小知识：IN和EXISTS的用法及效率验证
环境： Oracle 19.16 多租户架构
经常会在网上看到有人写exists和in的效率区别，其实在新版本的数据库中，是不存在这个问题的，优化器会自己判断选择最优的执行计划。
为了直观的说明，我在PDB中构造如下测试用例：
vi 1.sql
select count(*) from v$active_session_history;
select count(*) from dba_hist_active_sess_history;
create table T1 as select * from v$active_session_history;
create table T2 as select * from dba_hist_active_sess_history;
构造小表T1，大表T2。
SQL> set timing on
SQL> @1
COUNT(*)
----------
383
Elapsed: 00:00:00.05
COUNT(*)
----------
215636
Elapsed: 00:00:00.95
Table created.
Elapsed: 00:00:00.20
Table created.
Elapsed: 00:00:07.90
网上说，当T1数据量小，而T2数据量非常大时，使用exists的查询效率会高。
验证下，是否事实真是如此？
select /*+ monitor */ * from T1 where exists(select 1 from T2 where T1.sql_id = T2.sql_id) ;
select /*+ monitor */ * from T1 where T1.sql_id in (select T2.sql_id from T2) ;
SQL> select sql_id, sql_text from v$sql where sql_text like '%T2.sql_id%'
SQL_ID SQL_TEXT
------------- ------------------------------------------------------------------------------------------
4xu586p9h0qcq select /*+ monitor */ * from T1 where T1.sql_id in (select T2.sql_id from T2)
3qgrm97t5jgwj select /*+ monitor */ * from T1 where exists(select 1 from T2 where T1.sql_id = T2.sql_id)
使用sqlmon取到两个SQL对应的SQL Monitor Report，对比分析发现：
二者执行计划完全一样，对应Plan Hash Value 1713220790，都走的是Hash Join Semi，执行时间也没差别。
所以这个说法最起码在Oracle 19c的版本中是不存在的，你想怎么写都OK，优化器会帮你做查询转换。
为了进一步验证，构造4个典型SQL，分别使用in和exists的写法：
--SQL1:
select /*+ monitor */ SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID, count(*)
from T1
where T1.sql_id in (select T2.sql_id from T2)
group by SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID
order by 1;
--SQL2:
select /*+ monitor */ SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID, count(*)
from T1
where exists (select 1 from T2 where T2.sql_id = T1.sql_id)
group by SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID
order by 1;
--SQL3:
select /*+ monitor */ SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID, count(*)
from T2
where T2.sql_id in (select T1.sql_id from T1)
group by SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID
order by 1;
--SQL4:
select /*+ monitor */ SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID, count(*)
from T2
where exists (select 1 from T1 where T1.sql_id = T2.sql_id)
group by SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID
order by 1;
SQL Monitor的截图就不贴了，直接给大家看下文本格式的执行计划，方便对比和检索：
SQL1:
SQL> select /*+ monitor */ SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID, count(*)
2 from T1
3 where T1.sql_id in (select T2.sql_id from T2)
4 group by SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID
5 order by 1;
SQL_ID SQL_PLAN_HASH_VALUE SQL_PLAN_LINE_ID COUNT(*)
------------- ------------------- ---------------- ----------
3dbzmtf9ahvzt 3238164414 1 1
3kqrku32p6sfn 2977818336 14 1
3zbvwad7h1pgt 2360206614 1 2
3zbvwad7h1pgt 2360206614 6
87gaftwrm2h68 0 1
9wncfacx0nj9h 0 2
9wncfacx0nj9h 3312548573 9
avf5k3k0x0cxn 3746835944 1 1
b13g21mgg8y98 212733457 9 1
b13g21mgg8y98 212733457 12 2
ggh55rhz95kyj 3124993369 8
gug127tbfzjcs 3645025857 0 1
12 rows selected.
Elapsed: 00:00:00.07
SQL> @xplan
PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
SQL_ID 250w6cua1mfa0, child number 2
-------------------------------------
select /*+ monitor */ SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID,
count(*) from T1 where T1.sql_id in (select T2.sql_id from T2) group by
SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID order by 1
Plan hash value: 910330555
-----------------------------------------------------------------------------------------------------------------
| Id | Operation | Name | Starts | E-Rows | A-Rows | A-Time | Buffers | OMem | 1Mem | Used-Mem |
-----------------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | | 12 |00:00:00.07 | 16132 | | | |
| 1 | SORT GROUP BY | | 1 | 228 | 12 |00:00:00.07 | 16132 | 2048 | 2048 | 2048 (0)|
|* 2 | HASH JOIN SEMI | | 1 | 228 | 35 |00:00:00.07 | 16132 | 1376K| 1376K| 1604K (0)|
|* 3 | TABLE ACCESS FULL| T1 | 1 | 228 | 228 |00:00:00.01 | 26 | | | |
|* 4 | TABLE ACCESS FULL| T2 | 1 | 177K| 177K|00:00:00.06 | 16106 | | | |
-----------------------------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - access("T1"."SQL_ID"="T2"."SQL_ID")
3 - filter("T1"."SQL_ID" IS NOT NULL)
4 - filter("T2"."SQL_ID" IS NOT NULL)
25 rows selected.
Elapsed: 00:00:00.04
SQL2:
SQL> select /*+ monitor */ SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID, count(*)
2 from T1
3 where exists (select 1 from T2 where T2.sql_id = T1.sql_id)
4 group by SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID
5 order by 1;
SQL_ID SQL_PLAN_HASH_VALUE SQL_PLAN_LINE_ID COUNT(*)
------------- ------------------- ---------------- ----------
3dbzmtf9ahvzt 3238164414 1 1
3kqrku32p6sfn 2977818336 14 1
3zbvwad7h1pgt 2360206614 1 2
3zbvwad7h1pgt 2360206614 6
87gaftwrm2h68 0 1
9wncfacx0nj9h 0 2
9wncfacx0nj9h 3312548573 9
avf5k3k0x0cxn 3746835944 1 1
b13g21mgg8y98 212733457 9 1
b13g21mgg8y98 212733457 12 2
ggh55rhz95kyj 3124993369 8
gug127tbfzjcs 3645025857 0 1
12 rows selected.
Elapsed: 00:00:00.06
SQL> @xplan
PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
SQL_ID cxn8artthq7p8, child number 0
-------------------------------------
select /*+ monitor */ SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID,
count(*) from T1 where exists (select 1 from T2 where T2.sql_id =
T1.sql_id) group by SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID order
by 1
Plan hash value: 910330555
-----------------------------------------------------------------------------------------------------------------
| Id | Operation | Name | Starts | E-Rows | A-Rows | A-Time | Buffers | OMem | 1Mem | Used-Mem |
-----------------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | | 12 |00:00:00.06 | 16132 | | | |
| 1 | SORT GROUP BY | | 1 | 228 | 12 |00:00:00.06 | 16132 | 2048 | 2048 | 2048 (0)|
|* 2 | HASH JOIN SEMI | | 1 | 228 | 35 |00:00:00.06 | 16132 | 1376K| 1376K| 1611K (0)|
|* 3 | TABLE ACCESS FULL| T1 | 1 | 228 | 228 |00:00:00.01 | 26 | | | |
|* 4 | TABLE ACCESS FULL| T2 | 1 | 177K| 177K|00:00:00.05 | 16106 | | | |
-----------------------------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - access("T2"."SQL_ID"="T1"."SQL_ID")
3 - filter("T1"."SQL_ID" IS NOT NULL)
4 - filter("T2"."SQL_ID" IS NOT NULL)
26 rows selected.
Elapsed: 00:00:00.03
SQL3:
SQL> select /*+ monitor */ SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID, count(*)
2 from T2
3 where T2.sql_id in (select T1.sql_id from T1)
4 group by SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID
5 order by 1;
SQL_ID SQL_PLAN_HASH_VALUE SQL_PLAN_LINE_ID COUNT(*)
------------- ------------------- ---------------- ----------
3dbzmtf9ahvzt 3238164414 1 1
3kqrku32p6sfn 1774581179 20 2
3kqrku32p6sfn 1774581179 23 2
3kqrku32p6sfn 2977818336 14 2
3zbvwad7h1pgt 2360206614 1
87gaftwrm2h68 1072382624 2 2
9wncfacx0nj9h 3312548573 2
avf5k3k0x0cxn 3746835944 1 1
b13g21mgg8y98 212733457 9 1
b13g21mgg8y98 2612542848 1 2
ggh55rhz95kyj 3124993369 4
gug127tbfzjcs 3645025857 1
12 rows selected.
Elapsed: 00:00:00.09
SQL> @xplan
PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
SQL_ID 1588n6cc48yv0, child number 0
-------------------------------------
select /*+ monitor */ SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID,
count(*) from T2 where T2.sql_id in (select T1.sql_id from T1) group by
SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID order by 1
Plan hash value: 3152222881
-------------------------------------------------------------------------------------------------------------------
| Id | Operation | Name | Starts | E-Rows | A-Rows | A-Time | Buffers | OMem | 1Mem | Used-Mem |
-------------------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | | 12 |00:00:00.08 | 16132 | | | |
| 1 | SORT GROUP BY | | 1 | 3684 | 12 |00:00:00.08 | 16132 | 2048 | 2048 | 2048 (0)|
|* 2 | HASH JOIN RIGHT SEMI| | 1 | 3684 | 21 |00:00:00.08 | 16132 | 1651K| 1651K| 1520K (0)|
|* 3 | TABLE ACCESS FULL | T1 | 1 | 228 | 228 |00:00:00.01 | 26 | | | |
|* 4 | TABLE ACCESS FULL | T2 | 1 | 177K| 177K|00:00:00.08 | 16106 | | | |
-------------------------------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - access("T2"."SQL_ID"="T1"."SQL_ID")
3 - filter("T1"."SQL_ID" IS NOT NULL)
4 - filter("T2"."SQL_ID" IS NOT NULL)
25 rows selected.
Elapsed: 00:00:00.03
SQL4:
SQL> select /*+ monitor */ SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID, count(*)
2 from T2
3 where exists (select 1 from T1 where T1.sql_id = T2.sql_id)
4 group by SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID
5 order by 1;
SQL_ID SQL_PLAN_HASH_VALUE SQL_PLAN_LINE_ID COUNT(*)
------------- ------------------- ---------------- ----------
3dbzmtf9ahvzt 3238164414 1 1
3kqrku32p6sfn 1774581179 20 2
3kqrku32p6sfn 1774581179 23 2
3kqrku32p6sfn 2977818336 14 2
3zbvwad7h1pgt 2360206614 1
87gaftwrm2h68 1072382624 2 2
9wncfacx0nj9h 3312548573 2
avf5k3k0x0cxn 3746835944 1 1
b13g21mgg8y98 212733457 9 1
b13g21mgg8y98 2612542848 1 2
ggh55rhz95kyj 3124993369 4
gug127tbfzjcs 3645025857 1
12 rows selected.
Elapsed: 00:00:00.09
SQL> @xplan
PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
SQL_ID 99fkm9p94agcf, child number 0
-------------------------------------
select /*+ monitor */ SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID,
count(*) from T2 where exists (select 1 from T1 where T1.sql_id =
T2.sql_id) group by SQL_ID, SQL_PLAN_HASH_VALUE, SQL_PLAN_LINE_ID order
by 1
Plan hash value: 3152222881
-------------------------------------------------------------------------------------------------------------------
| Id | Operation | Name | Starts | E-Rows | A-Rows | A-Time | Buffers | OMem | 1Mem | Used-Mem |
-------------------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | | 12 |00:00:00.09 | 16132 | | | |
| 1 | SORT GROUP BY | | 1 | 3684 | 12 |00:00:00.09 | 16132 | 2048 | 2048 | 2048 (0)|
|* 2 | HASH JOIN RIGHT SEMI| | 1 | 3684 | 21 |00:00:00.09 | 16132 | 1651K| 1651K| 1520K (0)|
|* 3 | TABLE ACCESS FULL | T1 | 1 | 228 | 228 |00:00:00.01 | 26 | | | |
|* 4 | TABLE ACCESS FULL | T2 | 1 | 177K| 177K|00:00:00.08 | 16106 | | | |
-------------------------------------------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - access("T1"."SQL_ID"="T2"."SQL_ID")
3 - filter("T1"."SQL_ID" IS NOT NULL)
4 - filter("T2"."SQL_ID" IS NOT NULL)
26 rows selected.
Elapsed: 00:00:00.03
可以看到对比前2个SQL：
250w6cua1mfa0
cxn8artthq7p8
执行计划一样，都是HASH JOIN SEMI。
对比后两个SQL：
1588n6cc48yv0
99fkm9p94agcf
执行计划也一样，都是HASH JOIN RIGHT SEMI。
所以，在新版本的数据库中，确实是不用再关注这个问题，优化器会帮助我们做好最优的查询转换。
小知识：IN和EXISTS的用法及效率验证的更多相关文章
- WPF小知识，MessageBox的多种用法
我们在程序中经常会用到MessageBox. 现将其常见用法总结如下: 1.MessageBox.Show("Hello~~~~"); 最简单的,只显示提示信息. 2.Messag ...
- 12个你未必知道的CSS小知识
虽然CSS并不是一种很复杂的技术,但就算你是一个使用CSS多年的高手,仍然会有很多CSS用法/属性/属性值你从来没使用过,甚至从来没听说过. 1.CSS的color属性并非只能用于文本显示 对于CSS ...
- python小技巧 小知识
python小技巧 小知识 python系统变量(修改调用shell命令路径)或用户空间说明 20150418 python调用系统命令,报找不到.怎么办? 类似执行shell的: [ -f /etc ...
- Mysql中EXISTS关键字用法、总结
在做教务系统的时候,一个学生(alumni_info)有多个教育经历(alumni_education),使用的数据库是mysql,之前使用左链接查询的,发现数据量才只有几万条时,查询就很慢了,早上想 ...
- Linux小知识：rm -rf/*会将系统全部删除吗
Linux小知识:rm -rf/*会将系统全部删除吗 本文是学习笔记,视频地址为:https://www.bilibili.com/video/av62839850 执行上面的命令并不会删除所有内容( ...
- Oracle-where exists()、not exists() 、in()、not in()用法以及效率差异
0.exists() 用法: select * from T1 where exists(select 1 from T2 where T1.a=T2.a) 其中 "select 1 fro ...
- 蓝牙Bluetooth技术小知识
蓝牙Bluetooth技术以及广泛的应用于各种设备,并将继续在物联网IoT领域担任重要角色.下面搜集整理了一些关于蓝牙技术的小知识,以备参考. 蓝牙Bluetooth技术始创于1994年,其名字来源于 ...
- HTML+CSS中的一些小知识
今天分享一些HTML.CSS的小知识,希望能够对大家有所帮助! 1.解决网页乱码的问题:最重要的是要保证各个环节的字符编码一致! (1)编辑器的编辑环境的字符集(默认字符集):Crtl+U 常见的编码 ...
- iOS APP开发的小知识(分享）
亿合科技小编发现从2007年第一款智能手机横空出世,由此开启了人们的移动智能时代.我们从一开始对APP的陌生,到现在的爱不释手,可见APP开发的出现对我们的生活改变有多巨大.而iOS AP ...
- SQL中EXISTS的用法
比如在Northwind数据库中有一个查询为SELECT c.CustomerId,CompanyName FROM Customers cWHERE EXISTS(SELECT OrderID FR ...
随机推荐
- 一切混乱开端的透明效果——ShaderCp8
--20.8.18 Unity中,通常用两种方法来实现透明效果 1)透明度测试 2)透明度混合 这两个分别是什么呢 1.透明度测试指的只要不符合条件(即在物体颜色中的alpha通道的值小于某一个阈值) ...
- git常用命令查询手册
默认已经连接到远程仓库的情况下 本地文件夹初始化成git仓库.提交本地仓库并添加注释 git init git add 文件1(文件夹1) 文件2(文件夹2)... git commit -m &qu ...
- 19 request请求数据大全
1 request.path_info # 获取当前用户请求的url # http://127.0.0.1:8000/customer/list/ ----> /customer/list/ # ...
- vue后台管理系统——项目优化上线
1. 项目优化 项目优化策略: 使用进度条nprogress 生成打包报告 第三方库启用 CDN Element-UI 组件按需加载 路由懒加载 首页内容定制 ① 使用进度条nprogress 在页面 ...
- Mybatis框架中 collection 标签 和 association标签中关于 columnPrefix 属性的底层逻辑
columnPrefix的作用是给column自动拼接上前缀, 已知多重嵌套的collection 和 association的columnPrefix属性的值是会叠加的 <associatio ...
- Mybatis配置报错：Failed to configure a DataSource: 'url' attribute is not specified and no embe...
问题分析及解决方案 问题原因: Mybatis没有找到合适的加载类,其实是大部分spring - datasource - url没有加载成功,分析原因如下所示. DataSourceAutoConf ...
- 【python】第一模块 步骤四 第二课、实现飞机大战（未完待续）
第二课.实现飞机大战 一.项目介绍 项目实战:飞机大战 课程目标 掌握面向对象分析和开发的思想 能对项目进行拆分,进行模块化开发 了解项目开发的基本流程 理解并运用python的包.模块相关知识 理解 ...
- 对css及css3的大致整理（查漏补缺）
- 代理模式_v1
代理模式 概念: 1.真实对象:要被代理的对象 2.代理对象 3.代理模式 : 代理对象代理真实对象,达到增强真实对象功能的作用 实现方式: 1.静态代理:有一个类文件描述代理模式 2.动态代理:在内 ...
- 加热算法，加热温度控制加热功率，加热功率控制加热速度(PWM)
uint8_t user_heating_algorithmPID(void) { uint32_t temp_1; uint16_t Adcn; nrfx_err_t err_code; HEATI ...