ORACLE SPA（SQL Performance Analyzer）讲解

Posted 2021-04-26 Oracle一体机用户组

作者简介

孙显鹏，Oracle十年从业经验，拥有11G ocp认证，精通内部原理，擅长调优，解决疑难问题，致力于帮助客户解决生产过程过出现的性能问题，提高生产效率！爱好书法！

前言：

迁移数据库，修改数据库参数或者升级数据库后，如何确定原库的SQL在目标库执行时没有出现性能衰退呢？使用ORACLE的SPA（SQL Performance Analyzer）包可完成该功能。SPA通过比较原库和目标库收集的SQL集合的执行统计数据，最终给我们一个比较结果，通过分析对比结果数据可明确哪些SQL出现性能衰退。查找衰退原因修复问题，确保SQL在迁移后能满足业务需求。这样的迁移才是完整的。接下来演示如何使用SPA测试SQL的执行情况。

场景：

原库为11.2.0.3，目标库为12.2.0.1，把目标库当做要升级的数据库（通过xtts迁移了两个表空间），在目标库中建立spa-sts，首先通过dblink在原库执行sql trial，然后在目标库执行SQL trial，最后比较两个SQL trial。这样就达到了对比升级后系统是否可以符合业务实际要求。

操作步骤：

准备测试SQL：

delete from yaclaim.c_pay where rownum<100;
delete from yaclaim.c_pay where rownum<10;
select * from yaclaim.c_pay where rownum<100;
select count(*) from yaclaim.c_pay where CLAIMTIME>to_date('2018-03-20','yyyy-mm-dd');
select count(*) from yaclaim.c_pay where CLAIMTIME<to_date('2018-03-20','yyyy-mm-dd');

create sts ：

sqlplus / as sysdba 
exec dbms_sqltune.create_sqlset(sqlset_name => 'test_set', description =>'11g workload upgrade to 12c'); 
declare 
mycur dbms_sqltune.sqlset_cursor; 
begin 
open mycur for 
select value (P) 
from table(dbms_sqltune.select_cursor_cache('sql_text like ''%c_pay%''', null, null, null, null,1, null, 'ALL')
) P;
dbms_sqltune.load_sqlset(sqlset_name => 'test_set', populate_cursor => mycur); 
end; 
/

如果获取特定schema信息如下：

–dbms_sqltune.select_cursor_cache(‘parsing_schema_name<> ”YACLAIM”’, null, null, null, null,1, null, ‘ALL’)

select sql_text from dba_sqlset_statements where sqlset_name='test_set';

Create a task to run Sql Performance Analyzer：

create_sqlpa_task.sql 
---create sql task 
-- declare vars 
var tname varchar2(30); 
var sname varchar2(30); 
-- init vars 
exec :sname := 'test_set'; 
exec :tname := 'my_sqlpa_demo_task'; 

— 1.create a task with a purpose of change impact analysis

exec :tname := dbms_sqlpa.create_analysis_task(sqlset_name => :sname, task_name => :tname);

— 2.check task status

SELECT task_name, status 
FROM user_advisor_tasks 
WHERE task_name = :tname;

Execute Before Change TEST EXECUTE (Pre-Change SQL Trial)：

beforechange.sql

 

–Now I am ready to run the Before Change Execute

begin
DBMS_SQLPA.EXECUTE_ANALYSIS_TASK(
task_name => 'my_sqlpa_demo_task',
execution_type => 'TEST EXECUTE',
execution_name => 'BEFORE',
execution_params => dbms_advisor.arglist('DATABASE_LINK', 'TTSLINK'));
end;
/

Execute After Change TEST EXECUTE (Post-Change SQL Trial)：

begin 
DBMS_SQLPA.EXECUTE_ANALYSIS_TASK( 
task_name => 'my_sqlpa_demo_task', 
execution_type => 'TEST EXECUTE', 
execution_name => 'AFTER'); 
end; 
/

Comparing SQL Trials：

–Now we need to compare the 2 executions Before Change and After Change.

 

–I am selecting the comparison matrix as BUFFER_GETS.

 

–compare_runs.sql

begin 
DBMS_SQLPA.EXECUTE_ANALYSIS_TASK( 
task_name => 'my_sqlpa_demo_task', 
execution_type => 'COMPARE PERFORMANCE', 
execution_name => 'DEMOTASK03', 
execution_params => dbms_advisor.arglist( 
'comparison_metric', 
'buffer_gets',
'optimizer_cost',
'physical_i/o')); 
end; 
/

Generate compare report：

–Now I will generate a compare report

 

–The report format can be TEXT,html,OR XML

 

–report.sql

set long 100000 longchunksize 100000 linesize 200 head off feedback off echo off 
spool report.html 
SELECT dbms_sqlpa.report_analysis_task('my_sqlpa_demo_task', 'HTML', 'ALL','ALL') FROM dual; 
spool off

analyze report：

这里贴出部分报告

原创文章，版权归本文作者所有，如需转载请注明出处

喜欢本文请长按下方的二维码订阅Oracle一体机用户组