.Oracle固定执行计划之SQL PROFILE概要文件
1. 引子
oracle系统为了合理分配和使用系统的资源提出了概要文件的概念。所谓概要文件，就是一份描述如何使用系统的资源(主要是cpu资源)的配置文件。将概要文件赋予某个数据库用户，在用户连接并访问数据库服务器时，系统就按照概要文件给他分配资源。
包括:
1、管理数据库系统资源。
利用profile来分配资源限额，必须把初始化参数resource_limit设置为true默认是true的。
2、管理数据库口令及验证方式。
默认给用户分配的是default概要文件，将该文件赋予了每个创建的用户。但该文件对资源没有任何限制，因此管理员常常需要根据自己数据库系统的环境自行建立概要文件。
2. 概要文件限制
概要文件主要可以对数据库系统如下指标进行限制。
1）用户的最大并发会话数（session_per_user)
2）每个会话的cpu时钟限制（cpu_per_session）
3）每次调用的cpu时钟限制，调用包含解析、执行命令和获取数据等等。（cpu_per_call）
4）最长连接时间。一个会话的连接时间超过指定时间之后，oracle会自动的断开连接（connect_time）
5）最长空闲时间。如果一个会话处于空闲状态超过指定时间，oracle会自动断开连接（idle_time）
6）每个会话可以读取的最大数据块数量（logical_reads_per_session）
7）每次调用可以读取的最大数据块数量（logical_reads_per_call）
8）sga私有区域的最大容量（private_sga）
概要文件对口令的定义和限制如下：
1）登录失败的最大尝试次数（failed_login_attempts）
2）口令的最长有效期(password_life_time)
3）口令在可以重用之前必须修改的次数(password_reuse_max)
4）口令在可以重用之前必须经过的天数(password_reuse_time)
5）超过登录失败的最大允许尝试次数后，账户被锁定的天数
6）指定用于判断口令复杂度的函数名
在指定概要文件之后，dba可以手工的将概要文件赋予每个用户。但是概要文件不是立即生效，而是要将初始化参数文件中的参数resource_limit设置为true之后，概要文件才会生效。
3. sql profile
sql profile在oracle10g中引入，主要目的侧重于sql优化，弥补了存储概要的缺点.
dba可以使用sql调整顾问(sta)或sql访问顾问(saa)来识别可以得到更好性能的sql语句，
这些语句可以保存在sql调整集、一个awr快照或保存在当前的库缓存中，一旦识别出调整候选者， 这些顾问程序就开始分析捕获到的语句以期获得更好的性能，然后生成专用的语句扩展(就叫做sql配置文件)并重写sql语句以在执行期间获取更佳的性能。
与存储概要类似，一个sql配置文件提供了使用更好的执行计划的能力(如果这个执行计
划是可用的)，sql配置文件也可以象存储概要一样分阶段执行，或限制到对特定会话才能执行该sql配置文件，但是大多数重要的改进超过了存储概要.
sqlprofile对于一下类型语句有效：
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update和insert操作时有效）。
另外，使用sql profile还必须有create any sql profile、drop any sql profile和alter any sql profile等系统权限。
4. 测试一
创建表
tpcc@toaddb> create table t1 as selectobject_id,object_name from dba_objects where rownum<=50000;
table created.
tpcc@toaddb> create table t2 as select * fromdba_objects;
table created.
创建索引：
tpcc@toaddb> create index t2_idx on t2(object_id);
index created.
收集统计信息：
tpcc@toaddb> execdbms_stats.gather_table_stats(user,'t1',cascade=>true,method_opt=>'forall columns size 1');
pl/sql procedure successfully completed.
tpcc@toaddb> execdbms_stats.gather_table_stats(user,'t1',cascade=>true,method_opt=>'forall columns size 1');
pl/sql procedure successfully completed.
执行无hint的sql
tpcc@toaddb> set autotrace on
tpcc@toaddb> select t1.*,t2.owner from t1,t2 wheret1.object_name like '%t1%' and t1.object_id=t2.object_id;
42 rows selected.
execution plan
----------------------------------------------------------
plan hash value: 1838229974
---------------------------------------------------------------------------
| id | operation | name | rows | bytes | cost (%cpu)| time |
---------------------------------------------------------------------------
| 0| select statement | | 2500 | 97k| 498 (1)| 00:00:01 |
|* 1| hash join | | 2500 | 97k| 498 (1)| 00:00:01 |
|* 2| table access full| t1 | 2500 | 72500 | 68 (0)| 00:00:01 |
| 3| table access full| t2 | 92021 | 988k| 430 (1)| 00:00:01 |
---------------------------------------------------------------------------
predicate information (identified byoperation id):
---------------------------------------------------
1-access("t1"."object_id"="t2"."object_id")
2- filter("t1"."object_name" like '%t1%' and"t1"."object_name" is
not null)
statistics
----------------------------------------------------------
1 recursive calls
0 dbblock gets
1789 consistent gets
0 physical reads
0 redosize
2350 bytes sent via sql*net toclient
573 bytes received via sql*net from client
4 sql*net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
42 rowsprocessed
执行带hint的sql
sql>select /*+ use_nl(t1 t2) index(t2)*/ t1.*,t2.owner from t1,t2 where t1.object_name like '%t1%' and t1.object_id=t2.object_id;
42 rows selected.
execution plan
----------------------------------------------------------
plan hash value: 1022743391
---------------------------------------------------------------------------------------
| id | operation | name | rows | bytes | cost (%cpu)|time |
---------------------------------------------------------------------------------------
| 0| select statement | | 2500 | 97k| 5069 (1)|00:00:01 |
| 1| nested loops | | 2500| 97k| 5069 (1)| 00:00:01 |
| 2| nested loops | | 2500 | 97k| 5069 (1)| 00:00:01 |
|* 3| table access full | t1 | 2500 | 72500 | 68 (0)|00:00:01 |
|* 4| index range scan |t2_idx | 1 | | 1 (0)| 00:00:01 |
| 5| table access by index rowid| t2 | 1 | 11 | 2 (0)|00:00:01 |
---------------------------------------------------------------------------------------
predicate information (identified byoperation id):
---------------------------------------------------
3- filter("t1"."object_name" like '%t1%' and"t1"."object_name" is not
null)
4-access("t1"."object_id"="t2"."object_id")
statistics
----------------------------------------------------------
1 recursive calls
0 dbblock gets
304 consistent gets
24 physical reads
0 redosize
2350 bytes sent via sql*net toclient
573 bytes received via sql*net from client
4 sql*net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
42 rowsprocessed
使用sql profile
查找执行sql的sql_id
tpcc@toaddb> select sql_id,sql_text from v$sqlwhere sql_text like '%t1.object_name%';
sql_id
-------------
sql_text
----------------------------------------------------------------------------------------------------
4zbqykx89yc8v
select t1.*,t2.owner from t1,t2 wheret1.object_name like '%t1%' and t1.object_id=t2.object_id
18bphz37dajq9
select /*+ use_nl(t1 t2) index(t2) */t1.*,t2.owner from t1,t2 where t1.object_name like '%t1%' and
t1.object_id=t2.object_id
运行存储过程如下：
var tuning_task varchar2(100);
declare
l_sql_id v$session.prev_sql_id%type;
l_tuning_task varchar2(30);
begin
l_sql_id:='4zbqykx89yc8v';
l_tuning_task := dbms_sqltune.create_tuning_task(sql_id =>l_sql_id);
:tuning_task:=l_tuning_task;
dbms_sqltune.execute_tuning_task(l_tuning_task);
dbms_output.put_line(l_tuning_task);
end;
/
task_114
pl/sql procedure successfully completed.
查看task的名字
tpcc@toaddb> print tuning_task;
tuning_task
----------------------------------------------------------------------------------------------------
task_114
查看执行报告
set long 99999
col comments format a200
select dbms_sqltune.report_tuning_task(:tuning_task)comments from dual;
comments
----------------------------------------------------------------------------------------------------
general information section
-------------------------------------------------------------------------------
tuning task name : task_114
tuning task owner : tpcc
workload type : single sql statement
scope : comprehensive
time limit(seconds): 1800
completion status : completed
started at : 03/06/2016 05:27:21
completed at : 03/06/2016 05:27:24
-------------------------------------------------------------------------------
schema name: tpcc
sql id : 4zbqykx89yc8v
sql text : select t1.*,t2.owner from t1,t2 where t1.object_name like '%t1%'
and t1.object_id=t2.object_id
-------------------------------------------------------------------------------
findings section (1 finding)
-------------------------------------------------------------------------------
1- sql profile finding (see explain planssection below)
--------------------------------------------------------
apotentially better execution plan was found for this statement.
recommendation (estimated benefit: 83.08%)
------------------------------------------
-consider accepting the recommended sql profile.
executedbms_sqltune.accept_sql_profile(task_name => 'task_114',
task_owner =>'tpcc', replace => true);
validation results
------------------
thesql profile was tested by executing both its plan and the original plan
andmeasuring their respective execution statistics. a plan may have been
only partially executed if the other could be run to completion in lesstime.
original plan with sql profile % improved
------------- ---------------- ----------
completion status: complete complete
elapsed time (s): .012865 .004556 64.58 %
cputime (s): .0124 .0045 63.7%
user i/o time (s): 0 0
buffer gets: 1787 302 83.1%
physical read requests: 0 0
physical write requests: 0 0
physical read bytes: 0 0
physical write bytes: 0 0
rows processed: 42 42
fetches: 42 42
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: 1838229974
---------------------------------------------------------------------------
| id | operation | name | rows | bytes | cost (%cpu)| time |
---------------------------------------------------------------------------
| 0| select statement | | 42 | 1680 | 498 (1)| 00:00:01 |
|* 1| hash join | | 42 | 1680 | 498 (1)| 00:00:01 |
|* 2| table access full| t1 | 42 | 1218 | 68 (0)| 00:00:01 |
| 3| table access full| t2 | 92021 | 988k| 430 (1)| 00:00:01 |
---------------------------------------------------------------------------
predicate information (identified byoperation id):
---------------------------------------------------
1-access("t1"."object_id"="t2"."object_id")
2- filter("t1"."object_name" like '%t1%' and"t1"."object_name" is
not null)
2- using sql profile
--------------------
plan hash value: 1022743391
---------------------------------------------------------------------------------------
| id | operation | name | rows | bytes | cost (%cpu)|time |
---------------------------------------------------------------------------------------
| 0| select statement | | 42 | 1680 | 152 (0)|00:00:01 |
| 1| nested loops | | 42 | 1680 | 152 (0)| 00:00:01 |
| 2| nested loops | | 42 | 1680 | 152 (0)| 00:00:01 |
|* 3| table access full | t1 | 42 | 1218 | 68 (0)| 00:00:01 |
|* 4| index range scan |t2_idx | 1 | | 1 (0)| 00:00:01 |
| 5| table access by index rowid| t2 | 1 | 11 | 2 (0)|00:00:01 |
---------------------------------------------------------------------------------------
predicate information (identified byoperation id):
---------------------------------------------------
3- filter("t1"."object_name" like '%t1%' and"t1"."object_name" is not
null)
4-access("t1"."object_id"="t2"."object_id")
-------------------------------------------------------------------------------
接受分析建议
报告中给出了执行方法，如上红色部分
接受报告的建议，验证一下如下：
tpcc@toaddb> execute dbms_sqltune.accept_sql_profile(task_name=> 'task_114',task_owner => 'tpcc', replace => true);
pl/sql procedure successfully completed.
执行测试
再执行原先命令如下：
tpcc@toaddb> select t1.*,t2.owner from t1,t2 wheret1.object_name like '%t1%' and t1.object_id=t2.object_id;
42 rows selected.
execution plan
----------------------------------------------------------
plan hash value: 1022743391
---------------------------------------------------------------------------------------
| id | operation | name | rows | bytes | cost (%cpu)|time |
---------------------------------------------------------------------------------------
| 0| select statement | | 42 | 1680 | 152 (0)|00:00:01 |
| 1| nested loops | | 42| 1680 | 152 (0)|00:00:01 |
| 2| nested loops | | 42 | 1680 | 152 (0)| 00:00:01 |
|* 3| table access full | t1 | 42 | 1218 | 68 (0)| 00:00:01 |
|* 4| index range scan |t2_idx | 1 | | 1 (0)| 00:00:01 |
| 5| table access by index rowid| t2 | 1 | 11 | 2 (0)|00:00:01 |
---------------------------------------------------------------------------------------
predicate information (identified byoperation id):
---------------------------------------------------
3- filter("t1"."object_name" like '%t1%' and"t1"."object_name" is not
null)
4-access("t1"."object_id"="t2"."object_id")
note
-----
- sql profile"sys_sqlprof_01534b8309b90000" used for this statement
- this is an adaptive plan
statistics
----------------------------------------------------------
35 recursive calls
0 dbblock gets
317 consistent gets
1 physical reads
0 redosize
2350 bytes sent via sql*net toclient
573 bytes received via sql*net from client
4 sql*net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
42 rowsprocessed
启用了profile，ps：如果执行中多加几个空格，并不会影响profile的生效的。
5. 维护操作
禁用命令
如下：
begin
dbms_sqltune.alter_sql_profile(
name => 'sys_sqlprof_01534b8309b90000',
attribute_name => 'status',
value => 'disabled');
end;
/
启用命令
如下：
begin
dbms_sqltune.alter_sql_profile(
name => 'sys_sqlprof_01534b8309b90000',
attribute_name => 'status',
value => 'enabled');
end;
/
查看使用的profile
如下：
sql>select task_name,status fromuser_advisor_tasks ;
删除profile
begin
dbms_sqltune.drop_sql_profile(name => 'sys_sqlprof_01534b8309b90000');
end;
/
---------------------
作者：badman250
来源：csdn
原文：https://blog.csdn.net/notbaron/article/details/50830910
版权声明：本文为博主原创文章，转载请附上博文链接！