Oracle 中定位重要(消耗资源多)的SQL

为什么80%的码农都做不了架构师？>>>   

在分析SQL性能的时候，经常需要确定资源消耗多的SQL，总结如下：

1 查看值得怀疑的SQL
select substr(to_char(s.pct,'99.00'),2)||'%'load,
 
       s.executions executes,
 
       p.sql_text
 
from(select address,
 
            disk_reads,
 
            executions,
 
            pct,
 
            rank() over(order by disk_reads desc) ranking
 
         from(select address,
 
                     disk_reads,
 
                     executions,
 
                     100*ratio_to_report(disk_reads) over() pct
 
                 from sys.v_$sql
 
                where command_type!=47)
 
        where disk_reads>50*executions) s,
 
       sys.v_$sqltext p
 
where s.ranking<=5
 
  and p.address=s.address
 
order by 1, s.address, p.piece;
 
2 查看消耗内存多的sql
 
select b.username ,a.buffer_gets ,a.executions,
 
       a.disk_reads/decode(a.executions,0,1,a.executions),a.sql_text SQL
 
from v$sqlarea a,dba_users b
 
where a.parsing_user_id = b.user_id
 
 and a.disk_reads >10000
 
order by disk_reads desc;
 
3 查看逻辑读多的SQL
select*
 
from(select buffer_gets, sql_text
 
     from v$sqlarea
 
     where buffer_gets>500000
 
     order by buffer_gets desc)
 
where rownum<=30;
 
4 查看执行次数多的SQL
 
select sql_text, executions
 
from(select sql_text, executions from v$sqlarea order by executions desc)
 
where rownum<81;
 
5 查看读硬盘多的SQL
 
select sql_text, disk_reads
 
from(select sql_text, disk_reads from v$sqlarea order by disk_reads desc)
 
where rownum<21;
 
6 查看排序多的SQL
 
select sql_text, sorts
 
from(select sql_text, sorts from v$sqlarea order by sorts desc)
 
where rownum<21;
 
7 分析的次数太多，执行的次数太少，要用绑变量的方法来写sql
 
set pagesize 600;
 
set linesize 120;
 
select substr(sql_text,1,80) "sql",count(*),sum(executions) "totexecs"
 
from v$sqlarea

where executions<5
 
group by substr(sql_text,1,80)
 
having count(*)>30
 
order by 2;
 
8 游标的观察
set pages 300;
 
select sum(a.value), b.name
 
from v$sesstat a, v$statname b
 
where a.statistic#=b.statistic#
 
and b.name='opened cursors current'
 
group by b.name;
 
 
 
select count(0) from v$open_cursor;
 
 
 
select user_name, sql_text,count(0)
 
from v$open_cursor
 
group by user_name, sql_text
 
having count(0)>30;
 
9 查看当前用户&username执行的SQL
select sql_text
 
from v$sqltext_with_newlines
 
where(hash_value, address) in
 
     (select sql_hash_value, sql_address
 
      from v$session
 
      where username='&username')
 
order by address, piece;

本篇文章来源于 Linux公社网站(www.linuxidc.com)  原文链接：http://www.linuxidc.com/Linux/2011-04/34972.htm

转载于:https://my.oschina.net/oaoa/blog/228708