR_20160518';

 

no rows selected

 

 

Execution Plan

----------------------------------------------------------

Plan hash value: 2186742855

 

---------------------------------------------------------------------------------------------------

| Id  | Operation                   | Name                | Rows  | Bytes | Cost (%CPU)| Time     |

---------------------------------------------------------------------------------------------------

|   0 | SELECT STATEMENT            |                     |    89 |  3026 |     6   (0)| 00:00:01 |

|   1 |  TABLE ACCESS BY INDEX ROWID| TB_LHR_20160518     |    89 |  3026 |     6   (0)| 00:00:01 |

|*  2 |   INDEX RANGE SCAN          | IDX_TB_LHR_20160518 |    36 |       |     1   (0)| 00:00:01 |

---------------------------------------------------------------------------------------------------

 

Predicate Information (identified by operation id):

---------------------------------------------------

 

   2 - access("TABLE_NAME"='TB_LHR_20160518')

 

Note

-----

   - outline "TB_LHR_20160518_1" used for this statement

 

 

Statistics

----------------------------------------------------------

         34  recursive calls

        147  db block gets

         24  consistent gets

          0  physical reads

        584  redo size

        333  bytes sent via SQL*Net to client

        508  bytes received via SQL*Net from client

          1  SQL*Net roundtrips to/from client

          2  sorts (memory)

          0  sorts (disk)

          0  rows processed

 

LHR@test>

 

 

 

2.2  SQL Profile

2.2.1  SQL Profile基础知识

在oracle 11g的后续版本中,use_stored_outlines这个参数已经不存在了。意味着我们不能像以前的版本中使用create outline的方式来为一个sql创建hint,然后使用store outline来固定执行计划这种方式了.

SQL Profile就是为某一SQL语句提供除了系统统计信息、对象(表和索引等)统计信息之外的其他信息,比如运行环境、额外的更准确的统计信息,以帮助优化器为SQL语句选择更适合的执行计划。SQL Profiles可以说是Outlines的进化。Outlines能够实现的功能SQL Profiles也完全能够实现,而SQL Profiles具有Outlines不具备的优化,最重要的有二点: 

①  SQL Profiles更容易生成、更改和控制。 

②  SQL Profiles在对SQL语句的支持上做得更好,也就是适用范围更广。 

 

使用SQL Profiles两个目的: 

(一)  锁定或者说是稳定执行计划。 

(二)  在不能修改应用中的SQL的情况下使SQL语句按指定的执行计划运行。 

 10g之前有outlines,10g之后sql profile作为新特性之一出现。如果针对非绑定变量的sql,outlines则力不从心。sql profile最大的优点是在不修改sql语句和会话执行环境的情况下去优化sql的执行效率,适合无法在应用程序中修改sql.

SQL Profile以下类型语句有效:

     SELECT语句;

     UPDATE语句;

     INSERT语句(仅当使用SELECT子句时有效);

     DELETE语句;

     CREATE语句(仅当使用SELECT子句时有效);

     MERGE语句(仅当作UPDATEINSERT操作时有效)。

另外,使用SQL Profile还必须有CREATE ANY SQL PROFILEDROP ANY SQL PROFILEALTER ANY SQL PROFILE等系统权限。

 

2.2.2  SQL Profile使用演示

2种生成SQL Profile的方法,手动和采用STA来生成。

2.2.2.1  SQL Profile使用示例--手工创建SQL Profile

创建测试表,根据DBA_OBJECTS创建,OBJECT_ID上有索引

LHR@dlhr> select * from v$version;

 

BANNER

--------------------------------------------------------------------------------

Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production

PL/SQL Release 11.2.0.4.0 - Production

CORE    11.2.0.4.0      Production

TNS for IBM/AIX RISC System/6000: Version 11.2.0.4.0 - Production

NLSRTL Version 11.2.0.4.0 - Production

 

LHR@dlhr> Create table TB_LHR_20160525 as select * from dba_objects;

 

Table created.

 

LHR@dlhr> create index IND_TB_LHR_ID on TB_LHR_20160525(object_id);

 

Index created.

 

 

查看SQL默认执行计划,走了索引,通过指定outline可以获取到系统为我们生成的hint

LHR@dlhr> explain plan for select * from TB_LHR_20160525 where object_id= :a;

 

Explained.

 

LHR@dlhr> select * from table(dbms_xplan.display(null,null,'outline'));

 

PLAN_TABLE_OUTPUT

------------------------------------------------------------------------------------------------------------------------------------

Plan hash value: 4254050152

 

-----------------------------------------------------------------------------------------------

| Id  | Operation                   | Name            | Rows  | Bytes | Cost (%CPU)| Time     |

-----------------------------------------------------------------------------------------------

|   0 | SELECT STATEMENT            |                 |   886 |   179K|     7   (0)| 00:00:01 |

|   1 |  TABLE ACCESS BY INDEX ROWID| TB_LHR_20160525 |   886 |   179K|     7   (0)| 00:00:01 |

|*  2 |   INDEX RANGE SCAN          | IND_TB_LHR_ID   |   354 |       |     1   (0)| 00:00:01 |

-----------------------------------------------------------------------------------------------

 

Outline Data

-------------

 

  /*+

      BEGIN_OUTLINE_DATA

      INDEX_RS_ASC(@"SEL$1" "TB_LHR_20160525"@"SEL$1" ("TB_LHR_20160525"."OBJECT_ID"))

      OUTLINE_LEAF(@"SEL$1")

      ALL_ROWS

      DB_VERSION('11.2.0.4')

      OPTIMIZER_FEATURES_ENABLE('11.2.0.4')

      IGNORE_OPTIM_EMBEDDED_HINTS

      END_OUTLINE_DATA

  */

 

Predicate Information (identified by operation id):

---------------------------------------------------

 

   2 - access("OBJECT_ID"=TO_NUMBER(:A))

 

Note

-----

   - dynamic sampling used for this statement (level=2)

 

32 rows selected.

 

 

如果我们想让它走全表扫描,首先获取全表扫描HINT

 

LHR@dlhr> explain plan for select /*+ full(TB_LHR_20160525) */* from TB_LHR_20160525 where object_id= :a;

 

Explained.

 

LHR@dlhr> select * from table(dbms_xplan.display(null,null,'outline'));

 

PLAN_TABLE_OUTPUT

------------------------------------------------------------------------------------------------------------------------------------

Plan hash value: 345881005

 

-------------------------------------------------------------------------------------

| Id  | Operation         | Name            | Rows  | Bytes | Cost (%CPU)| Time     |

-------------------------------------------------------------------------------------

|   0 | SELECT STATEMENT  |                 |   886 |   179K|   352   (2)| 00:00:05 |

|*  1 |  TABLE ACCESS FULL| TB_LHR_20160525 |   886 |   179K|   352   (2)| 00:00:05 |

-------------------------------------------------------------------------------------

 

Outline Data

-------------

 

  /*+

      BEGIN_OUTLINE_DATA

     FULL(@"SEL$1" "TB_LHR_20160525"@"SEL$1")

      OUTLINE_LEAF(@"SEL$1")

      ALL_ROWS

      DB_VERSION('11.2.0.4')

      OPTIMIZER_FEATURES_ENABLE('11.2.0.4')

      IGNORE_OPTIM_EMBEDDED_HINTS

      END_OUTLINE_DATA

  */

 

Predicate Information (identified by operation id):

---------------------------------------------------

 

   1 - filter("OBJECT_ID"=TO_NUMBER(:A))

 

Note

-----

   - dynamic sampling used for this statement (level=2)

 

31 rows selected.

 

 

可以看到全表扫描的hint已经为我们生成了,我们选取必要的hintOK了,其他的可以不要,使用sql profile

 

LHR@dlhr> declare

  2        v_hints sys.sqlprof_attr;

  3  begin

  4        v_hints := sys.sqlprof_attr('FULL(@"SEL$1" "TB_LHR_20160525"@"SEL$1")'); ----------从上面Outline Data部分获取到的HINT

  5        dbms_sqltune.import_sql_profile('select * from TB_LHR_20160525 where object_id= :a', ----------SQL语句部分

  6                                 v_hints,

  7                                 'TB_LHR_20160525', --------PROFILE 的名字

  8                                  force_match => true);

  9  end;

10  /

 

PL/SQL procedure successfully completed.

 

 

查看是否生效,已经生效了

 

LHR@dlhr> explain plan for select * from TB_LHR_20160525 where object_id= :a;

 

Explained.

 

LHR@dlhr> select * from table(dbms_xplan.display);

 

PLAN_TABLE_OUTPUT

------------------------------------------------------------------------------------------------------------------------------------

Plan hash value: 345881005

 

-------------------------------------------------------------------------------------

| Id  | Operation         | Name            | Rows  | Bytes | Cost (%CPU)| Time     |

-------------------------------------------------------------------------------------

|   0 | SELECT STATEMENT  |                 |   886 |   179K|   352   (2)| 00:00:05 |

|*  1 |  TABLE ACCESS FULL| TB_LHR_20160525 |   886 |   179K|   352   (2)| 00:00:05 |

-------------------------------------------------------------------------------------

 

Predicate Information (identified by operation id):

---------------------------------------------------

 

   1 - filter("OBJECT_ID"=TO_NUMBER(:A))

 

Note

-----

   - dynamic sampling used for this statement (level=2)

   - SQL profile "TB_LHR_20160525" used for this statement

 

18 rows selected.

 

LHR@dlhr>  SELECT b.name,d.sql_text,  extractvalue(value(h),'.') as hints

  2     FROM dba_sql_profiles d,SYS.SQLOBJ$DATA A,

  3          SYS.SQLOBJ$ B,

  4          TABLE(XMLSEQUENCE(EXTRACT(XMLTYPE(A.COMP_DATA),

  5                                    '/outline_data/hint'))) h

  6    where a.signature = b.signature

  7      and a.category = b.category

  8      and a.obj_type = b.obj_type

  9      and a.plan_id = b.plan_id

10             and a.signature=d.signature

11             and D.name = 'TB_LHR_20160525';

 

NAME                           SQL_TEXT                                                                         HINTS

------------------------------ -------------------------------------------------------------------------------- -------------------------------------------------------

TB_LHR_20160525                select * from TB_LHR_20160525 where object_id= :a                                FULL(@"SEL$1" "TB_LHR_20160525"@"SEL$1")

 

LHR@dlhr>

 

 

 

 

一、 使用coe_xfr_sql_profile.sql 脚本生成sqlprof_attr数据

最麻烦的sqlprof_attr('FULL(t1@SEL$1)')是这里的格式如何写.mos上的文章note 215187.1中的sqlt.zip的目录utl中提供了脚本coe_xfr_sql_profile.sql可以生成这些信息.


1.建立测试表和数据

SYS@dlhr> select * from v$version;

 

BANNER

--------------------------------------------------------------------------------

Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production

PL/SQL Release 11.2.0.4.0 - Production

CORE    11.2.0.4.0      Production

TNS for IBM/AIX RISC System/6000: Version 11.2.0.4.0 - Production

NLSRTL Version 11.2.0.4.0 - Production

 

 

LHR@dlhr> create table scott.test as select * from dba_objects;

 

Table created.

 

LHR@dlhr> create index scott.idx_test_01 on scott.test(object_id);

 

Index created.

 

LHR@dlhr> exec dbms_stats.gather_table_stats('scott','test',cascade=>true);

 

PL/SQL procedure successfully completed.

 

LHR@dlhr> update scott.test set object_id=10 where object_id>10;

 

 

LHR@dlhr> commit;

Commit complete.

 

 

 

LHR@dlhr> select OBJECT_ID ,count(1) from scott.test group by OBJECT_ID;

 

OBJECT_ID   COUNT(1)

---------- ----------

         6          1

         7          1

         5          1

         8          1

         3          1

         2          1

        10      87076

         4          1

         9          1

 

9 rows selected.

 

 


2.执行查询语句
--执行原有的查询语句,查看执行计划发现走索引,实际上这时表中大部分行的object_id都已经被更新为10,所以走索引是不合理的.

LHR@dlhr>

LHR@dlhr> set autot traceonly explain stat

LHR@dlhr>

LHR@dlhr> select * from scott.test where object_id=10;

 

87076 rows selected.

 

 

Execution Plan

----------------------------------------------------------

Plan hash value: 3384190782

 

-------------------------------------------------------------------------------------------

| Id  | Operation                   | Name        | Rows  | Bytes | Cost (%CPU)| Time     |

-------------------------------------------------------------------------------------------

|   0 | SELECT STATEMENT            |             |     1 |    98 |     2   (0)| 00:00:01 |

|   1 |  TABLE ACCESS BY INDEX ROWID| TEST        |     1 |    98 |     2   (0)| 00:00:01 |

|*  2 |   INDEX RANGE SCAN          | IDX_TEST_01 |     1 |       |     1   (0)| 00:00:01 |

-------------------------------------------------------------------------------------------

 

Predicate Information (identified by operation id):

---------------------------------------------------

 

   2 - access("OBJECT_ID"=10)

 

 

Statistics

----------------------------------------------------------

          0  recursive calls

          0  db block gets

      13060  consistent gets

          0  physical reads

          0  redo size

    9855485  bytes sent via SQL*Net to client

      64375  bytes received via SQL*Net from client

       5807  SQL*Net roundtrips to/from client

          0  sorts (memory)

          0  sorts (disk)

      87076  rows processed

 

LHR@dlhr> select /*+ full(test)*/* from scott.test where object_id=10;

 

87076 rows selected.

 

 

Execution Plan

----------------------------------------------------------

Plan hash value: 217508114

 

--------------------------------------------------------------------------

| Id  | Operation         | Name | Rows  | Bytes | Cost (%CPU)| Time     |

--------------------------------------------------------------------------

|   0 | SELECT STATEMENT  |      |     1 |    98 |   351   (2)| 00:00:05 |

|*  1 |  TABLE ACCESS FULL| TEST |     1 |    98 |   351   (2)| 00:00:05 |

--------------------------------------------------------------------------

 

Predicate Information (identified by operation id):

---------------------------------------------------

 

   1 - filter("OBJECT_ID"=10)

 

 

Statistics

----------------------------------------------------------

          1  recursive calls

          0  db block gets

       6973  consistent gets

          0  physical reads

          0  redo size

    4159482  bytes sent via SQL*Net to client

      64375  bytes received via SQL*Net from client

       5807  SQL*Net roundtrips to/from client

          0  sorts (memory)

          0  sorts (disk)

      87076  rows processed

 

 


3.查询上面两个语句的sql_id,plan_hash_value

 

LHR@dlhr> set autot off

LHR@dlhr>

LHR@dlhr> col sql_text format a100

LHR@dlhr> select sql_text,sql_id,plan_hash_value from v$sql

  2  where sql_text like 'select * from scott.test where object_id=10%';

 

SQL_TEXT                                                                                             SQL_ID        PLAN_HASH_VALUE

---------------------------------------------------------------------------------------------------- ------------- ---------------

select * from scott.test where object_id=10                                                          cpk9jsg2qt52r      3384190782

 

LHR@dlhr> select sql_text,sql_id,plan_hash_value from v$sql

  2  where sql_text like 'select /*+ full(test)*/* from scott.test where object_id=10%';

 

SQL_TEXT                                                                                             SQL_ID        PLAN_HASH_VALUE

---------------------------------------------------------------------------------------------------- ------------- ---------------

select /*+ full(test)*/* from scott.test where object_id=10                                          06c2mucgn6t5g       217508114

 

 

 


4.coe_xfr_sql_profile.sql放在$ORACLE_HOME/rdbms/admin下,或者放在/tmp下都可以。

wps83.tmp


5.对上面的两个sql产生outline datasql.

[ZHLHRSPMDB2:oracle]:/oracle>cd /tmp

[ZHLHRSPMDB2:oracle]:/tmp>

[ZHLHRSPMDB2:oracle]:/tmp>

[ZHLHRSPMDB2:oracle]:/tmp>

[ZHLHRSPMDB2:oracle]:/tmp>

[ZHLHRSPMDB2:oracle]:/tmp>sqlplus / as sysdba

 

SQL*Plus: Release 11.2.0.4.0 Production on Thu May 26 09:15:14 2016

 

Copyright (c) 1982, 2013, Oracle.  All rights reserved.

 

 

Connected to:

Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production

With the Partitioning, Real Application Clusters, OLAP, Data Mining

and Real Application Testing options

 

SYS@dlhr> @$ORACLE_HOME/rdbms/admin/coe_xfr_sql_profile.sql cpk9jsg2qt52r 3384190782

 

Parameter 1:

SQL_ID (required)

 

 

 

PLAN_HASH_VALUE AVG_ET_SECS

--------------- -----------

     3384190782        .046

 

Parameter 2:

PLAN_HASH_VALUE (required)

 

 

Values passed to coe_xfr_sql_profile:

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

SQL_ID         : "cpk9jsg2qt52r"

PLAN_HASH_VALUE: "3384190782"

 

SQL>BEGIN

  2    IF :sql_text IS NULL THEN

  3      RAISE_APPLICATION_ERROR(-20100, 'SQL_TEXT for SQL_ID &&sql_id. was not found in memory (gv$sqltext_with_newlines) or AWR (dba_hist_sqltext).');

  4    END IF;

  5  END;

  6  /

SQL>SET TERM OFF;

SQL>BEGIN

  2    IF :other_xml IS NULL THEN

  3      RAISE_APPLICATION_ERROR(-20101, 'PLAN for SQL_ID &&sql_id. and PHV &&plan_hash_value. was not found in memory (gv$sql_plan) or AWR (dba_hist_sql_plan).');

  4    END IF;

  5  END;

  6  /

SQL>SET TERM OFF;

 

Execute coe_xfr_sql_profile_cpk9jsg2qt52r_3384190782.sql

on TARGET system in order to create a custom SQL Profile

with plan 3384190782 linked to adjusted sql_text.

 

 

COE_XFR_SQL_PROFILE completed.

 

SQL>@$ORACLE_HOME/rdbms/admin/coe_xfr_sql_profile.sql 06c2mucgn6t5g 217508114

 

Parameter 1:

SQL_ID (required)

 

 

 

PLAN_HASH_VALUE AVG_ET_SECS

--------------- -----------

      217508114        .113

 

Parameter 2:

PLAN_HASH_VALUE (required)

 

 

Values passed to coe_xfr_sql_profile:

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

SQL_ID         : "06c2mucgn6t5g"

PLAN_HASH_VALUE: "217508114"

 

SQL>BEGIN

  2    IF :sql_text IS NULL THEN

  3      RAISE_APPLICATION_ERROR(-20100, 'SQL_TEXT for SQL_ID &&sql_id. was not found in memory (gv$sqltext_with_newlines) or AWR (dba_hist_sqltext).');

  4    END IF;

  5  END;

  6  /

SQL>SET TERM OFF;

SQL>BEGIN

  2    IF :other_xml IS NULL THEN

  3      RAISE_APPLICATION_ERROR(-20101, 'PLAN for SQL_ID &&sql_id. and PHV &&plan_hash_value. was not found in memory (gv$sql_plan) or AWR (dba_hist_sql_plan).');

  4    END IF;

  5  END;

  6  /

SQL>SET TERM OFF;

 

Execute coe_xfr_sql_profile_06c2mucgn6t5g_217508114.sql

on TARGET system in order to create a custom SQL Profile

with plan 217508114 linked to adjusted sql_text.

 

 

COE_XFR_SQL_PROFILE completed.

 


6.替换文件coe_xfr_sql_profile_cpk9jsg2qt52r_3384190782.sql中的SYS.SQLPROF_ATTR部分,把它更改为
coe_xfr_sql_profile_06c2mucgn6t5g_217508114.sql中产生的SYS.SQLPROF_ATTR部分,其中:


coe_xfr_sql_profile_cpk9jsg2qt52r_3384190782.sqlSYS.SQLPROF_ATTR

h := SYS.SQLPROF_ATTR(

q'[BEGIN_OUTLINE_DATA]',

q'[IGNORE_OPTIM_EMBEDDED_HINTS]',

q'[OPTIMIZER_FEATURES_ENABLE('11.2.0.4')]',

q'[DB_VERSION('11.2.0.4')]',

q'[ALL_ROWS]',

q'[OUTLINE_LEAF(@"SEL$1")]',

q'[INDEX_RS_ASC(@"SEL$1" "TEST"@"SEL$1" ("TEST"."OBJECT_ID"))]',

q'[END_OUTLINE_DATA]');

 

----coe_xfr_sql_profile_06c2mucgn6t5g_217508114.sqlSYS.SQLPROF_ATTR

h := SYS.SQLPROF_ATTR(

q'[BEGIN_OUTLINE_DATA]',

q'[IGNORE_OPTIM_EMBEDDED_HINTS]',

q'[OPTIMIZER_FEATURES_ENABLE('11.2.0.4')]',

q'[DB_VERSION('11.2.0.4')]',

q'[ALL_ROWS]',

q'[OUTLINE_LEAF(@"SEL$1")]',

q'[FULL(@"SEL$1" "TEST"@"SEL$1")]',

q'[END_OUTLINE_DATA]');

 

生成的文件在当前目录:

wps94.tmp

 

wps95.tmp

 

7.执行替换过SYS.SQLPROF_ATTRSQL,coe_xfr_sql_profile_cpk9jsg2qt52r_3384190782.sql

SQL> @/tmp/coe_xfr_sql_profile_cpk9jsg2qt52r_3384190782.sql

SQL>@coe_xfr_sql_profile_cpk9jsg2qt52r_3384190782.sql

SQL>REM

SQL>REM $Header: 215187.1 coe_xfr_sql_profile_cpk9jsg2qt52r_3384190782.sql 11.4.4.4 2016/05/26 carlos.sierra $

SQL>REM

SQL>REM Copyright (c) 2000-2012, Oracle Corporation. All rights reserved.

SQL>REM

SQL>REM AUTHOR

SQL>REM   carlos.sierra@oracle.com

SQL>REM

SQL>REM SCRIPT

SQL>REM   coe_xfr_sql_profile_cpk9jsg2qt52r_3384190782.sql

SQL>REM

SQL>REM DESCRIPTION

SQL>REM   This script is generated by coe_xfr_sql_profile.sql

SQL>REM   It contains the SQL*Plus commands to create a custom

SQL>REM   SQL Profile for SQL_ID cpk9jsg2qt52r based on plan hash

SQL>REM   value 3384190782.

SQL>REM   The custom SQL Profile to be created by this script

SQL>REM   will affect plans for SQL commands with signature

SQL>REM   matching the one for SQL Text below.

SQL>REM   Review SQL Text and adjust accordingly.

SQL>REM

SQL>REM PARAMETERS

SQL>REM   None.

SQL>REM

SQL>REM EXAMPLE

SQL>REM   SQL> START coe_xfr_sql_profile_cpk9jsg2qt52r_3384190782.sql;

SQL>REM

SQL>REM NOTES

SQL>REM   1. Should be run as SYSTEM or SYSDBA.

SQL>REM   2. User must have CREATE ANY SQL PROFILE privilege.

SQL>REM   3. SOURCE and TARGET systems can be the same or similar.

SQL>REM   4. To drop this custom SQL Profile after it has been created:

SQL>REM  EXEC DBMS_SQLTUNE.DROP_SQL_PROFILE('coe_cpk9jsg2qt52r_3384190782');

SQL>REM   5. Be aware that using DBMS_SQLTUNE requires a license

SQL>REM  for the Oracle Tuning Pack.

SQL>REM   6. If you modified a SQL putting Hints in order to produce a desired

SQL>REM  Plan, you can remove the artifical Hints from SQL Text pieces below.

SQL>REM  By doing so you can create a custom SQL Profile for the original

SQL>REM  SQL but with the Plan captured from the modified SQL (with Hints).

SQL>REM

SQL>WHENEVER SQLERROR EXIT SQL.SQLCODE;

SQL>REM

SQL>VAR signature NUMBER;

SQL>VAR signaturef NUMBER;

SQL>REM

SQL>DECLARE

  2  sql_txt CLOB;

  3  h       SYS.SQLPROF_ATTR;

  4  PROCEDURE wa (p_line IN VARCHAR2) IS

  5  BEGIN

  6  DBMS_LOB.WRITEAPPEND(sql_txt, LENGTH(p_line), p_line);

  7  END wa;

  8  BEGIN

  9  DBMS_LOB.CREATETEMPORARY(sql_txt, TRUE);

10  DBMS_LOB.OPEN(sql_txt, DBMS_LOB.LOB_READWRITE);

11  -- SQL Text pieces below do not have to be of same length.

12  -- So if you edit SQL Text (i.e. removing temporary Hints),

13  -- there is no need to edit or re-align unmodified pieces.

14  wa(q'[select * from scott.test where object_id=10]');

15  DBMS_LOB.CLOSE(sql_txt);

16  h := SYS.SQLPROF_ATTR(

17  q'[BEGIN_OUTLINE_DATA]',

18  q'[IGNORE_OPTIM_EMBEDDED_HINTS]',

19  q'[OPTIMIZER_FEATURES_ENABLE('11.2.0.4')]',

20  q'[DB_VERSION('11.2.0.4')]',

21  q'[ALL_ROWS]',

22  q'[OUTLINE_LEAF(@"SEL$1")]',

23  q'[FULL(@"SEL$1" "TEST"@"SEL$1")]',

24  q'[END_OUTLINE_DATA]');

25  :signature := DBMS_SQLTUNE.SQLTEXT_TO_SIGNATURE(sql_txt);

26  :signaturef := DBMS_SQLTUNE.SQLTEXT_TO_SIGNATURE(sql_txt, TRUE);

27  DBMS_SQLTUNE.IMPORT_SQL_PROFILE (

28  sql_text    => sql_txt,

29  profile     => h,

30  name        => 'coe_cpk9jsg2qt52r_3384190782',

31  description => 'coe cpk9jsg2qt52r 3384190782 '||:signature||' '||:signaturef||'',

32  category    => 'DEFAULT',

33  validate    => TRUE,

34  replace     => TRUE,

35  force_match => FALSE /* TRUE:FORCE (match even when different literals in SQL). FALSE:EXACT (similar to CURSOR_SHARING) */ );

36  DBMS_LOB.FREETEMPORARY(sql_txt);

37  END;

38  /

 

PL/SQL procedure successfully completed.

 

SQL>WHENEVER SQLERROR CONTINUE

SQL>SET ECHO OFF;

 

            SIGNATURE

---------------------

10910590721604799112

 

 

           SIGNATUREF

---------------------

15966118871002195466

 

 

... manual custom SQL Profile has been created

 

 

COE_XFR_SQL_PROFILE_cpk9jsg2qt52r_3384190782 completed


8.查看产生的sql profile,此时原语句在不加hint的情况下也走全表扫了
select * from dba_sql_profiles;

 

SYS@dlhr> col sql_text for a50

SYS@dlhr> col hints for a50

SYS@dlhr>  SELECT b.name,to_char(d.sql_text) sql_text,  extractvalue(value(h),'.') as hints

  2     FROM dba_sql_profiles d,SYS.SQLOBJ$DATA A,

  3          SYS.SQLOBJ$ B,

  4          TABLE(XMLSEQUENCE(EXTRACT(XMLTYPE(A.COMP_DATA),

  5                                    '/outline_data/hint'))) h

  6    where a.signature = b.signature

  7      and a.category = b.category

  8      and a.obj_type = b.obj_type

  9      and a.plan_id = b.plan_id

10      and a.signature=d.signature

11      and D.name = 'coe_cpk9jsg2qt52r_3384190782';

 

NAME                           SQL_TEXT                                           HINTS

------------------------------ -------------------------------------------------- --------------------------------------------------

coe_cpk9jsg2qt52r_3384190782   select * from scott.test where object_id=10        BEGIN_OUTLINE_DATA

coe_cpk9jsg2qt52r_3384190782   select * from scott.test where object_id=10        IGNORE_OPTIM_EMBEDDED_HINTS

coe_cpk9jsg2qt52r_3384190782   select * from scott.test where object_id=10        OPTIMIZER_FEATURES_ENABLE('11.2.0.4')

coe_cpk9jsg2qt52r_3384190782   select * from scott.test where object_id=10        DB_VERSION('11.2.0.4')

coe_cpk9jsg2qt52r_3384190782   select * from scott.test where object_id=10        ALL_ROWS

coe_cpk9jsg2qt52r_3384190782   select * from scott.test where object_id=10        OUTLINE_LEAF(@"SEL$1")

coe_cpk9jsg2qt52r_3384190782   select * from scott.test where object_id=10        FULL(@"SEL$1" "TEST"@"SEL$1")

coe_cpk9jsg2qt52r_3384190782   select * from scott.test where object_id=10        END_OUTLINE_DATA

 

8 rows selected.

 

SYS@dlhr>

 

 

9.验证SQL Profile是否生效

 

SYS@dlhr> set autot traceonly explain stat

SYS@dlhr> select * from scott.test where object_id=10;

 

87076 rows selected.

 

 

Execution Plan

----------------------------------------------------------

Plan hash value: 217508114

 

--------------------------------------------------------------------------

| Id  | Operation         | Name | Rows  | Bytes | Cost (%CPU)| Time     |

--------------------------------------------------------------------------

|   0 | SELECT STATEMENT  |      |     1 |    98 |   351   (2)| 00:00:05 |

|*  1 |  TABLE ACCESS FULL| TEST |     1 |    98 |   351   (2)| 00:00:05 |

--------------------------------------------------------------------------

 

Predicate Information (identified by operation id):

---------------------------------------------------

 

   1 - filter("OBJECT_ID"=10)

 

Note

-----

   - SQL profile "coe_cpk9jsg2qt52r_3384190782" used for this statement

 

 

Statistics

----------------------------------------------------------

          0  recursive calls

          0  db block gets

       6973  consistent gets

          0  physical reads

          0  redo size

    4159482  bytes sent via SQL*Net to client

      64375  bytes received via SQL*Net from client

       5807  SQL*Net roundtrips to/from client

          0  sorts (memory)

          0  sorts (disk)

      87076  rows processed



注意:
1.这个测试只是为了演示通过coe_xfr_sql_profile.sql
实现手动加hint的方法,实际上面的语句问题的处理最佳的方法应该是重新收集scott.test的统计信息才对.
2.当一条sql既有sql profile又有stored outline,优化器优先选择stored outline.
3.force_match参数,TRUE:FORCE (match even when different literals in SQL),FALSE:EXACT (similar to CURSOR_SHARING).
4.通过sql profile手动加hint的方法很简单,而为sql添加最合理的hint才是关键.
5.测试完后,可以通过 exec dbms_sqltune.drop_sql_profile(name =>'coe_cpk9jsg2qt52r_3384190782' );删除这个sql profile.

6.执行coe_xfr_sql_profile.sql脚本的时候用户需要对当前目录有生成文件的权限,最好当前目录是/tmp



2.2.2.2  SQL Profile使用示例--使用STA来生成SQL Profile

利用STA对语句进行优化后,STA会对语句进行分析,采用最优的优化策略,并给出优化后的查询计划。你可以按照STA给出的建议重写语句。但是,有些情况下,你可能无法重写语句(比如在生产环境中,你的语句又在一个包中)。这个时候就可以利用sql profile,将优化策略存储在profile中,Oracle在构建这条语句的查询计划时,就不会使用已有相关统计数据,而使用profile的策略,生成新的查询计划。

 

一、 第一步:给用户赋权限

[ZHLHRSPMDB2:oracle]:/oracle>sqlplus / as sysdba

 

SQL*Plus: Release 11.2.0.4.0 Production on Wed May 25 16:47:29 2016

 

Copyright (c) 1982, 2013, Oracle.  All rights reserved.

 

 

Connected to:

Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production

With the Partitioning, Real Application Clusters, OLAP, Data Mining

and Real Application Testing options

 

SYS@dlhr>

SYS@dlhr>

SYS@dlhr>

SYS@dlhr> GRANT CREATE ANY SQL PROFILE TO LHR;

 

Grant succeeded.

 

SYS@dlhr> GRANT DROP ANY SQL PROFILE TO LHR;

 

Grant succeeded.

 

SYS@dlhr> GRANT ALTER ANY SQL PROFILE TO LHR;

 

Grant succeeded.

 

SYS@dlhr> conn lhr/lhr

Connected.

LHR@dlhr>

LHR@dlhr> select * from v$version;

 

BANNER

--------------------------------------------------------------------------------

Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production

PL/SQL Release 11.2.0.4.0 - Production

CORE    11.2.0.4.0      Production

TNS for IBM/AIX RISC System/6000: Version 11.2.0.4.0 - Production

NLSRTL Version 11.2.0.4.0 - Production

 

LHR@dlhr> create table lhr.TB_LHR_20160525_01 as select * from dba_objects;

 

Table created.

 

LHR@dlhr> create index lhr.TB_LHR_20160525_01_idx on TB_LHR_20160525_01(object_id);  

 

Index created.

 

LHR@dlhr> exec dbms_stats.gather_table_stats('lhr','TB_LHR_20160525_01',cascade=>true,degree=>4);  

 

PL/SQL procedure successfully completed.

 

LHR@dlhr> set autot on

LHR@dlhr> select /*+no_index(TB_LHR_20160525_01 TB_LHR_20160525_01_idx)*/count(*) from lhr.TB_LHR_20160525_01 where object_id = 100 ;

 

  COUNT(*)

----------

         1

 

 

Execution Plan

----------------------------------------------------------

Plan hash value: 3612989399

 

-----------------------------------------------------------------------------------------

| Id  | Operation          | Name               | Rows  | Bytes | Cost (%CPU)| Time     |

-----------------------------------------------------------------------------------------

|   0 | SELECT STATEMENT   |                    |     1 |     5 |   351   (2)| 00:00:05 |

|   1 |  SORT AGGREGATE    |                    |     1 |     5 |            |          |

|*  2 |   TABLE ACCESS FULL| TB_LHR_20160525_01 |     1 |     5 |   351   (2)| 00:00:05 |

-----------------------------------------------------------------------------------------

 

Predicate Information (identified by operation id):

---------------------------------------------------

 

   2 - filter("OBJECT_ID"=100)

 

 

Statistics

----------------------------------------------------------

          1  recursive calls

          0  db block gets

       1249  consistent gets

          0  physical reads

          0  redo size

        526  bytes sent via SQL*Net to client

        520  bytes received via SQL*Net from client

          2  SQL*Net roundtrips to/from client

          0  sorts (memory)

          0  sorts (disk)

          1  rows processed

LHR@dlhr> set autot off

LHR@dlhr> SELECT v.SQL_ID, v.SQL_TEXT FROM v$sql v WHERE v.SQL_TEXT like '%no_index(TB_LHR_20160525_01%' and v.SQL_TEXT not like '%v$sql%' ;

 

SQL_ID

-------------

SQL_TEXT

------------------------------------------------------------------------------------------------------------------------------------

7jt1btjkcczb8

select /*+no_index(TB_LHR_20160525_01 TB_LHR_20160525_01_idx)*/count(*) from lhr.TB_LHR_20160525_01 where object_id = 100

 

7suktf0w95cry

EXPLAIN PLAN SET STATEMENT_ID='PLUS150249' FOR select /*+no_index(TB_LHR_20160525_01 TB_LHR_20160525_01_idx)*/count(*) from lhr.TB_L

HR_20160525_01 where object_id = 100

 

二、 第二步创建、执行优化任务

LHR@dlhr> DECLARE

  2      my_task_name VARCHAR2(30);

  3      my_sqltext       CLOB;

  4  BEGIN

  5      my_sqltext := 'select /*+no_index(TB_LHR_20160525_01 TB_LHR_20160525_01_idx)*/count(*) from lhr.TB_LHR_20160525_01 where object_id = 100';

  6      my_task_name := DBMS_SQLTUNE.CREATE_TUNING_TASK(

  7                              sql_text          => my_sqltext,

  8                              user_name       => 'LHR',

  9                              scope           => 'COMPREHENSIVE',

10                              time_limit    => 60,

11                              task_name       => 'sql_profile_test',

12                              description => 'Task to tune a query on a specified table');

13      DBMS_SQLTUNE.EXECUTE_TUNING_TASK( task_name => 'sql_profile_test');

14  END;

15  /

 

PL/SQL procedure successfully completed.

 

或者也可以使用sqlid来生成优化任务,如下:

 

LHR@dlhr> DECLARE

  2    a_tuning_task VARCHAR2(30);

  3  BEGIN

  4    a_tuning_task := dbms_sqltune.create_tuning_task(sql_id    => '7jt1btjkcczb8',

  5                                                     task_name => 'sql_profile_test_SQLID');

  6    dbms_sqltune.execute_tuning_task(a_tuning_task);

  7  END;

  8  /

 

PL/SQL procedure successfully completed.

三、 第三步:查看优化建议

LHR@dlhr> set autot off

LHR@dlhr> set long 10000

LHR@dlhr> set longchunksize 1000

LHR@dlhr> set linesize 100

LHR@dlhr> SELECT DBMS_SQLTUNE.REPORT_TUNING_TASK( 'sql_profile_test') from DUAL;

 

DBMS_SQLTUNE.REPORT_TUNING_TASK('SQL_PROFILE_TEST')

----------------------------------------------------------------------------------------------------

GENERAL INFORMATION SECTION

-------------------------------------------------------------------------------

Tuning Task Name   : sql_profile_test

Tuning Task Owner  : LHR

Workload Type      : Single SQL Statement

Scope              : COMPREHENSIVE

Time Limit(seconds): 60

Completion Status  : COMPLETED

Started at         : 05/25/2016 16:58:31

Completed at       : 05/25/2016 16:58:32

 

-------------------------------------------------------------------------------

Schema Name: LHR

SQL ID     : 9kzm8scz6t92z

SQL Text   : select /*+no_index(TB_LHR_20160525_01

             TB_LHR_20160525_01_idx)*/count(*) from lhr.TB_LHR_20160525_01

             where object_id = 100

 

-------------------------------------------------------------------------------

FINDINGS SECTION (1 finding)

-------------------------------------------------------------------------------

 

1- SQL Profile Finding (see explain plans section below)

--------------------------------------------------------

  A potentially better execution plan was found for this statement.

 

  Recommendation (estimated benefit: 99.83%)

  ------------------------------------------

  - Consider accepting the recommended SQL profile.

    execute dbms_sqltune.accept_sql_profile(task_name => 'sql_profile_test',

            task_owner => 'LHR', replace => TRUE);

 

  Validation results

  ------------------

  The SQL profile was tested by executing both its plan and the original plan

  and measuring their respective execution statistics. A plan may have been

  only partially executed if the other could be run to completion in less time.

 

                           Original Plan  With SQL Profile  % Improved

                           -------------  ----------------  ----------

  Completion Status:            COMPLETE          COMPLETE

  Elapsed Time (s):             .006278            .00004      99.36 %

  CPU Time (s):                 .003397           .000021      99.38 %

  User I/O Time (s):                  0                 0

  Buffer Gets:                     1249                 2      99.83 %

  Physical Read Requests:             0                 0

  Physical Write Requests:            0                 0

 

DBMS_SQLTUNE.REPORT_TUNING_TASK('SQL_PROFILE_TEST')

----------------------------------------------------------------------------------------------------

  Physical Read Bytes:                0                 0

  Physical Write Bytes:               0                 0

  Rows Processed:                     1                 1

  Fetches:                            1                 1

  Executions:                         1                 1

 

  Notes

  -----

  1. Statistics for the original plan were averaged over 10 executions.

  2. Statistics for the SQL profile plan were averaged over 10 executions.

 

-------------------------------------------------------------------------------

EXPLAIN PLANS SECTION

-------------------------------------------------------------------------------

 

1- Original With Adjusted Cost

------------------------------

Plan hash value: 3612989399

 

-----------------------------------------------------------------------------------------

| Id  | Operation          | Name               | Rows  | Bytes | Cost (%CPU)| Time     |

-----------------------------------------------------------------------------------------

|   0 | SELECT STATEMENT   |                    |     1 |     5 |   351   (2)| 00:00:05 |

|   1 |  SORT AGGREGATE    |                    |     1 |     5 |            |          |

|*  2 |   TABLE ACCESS FULL| TB_LHR_20160525_01 |     1 |     5 |   351   (2)| 00:00:05 |

-----------------------------------------------------------------------------------------

 

Predicate Information (identified by operation id):

---------------------------------------------------

 

   2 - filter("OBJECT_ID"=100)

 

2- Using SQL Profile

--------------------

Plan hash value: 661515879

 

--------------------------------------------------------------------------------------------

| Id  | Operation         | Name                   | Rows  | Bytes | Cost (%CPU)| Time     |

--------------------------------------------------------------------------------------------

|   0 | SELECT STATEMENT  |                        |     1 |     5 |     1   (0)| 00:00:01 |

|   1 |  SORT AGGREGATE   |                        |     1 |     5 |            |          |

|*  2 |   INDEX RANGE SCAN| TB_LHR_20160525_01_IDX |     1 |     5 |     1   (0)| 00:00:01 |

--------------------------------------------------------------------------------------------

 

Predicate Information (identified by operation id):

---------------------------------------------------

 

 

DBMS_SQLTUNE.REPORT_TUNING_TASK('SQL_PROFILE_TEST')

----------------------------------------------------------------------------------------------------

   2 - access("OBJECT_ID"=100)

 

-------------------------------------------------------------------------------

 

 

这里可以看到,在优化建议中给出了新的查询计划。现在,我们决定接受这个建议,并且不重写语句。

 

四、 步:接受profile

LHR@dlhr> set autot on

LHR@dlhr> select /*+no_index(TB_LHR_20160525_01 TB_LHR_20160525_01_idx)*/count(*) from lhr.TB_LHR_20160525_01 where object_id = 100 ;

 

  COUNT(*)

----------

         1

 

 

Execution Plan

----------------------------------------------------------

 

Plan hash value: 3612989399

 

-----------------------------------------------------------------------------------------

| Id  | Operation          | Name               | Rows  | Bytes | Cost (%CPU)| Time     |

-----------------------------------------------------------------------------------------

|   0 | SELECT STATEMENT   |                    |     1 |     5 |   351   (2)| 00:00:05 |

|   1 |  SORT AGGREGATE    |                    |     1 |     5 |            |          |

|*  2 |   TABLE ACCESS FULL| TB_LHR_20160525_01 |     1 |     5 |   351   (2)| 00:00:05 |

-----------------------------------------------------------------------------------------

 

Predicate Information (identified by operation id):

---------------------------------------------------

 

   2 - filter("OBJECT_ID"=100)

 

 

Statistics

----------------------------------------------------------

          0  recursive calls

          0  db block gets

       1249  consistent gets

          0  physical reads

          0  redo size

        526  bytes sent via SQL*Net to client

        520  bytes received via SQL*Net from client

          2  SQL*Net roundtrips to/from client

          0  sorts (memory)

          0  sorts (disk)

          1  rows processed

LHR@dlhr> execute dbms_sqltune.accept_sql_profile(task_name =>'sql_profile_test_SQLID', task_owner => 'LHR', replace => TRUE);

 

PL/SQL procedure successfully completed.

 

LHR@dlhr> set autot off

LHR@dlhr>  SELECT e.task_name, b.name, d.sql_text, extractvalue(value(h), '.') as hints

  2     FROM dba_sql_profiles d,

  3           dba_advisor_tasks e,

  4          SYS.SQLOBJ$DATA A,

  5          SYS.SQLOBJ$ B,

  6          TABLE(XMLSEQUENCE(EXTRACT(XMLTYPE(A.COMP_DATA),

  7                                    '/outline_data/hint'))) h

  8    where a.signature = b.signature

  9      and a.category = b.category

10      and a.obj_type = b.obj_type

11      and a.plan_id = b.plan_id

12      and a.signature = d.signature

13      and d.task_id=e.task_id

14      and d.name = 'SYS_SQLPROF_0154e728ad3f0000'

15     ;

 

TASK_NAME                      NAME

------------------------------ ------------------------------

SQL_TEXT

----------------------------------------------------------------------------------------------------

HINTS

----------------------------------------------------------------------------------------------------

sql_profile_test               SYS_SQLPROF_0154e728ad3f0000

select /*+no_index(TB_LHR_20160525_01 TB_LHR_20160525_01_idx)*/count(*) from lhr.TB_LHR_20160525_01

where object_id = 100

OPTIMIZER_FEATURES_ENABLE(default)

 

sql_profile_test               SYS_SQLPROF_0154e728ad3f0000

select /*+no_index(TB_LHR_20160525_01 TB_LHR_20160525_01_idx)*/count(*) from lhr.TB_LHR_20160525_01

where object_id = 100

IGNORE_OPTIM_EMBEDDED_HINTS

 

 

 

在这里用了包DBMS_SQLTUNE的另一个函数:ACCEPT_SQL_PROFILE。其中,参数task_name即我们创建的优化建议任务的名称,nameprofile的名字,可以是任意合法名称。此外这个函数还有其他一些函数,下面是这个函数的原型:

DBMS_SQLTUNE.ACCEPT_SQL_PROFILE (

   task_name    IN  VARCHAR2,

   object_id    IN  NUMBER   := NULL,

   name         IN  VARCHAR2 := NULL,

   description  IN  VARCHAR2 := NULL,

   category     IN  VARCHAR2 := NULL;

   task_owner   IN VARCHAR2  := NULL,

   replace      IN BOOLEAN   := FALSE,

   force_match  IN BOOLEAN   := FALSE)

RETURN VARCHAR2;

 

Descriptionprofile的描述信息;task_owner是优化建议任务的所有者;replaceTRUE时,如果这个profile已经存在,就代替它;force_matchTURE时,表示与语句强制匹配,即强制使用绑定变量,和系统参数cursor_sharing设置为FORCE时类似,为FALSE时,与cursor_sharing设置为EXACT时类似,即完全匹配。

这里要特别提到的是category这个参数,你可以通过设置这个参数,制定特定会话使用这个profile。在10g中,每个会话都有一个新参数SQLTUNE_CATEGORY,他的默认值是DEFAULT。而我们在调用这个函数时,如果没有指定这个参数,那它的值也是DEFAULT,而如果我们给这个profile指定了一个其它的CATEGORY值,如FOR_TUNING,那么只有会话参SQLTUNE_CATEGORY也为FOR_TUNING时,才会使用这个porfile。为什么说这个参数很有用呢?试想一个这样的环境:你在一个生产系统上利用STA调优一条语句,STA已经给出了优化建议,但是你又不敢贸然实施它给出的建议(毕竟它只是机器嘛,不能完全信任),你就可以创建一个有特殊CATEGORYprofile,然后在你自己的会话中制定SQLTUNE_CATEGORY为这个特殊的CATEGORY,那就既可以看优化建议的实际效果又不影响生产环境。

 此外可以通过视图DBA_SQL_PROFILES来查看已经创建的profile

五、 步:查看profile的效果

LHR@dlhr> set autot on

LHR@dlhr> select /*+no_index(TB_LHR_20160525_01 TB_LHR_20160525_01_idx)*/count(*) from lhr.TB_LHR_20160525_01 where object_id = 100 ;

 

  COUNT(*)

----------

         1

 

 

Execution Plan

----------------------------------------------------------

Plan hash value: 661515879

 

--------------------------------------------------------------------------------------------

| Id  | Operation         | Name                   | Rows  | Bytes | Cost (%CPU)| Time     |

--------------------------------------------------------------------------------------------

|   0 | SELECT STATEMENT  |                        |     1 |     5 |     1   (0)| 00:00:01 |

|   1 |  SORT AGGREGATE   |                        |     1 |     5 |            |          |

|*  2 |   INDEX RANGE SCAN| TB_LHR_20160525_01_IDX |     1 |     5 |     1   (0)| 00:00:01 |

--------------------------------------------------------------------------------------------

 

Predicate Information (identified by operation id):

---------------------------------------------------

 

   2 - access("OBJECT_ID"=100)

 

Note

-----

   - SQL profile "SYS_SQLPROF_0154e728ad3f0000" used for this statement

 

 

Statistics

----------------------------------------------------------

          1  recursive calls

          0  db block gets

          2  consistent gets

          0  physical reads

          0  redo size

        526  bytes sent via SQL*Net to client

        520  bytes received via SQL*Net from client

          2  SQL*Net roundtrips to/from client

          0  sorts (memory)

          0  sorts (disk)

          1  rows processed

 

 

NOTE部分可以看到,语句采用了profile中的数据,创建了新的查询计划。并且在查询计划中还有一些附加信息,表明这个语句是采用了SYS_SQLPROF_0154e728ad3f0000这个profile,而不是根据对象上面的统计数据来生成的查询计划。

但上述方法主要是依赖sql tuning advisor,如果它无法生成你想要的执行计划.你还可以通过手动的方式,通过sql profilehint加进去.复杂的SQLhint可以采用脚本coe_xfr_sql_profile.sql来产生原语句的outline data和加hint语句的outline data,然后替换对应的SYS.SQLPROF_ATTR,最后执行生成的sql就可以了.

 

使用PLSQL DEVELOPER 11查看执行计划,如下图,新版本的好处:

wpsA6.tmp 

 

 

2.3  SPM(SQL Plan Management)

2.3.1  SPM基础知识

 

SQL 语句的SQL 执行计划发生更改时,可能存在性能风险。

SQL 计划发生更改的原因有很多,如优化程序版本、优化程序统计信息、优化程序参数、方案定义、系统设计和SQL 概要文件创建等。

在以前版本的Oracle DB 中引入了各种计划控制技术(如存储的大纲(storedoutline(9i))SQL 概要文件等(SQLprofile(10g))),用于解决计划更改导致的性能回归。但是,这些技术都是需要手动干预的被动式进程。

SQL 计划管理是一种随Oracle Database 11g 引入的新功能,通过维护所谓的SQL 计划基线(SQL plan baseline(11g))来使系统能够自动控制SQL 计划演变。启用此功能后,只要证明新生成的SQL 计划与SQL 计划基线相集成不会导致性能回归,就可以进行此项集成。因此,在执行某个SQL 语句时,只能使用对应的SQL 计划基线中包括的计划。可以使用SQL 优化集自动加载或植入SQL 计划基线。

SQL 计划管理功能的主要优点是系统性能稳定,不会出现计划回归。此外,该功能还可以节省DBA 的许多时间,这些时间通常花费在确定和分析SQL 性能回归以及寻找可用的解决方案上。Oracle11g中,Oracle提供dbms_spm包来管理SQL PlanSPM是一个预防机制,它记录并评估sql的执行计划,将已知的高效的sql执行计划建立为SQL Plan BaselinesSQL Plan Baseline的功能是保持SQL的性能而不必关注系统的改变。

 

在SQL Plan BaseLines捕获阶段,Oracle记录SQL的执行计划并检测该执行计划是否已经改变,如果SQL改变后的执行计划是安全的,则SQL就使用新的执行计划,因此,Oracle维护单个SQL执行计划的历史信息,Oracle维护的SQL执行计划的历史仅仅针对重复执行的SQLSQL Plan Baseline可以手工load,也可以设置为自动捕获。

 

加载SQL 计划基线的方式有两种:

(1)  即时捕获,自动捕获(Automatic Plan Capture)

使用自动计划捕获,方法是:将初始化参数OPTIMIZER_CAPTURE_SQL_PLAN_BASELINES 设置为TRUE。默认情况下,该参数设置为FALSE。将该参数设置为TRUE 将打开自动标识可重复SQL 语句,以及自动为此类语句创建计划历史记录的功能。 如果要激活自动的SQL Plan Capture,则需要设置OPTIMIZER_CAPTURE_SQL_PLAN_BASELINES,该参数默认为False,如果设置为True,则表示自动捕获SQL Plan,则系统会自动创建并维护SQL Plan HistorySQL Plan History包括优化器关注的:比如an execution plan, SQL text, outline, bind variables, and compilation environment

 

(2)  成批加载Manual Plan Loading

使用DBMS_SPM 程序包;该程序包支持手动管理SQL 计划基线。使用此程序包,可以将SQL 计划从游标高速缓存或现有的SQL 优化集(STS) 直接加载到SQL计划基线中。对于要从STS 加载到SQL 计划基线的SQL 语句,需要将其SQL计划存储在STS中。使用DBMS_SPM 可以将基线计划的状态从已接受更改为未接受(以及从未接受更改为已接受),还可以从登台表导出基线计划,然后使用导出的基线计划将SQL 计划基线加载到其它数据库中。

 

也可以手动装载一个存在的SQL Plan作为SQL Plan Baseline,手动装载的SQL Plan并不校验它的性能:

--SQL Tuning Set中装载:

DECLARE

  my_plans pls_integer;

BEGIN

  my_plans := DBMS_SPM.LOAD_PLANS_FROM_SQLSET(sqlset_name => 'tset1');

END;

/

 

--Cursor Cache中装载

DECLARE my_plans pls_integer;

BEGIN

  my_plans := DBMS_SPM.LOAD_PLANS_FROM_CURSOR_CACHE(sql_id => '7qqnad1j615m7');

END;

/

 

在SQL 计划基线演化阶段,Oracle DB 会按常规方式评估新计划的性能,并将性能较好的计划集成到SQL 计划基线中。

优化程序为SQL 语句找到新的计划时,会将该计划作为未接受的计划添加到计划历史记录中。然后,相对于SQL 计划基线的性能,验证该计划的性能。如果经验证某个未接受的计划不会导致性能回归(手动或自动),则该计划会被更改为已接受计划,并集成到SQL 计划基线中。成功验证未接受计划的过程包括:对此计划的性能和从SQL计划基线中选择的一个计划的性能进行比较,确保其性能更佳。

 

演化SQL 计划基线的方式有两种:

(1)使用DBMS_SPM.EVOLVE_SQL_PLAN_BASELINE 函数。该函数将返回一个报表,显示是否已将一些现有的历史记录计划移到了计划基线中。也可以在历史记录中指定要测试的特定计划。

 

(2)运行SQL 优化指导:通过使用SQL 优化指导手动或自动优化SQL 语句,演化SQL计划基线。SQL优化指导发现已优化的计划,并确认其性能优于从相应的SQL 计划基线中选择的计划的性能时,就会生成一个建议案以接受SQL 概要文件。接受了该SQL 概要文件后,会将已优化的计划添加到相应的SQL 计划基线中。

 

在SQL Plan Baselines的演变阶段,Oracle评估新的Plan的性能并将性能较好的Plan存放SQL Plan Baselines中,可以使用dbms_spm package的过程EVOLVE_SQL_PLAN_BASELINE将新的SQL Plan存入已经存在的SQL Plan Baselines中,新的Plan将会作为已经Accept Plan加入到SQL Plan Baselines中。

SET SERVEROUTPUT ON

SET LONG 10000

DECLARE report clob;

BEGIN report := DBMS_SPM.EVOLVE_SQL_PLAN_BASELINE (sql_handle => 'SYS_SQL_593bc74fca8e6738');

DBMS_OUTPUT.PUT_LINE(report);

END;

/

 

如果将计划添加到计划历史记录中,则该计划将与一些重要的属性关联:

(1SIGNATURESQL_HANDLESQL_TEXT PLAN_NAME 是搜索操作的重要标识符。

 

(2)使用ORIGIN 可以确定计划是自动捕获的(AUTO-CAPTURE)、手动演化的(MANUALLOAD)、通过SQL 优化指导自动演化的(MANUAL-SQLTUNE) 还是通过自动SQL 优化自动演化的(AUTO-SQLTUNE)

 

(3)  ENABLED ACCEPTED:ENABLED属性表示计划已启用,可供优化程序使用。如果未设置ENABLED,则系统将不考虑此计划。ACCEPTED 属性表示用户在将计划更改为ACCEPTED 时计划已经过验证为有效计划(系统自动进行的或用户手动进行的)。如果将某个计划更改为ACCEPTED,则仅当使用DBMS_SPM.ALTER_SQL_PLAN_BASELINE()更改其状态时,该计划才是非ACCEPTED 的。可以通过删除ENABLED设置暂时禁用ACCEPTED 计划。计划必须为ENABLED 和ACCEPTED,优化程序才会考虑使用它。

 

(4)  FIXED 表示优化程序仅考虑标记为FIXED 的计划,而不考虑其它计划。例如,如果有10 个基线计划,其中的三个计划被标记为FIXED,则优化程序将仅使用这三个计划中的最佳计划,而忽略其它所有计划。如果某个SQL 计划基线至少包含一个已启用的已修复计划,则该SQL 计划基线就是FIXED 的。如果在修复的SQL 计划基线中添加了新计划,则在手动将这些新计划声明为FIXED 之前,无法使用这些新计划。

可以使用DBA_SQL_PLAN_BASELINES视图查看每个计划的属性。然后,可以使用DBMS_SPM.ALTER_SQL_PLAN_BASELINE 函数更改其中的某些属性。也可以使用DBMS_SPM.DROP_SQL_PLAN_BASELINE 函数删除计划或整个计划历史记录。

注:DBA_SQL_PLAN_BASELINES 视图包含了一些附加属性;使用这些属性可以确定各个计划的上次使用时间,以及是否应自动清除某个计划。

 

如果使用的是自动计划捕获,则第一次将某个SQL 语句标识为可重复时,其最佳成本计划将被添加到对应的SQL 计划基线中。然后,该计划将用于执行相应的语句。