利用sqlprofile固定執行計劃并将執行計劃導入到新庫

本文主要講的是如何利用sqlprofile固定執行計劃，已經執行計劃的導出導入。對于資料遷移導緻執行計劃不穩定有一定的幫助。

1 實驗環境

Linux 11G R2 導入到 windows 11G R2

源庫：

SQL> 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 Linux: Version 11.2.0.4.0 - Production

NLSRTL Version 11.2.0.4.0 - Production

目标庫：

SQL> 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 64-bit Windows: Version 11.2.0.4.0 - Production

NLSRTL Version 11.2.0.4.0 - Production

1.1 建立表

SQL> conn lei/lei
Connected.

SQL> create table tt as select * from dba_objects;
Table created.

SQL> create index idex_01 on tt(object_id);
Index created.

1.2 收集統計資訊

SQL> exec dbms_stats.gather_table_stats('LEI','TT',cascade=>true);
PL/SQL procedure successfully completed.

1.3 生成執行計劃

SQL> explain plan for select object_NAME FROM TT WHERE object_id=2;
Exlained.

SQL>  select * from table(dbms_xplan.display());
PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------
Plan hash value: 2974445191
--------------------------------------------------------------------------------
| Id  | Operation     | Name    | Rows  | Bytes | Cost (%CPU)| Time     |
--------------------------------------------------------------------------------
|   0 | SELECT STATEMENT      |       |     1 |    30 |     2 (0)| 00:00:01 |
|   1 |  TABLE ACCESS BY INDEX ROWID| TT      |     1 |    30 |     2 (0)| 00:00:01 |
|*  2 |   INDEX RANGE SCAN     | IDEX_01 |     1 |       |     1  (0)| 00:00:01 |
--------------------------------------------------------------------------------
PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
 2 - access("OBJECT_ID"=2)


14 rows selected.

可以看到是走索引的。

1.4 使用HINT改變執行計劃

SQL> select /*+ full(tt) */* from tt where object_id=2;


OWNER

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

OBJECT_NAME

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

SUBOBJECT_NAME  OBJECT_ID DATA_OBJECT_ID OBJECT_TYPE

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

CREATED   LAST_DDL_ TIMESTAMP STATUS  T G S  NAMESPACE

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

EDITION_NAME

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

SYS

C_OBJ#

2        2 CLUSTER


OWNER

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

OBJECT_NAME

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

SUBOBJECT_NAME  OBJECT_ID DATA_OBJECT_ID OBJECT_TYPE

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

CREATED   LAST_DDL_ TIMESTAMP STATUS  T G S  NAMESPACE

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

EDITION_NAME

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

24-AUG-13 24-AUG-13 2013-08-24:11:37:35 VALID N N N        5

1.5 檢視outline

SQL> explain plan for select /*+ full(tt) */* from tt where object_id=2;

Explained.


SQL> select * from table(dbms_xplan.display(null,null,'outline'));


PLAN_TABLE_OUTPUT

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

Plan hash value: 264906180

--------------------------------------------------------------------------
| Id  | Operation   | Name | Rows  | Bytes | Cost (%CPU)| Time   |
--------------------------------------------------------------------------
|   0 | SELECT STATEMENT  |  |     1 |    98 |   344   (1)| 00:00:05 |
|*  1 |  TABLE ACCESS FULL| TT   |     1 |    98 |   344   (1)| 00:00:05 |
--------------------------------------------------------------------------
PLAN_TABLE_OUTPUT

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

  /*+

      BEGIN_OUTLINE_DATA

      FULL(@"SEL$1" "TT"@"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

  */

PLAN_TABLE_OUTPUT
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   1 - filter("OBJECT_ID"=2)

27 rows selected.

1.6 生成sql profile

SQL> declare

        v_hints sys.sqlprof_attr;

  begin

        v_hints := sys.sqlprof_attr('FULL(@"SEL$1" "TT"@"SEL$1")');   --從上面獲得

        dbms_sqltune.import_sql_profile('select * from tt where object_id= 2',    --sql語句

         v_hints, 'TT_LEI_20170510',   --profile名稱

        force_match => true);

  end;

 /  8    9   10  

 

PL/SQL procedure successfully completed.

1.7 檢視profile是否生效

SQL> explain plan for select * from tt where object_id=2;

Explained.
SQL> select * from table(dbms_xplan.display);


PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Plan hash value: 264906180
--------------------------------------------------------------------------
| Id  | Operation   | Name | Rows  | Bytes | Cost (%CPU)| Time  |
--------------------------------------------------------------------------
|   0 | SELECT STATEMENT  |    |     1 |    98 |   344   (1)| 00:00:05 |
|*  1 |  TABLE ACCESS FULL| TT    |     1 |    98 |   344   (1)| 00:00:05 |
--------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
   1 - filter("OBJECT_ID"=2)

Note
--------
   - SQL profile "TT_LEI_20170510" used for this statement

17 rows selected.

可以看到已經生效了。

2 導出表和打包執行計劃

2.1 打包執行計劃

SQL> exec DBMS_SQLTUNE.CREATE_STGTAB_SQLPROF(table_name=>'TEST_TT_PROFILE1',schema_name=>'LEI');

PL/SQL procedure successfully completed.

SQL> exec DBMS_SQLTUNE.PACK_STGTAB_SQLPROF(staging_table_name
=>'TEST_TT_PROFILE1',profile_name=>'TT_LEI_20170510');

PL/SQL procedure successfully completed

名稱随便。

更多關于DBMS_SQLTUNE.CREATE_STGTAB_SQLPROF的說明，請檢視官方文檔：

http://docs.oracle.com/cd/B28359_01/appdev.111/b28419/d_sqltun.htm#CACBCEEH

2.2 導出使用者LEI

[oracle@dg-p ~]$ expdp system/oracle dumpfile=tt.dmp directory=lei_dir schemas=lei
 
Export: Release 11.2.0.4.0 - Production on Wed May 10 20:09:28 2017 
Copyright (c) 1982, 2011, Oracle and/or its affiliates.  All rights reserved.

Connected to: Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
Starting "SYSTEM"."SYS_EXPORT_SCHEMA_01":  system/******** dumpfile=tt.dmp directory=lei_dir schemas=lei
Estimate in progress using BLOCKS method...
Processing object type SCHEMA_EXPORT/TABLE/TABLE_DATA
Total estimation using BLOCKS method: 10.18 MB
Processing object type SCHEMA_EXPORT/USER
Processing object type SCHEMA_EXPORT/SYSTEM_GRANT
Processing object type SCHEMA_EXPORT/ROLE_GRANT
Processing object type SCHEMA_EXPORT/DEFAULT_ROLE
Processing object type SCHEMA_EXPORT/PRE_SCHEMA/PROCACT_SCHEMA
Processing object type SCHEMA_EXPORT/TABLE/TABLE
Processing object type SCHEMA_EXPORT/TABLE/INDEX/INDEX
Processing object type SCHEMA_EXPORT/TABLE/INDEX/STATISTICS/INDEX_STATISTICS
Processing object type SCHEMA_EXPORT/TABLE/STATISTICS/TABLE_STATISTICS
. . exported "LEI"."TT"                                  8.366 MB   86269 rows
. . exported "LEI"."TEST_TT_PROFILE1"                    22.02 KB       1 rows
Master table "SYSTEM"."SYS_EXPORT_SCHEMA_01" successfully loaded/unloaded
******************************************************************************
Dump file set for SYSTEM.SYS_EXPORT_SCHEMA_01 is:
  /u01/backup/tt.dmp
Job "SYSTEM"."SYS_EXPORT_SCHEMA_01" successfully completed at Wed May 10 20:09:40 2017 elapsed 0 00:00:12

2.3 導入到新環境

2.3.1 建立使用者

SQL> create user lei identified by lei;
使用者已建立。

SQL> grant dba,resource,connect to lei;
授權成功。

SQL>
C:/Users/Administrator>impdp system/oracle dumpfile=tt.dmp directory=lei_dir schemas=LEI
Import: Release 11.2.0.4.0 - Production on 星期三 5月 10 12:05:09 2017
Copyright (c) 1982, 2011, Oracle and/or its affiliates.  All rights reserved.
 
連接配接到: Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
已成功加載/解除安裝了主表 "SYSTEM"."SYS_IMPORT_SCHEMA_01"
啟動 "SYSTEM"."SYS_IMPORT_SCHEMA_01":  system/******** dumpfile=tt.dmp directory=lei_dir schemas=LEI
處理對象類型 SCHEMA_EXPORT/USER
ORA-31684: 對象類型 USER:"LEI" 已存在
處理對象類型 SCHEMA_EXPORT/SYSTEM_GRANT
處理對象類型 SCHEMA_EXPORT/ROLE_GRANT
處理對象類型 SCHEMA_EXPORT/DEFAULT_ROL
處理對象類型 SCHEMA_EXPORT/PRE_SCHEMA/PROCACT_SCHEMA
處理對象類型 SCHEMA_EXPORT/TABLE/TABLE
處理對象類型 SCHEMA_EXPORT/TABLE/TABLE_DATA
. . 導入了 "LEI"."TT"                                  8.366 MB   86269 行
. . 導入了 "LEI"."TEST_TT_PROFILE1"                    22.02 KB       1 行
處理對象類型 SCHEMA_EXPORT/TABLE/INDEX/INDEX
處理對象類型 SCHEMA_EXPORT/TABLE/INDEX/STATISTICS/INDEX_STAISTICS
處理對象類型 SCHEMA_EXPORT/TABLE/STATISTICS/TABLE_STATISTICS
作業 "SYSTEM"."SYS_IMPORT_SCHEMA_01" 已經完成, 但是有 1 個錯誤 (于 星期三 5月 10 12:05:12 2017 elapsed 0 00:00:03 完成)

2.3.2 檢視新庫中的執行計劃

SQL> conn lei/lei
已連接配接。

SQL> explain plan for select * from tt where object_id=2;
已解釋。

SQL>  select * from table(dbms_xplan.display());
PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------
Plan hash value: 2974445191
----------------------------------------------------------------------------------------
| Id  | Operation                   | Name    | Rows  | Bytes | Cost (%CPU)| Time     |
---------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT            |         |     1 |    98 |     2   (0)| 00:00:01 |
|   1 |  TABLE ACCESS BY INDEX ROWID| TT      |     1 |    98 |     2   (0)| 00:00:01 |
|*  2 |   INDEX RANGE SCAN          | IDEX_01 |     1 |       |     1   (0)| 00:00:01 |
---------------------------------------------------------------------------------------
PLAN_TABLE_OUTPUT
---------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
   2 - access("OBJECT_ID"=2)

已選擇14行。

2.3.3 解包sqlprofile,執行計劃變更為與源庫一樣的執行計劃。

SQL> EXEC DBMS_SQLTUNE.UNPACK_STGTAB_SQLPROF(replace => TRUE,staging_table_name => 'TEST_TT_PROFILE1');
PL/SQL 過程已成功完成。

2.3.4 再次檢視執行計劃

SQL> explain plan for select * from tt where object_id=2;
已解釋。

SQL>  select * from table(dbms_xplan.display());

PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------
Plan hash value: 264906180
--------------------------------------------------------------------------
| Id  | Operation         | Name | Rows  | Bytes | Cost (%CPU)| Time     |
-------------------------------------------------------------------------
|   0 | SELECT STATEMENT  |      |     1 |    98 |   344   (1)| 00:00:05 |
|*  1 |  TABLE ACCESS FULL| TT   |     1 |    98 |   344   (1)| 00:00:05 |
--------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------
   1 - filter("OBJECT_ID"=2)
 
Note
-----
   - SQL profile "TT_LEI_20170510" used for this statement

已選擇17行。

利用sqlprofile固定執行計劃并将執行計劃導入到新庫

1 實驗環境

1.1 建立表

1.2 收集統計資訊

1.3 生成執行計劃

1.4 使用HINT改變執行計劃

1.5 檢視outline

1.6 生成sql profile

1.7 檢視profile是否生效

2 導出表和打包執行計劃

2.1 打包執行計劃

2.2 導出使用者LEI

2.3 導入到新環境

2.3.1 建立使用者

2.3.2 檢視新庫中的執行計劃

2.3.3 解包sqlprofile,執行計劃變更為與源庫一樣的執行計劃。

2.3.4 再次檢視執行計劃

繼續閱讀

ASP程式設計中20個非常有用的例子

龍珠訓練營task04

阿裡雲天池龍珠計劃SQL訓練營打卡

阿裡雲天池龍珠計劃SQL訓練營day1

賦予使用者dba權限--解釋以上：

實驗樓sql進階之成績管理系統的資料操作(window)

Oracle的基本操作

為什麼要選擇UniDAC

SQL優化SQL語句優化的目的

JAVA高效程式設計指南

windows不能在本地計算機上運作oracleDbConsoleorcl

關于SQL語言

SQL語言基礎：常用的資料查詢語句

neo4j之cypher使用文檔

sqlServer根據經緯查距離

Oracle 批量查詢傳入List 傳回List