oracle一条查询语句慢会影响整个数据库的效率吗

Posted 2023-05-07

参考技术A oracle一条查询语句慢会影响整个数据库的效率吗
如果你的服务器性能已经负载很高了,这时在出现一条特别慢的sql语句,占用了cpu,内存,IO资源等,有可能会产生严重的后果。

常用sql 分页语句(Oracle)

常用的Oracle查询语句

1.无ORDER BY排序的写法。(效率最高)

经过测试，此方法成本最低，只嵌套一层，速度最快！即使查询的数据量再大，也几乎不受影响，速度依然！

sql语句如下：

SELECT *
        FROM (Select ROWNUM AS ROWNO, T.*
       from k_task T
      where Flight_date between to_date(‘20060501‘, ‘yyyymmdd‘) and
       to_date(‘20060731‘, ‘yyyymmdd‘)
       AND ROWNUM <= 20) TABLE_ALIAS
WHERE TABLE_ALIAS.ROWNO >= 10;

2.有ORDER BY排序的写法。(效率最高)

经过测试，此方法随着查询范围的扩大，速度也会越来越慢！ 

SELECT *
       FROM (SELECT TT.*, ROWNUM AS ROWNO
         FROM (Select t.*
         from k_task T
        where flight_date between to_date(‘20060501‘, ‘yyyymmdd‘) and
       to_date(‘20060531‘, ‘yyyymmdd‘)
      ORDER BY FACT_UP_TIME, flight_no) TT
WHERE ROWNUM <= 20) TABLE_ALIAS
where TABLE_ALIAS.rowno >= 10;

3.无ORDER BY排序的写法。(建议使用方法1代替)
此方法随着查询数据量的扩张，速度会越来越慢！  

    

SELECT *
      FROM (Select ROWNUM AS ROWNO, T.*
      from k_task T
       where Flight_date between to_date(‘20060501‘, ‘yyyymmdd‘) and
          to_date(‘20060731‘, ‘yyyymmdd‘)) TABLE_ALIAS
         WHERE TABLE_ALIAS.ROWNO <= 20
          AND TABLE_ALIAS.ROWNO >= 10;
TABLE_ALIAS.ROWNO between 10 and 100;

4.有ORDER BY排序的写法.(建议使用方法2代替)

此方法随着查询范围的扩大，速度也会越来越慢！      

 

SELECT *
FROM (SELECT TT.*, ROWNUM AS ROWNO
FROM (Select *
from k_task T
where flight_date between to_date(‘20060501‘, ‘yyyymmdd‘) and
to_date(‘20060531‘, ‘yyyymmdd‘)
ORDER BY FACT_UP_TIME, flight_no) TT) TABLE_ALIAS
where TABLE_ALIAS.rowno BETWEEN 10 AND 20;

5.另类语法。(有ORDER BY写法）

该语法风格与传统的SQL语法不同，不方便阅读与理解，为规范与统一标准，不推荐使用。此处贴出代码供大家参考之用。

 

With partdata as(
SELECT ROWNUM AS ROWNO, TT.* FROM (Select *
from k_task T
where flight_date between to_date(‘20060501‘, ‘yyyymmdd‘) and
to_date(‘20060531‘, ‘yyyymmdd‘)
ORDER BY FACT_UP_TIME, flight_no) TT
WHERE ROWNUM <= 20)
Select * from partdata where rowno >= 10;

6.另类语法 。(无ORDER BY写法）

With partdata as(
Select ROWNUM AS ROWNO, T.*
From K_task T
where Flight_date between to_date(‘20060501‘, ‘yyyymmdd‘) and
To_date(‘20060531‘, ‘yyyymmdd‘)
AND ROWNUM <= 20)
Select * from partdata where Rowno >= 10;