Sql server not in优化

使用EXISTS(或NOT EXISTS)通常将提高查询的效率，由于NOT IN子句将对子查询中的表执行了一个全表遍历。

oracle在执行IN子查询过程中，先执行子查询结果放入临时表再进行主查询；

 

而exists先运行主查询，再执行子查询查到第一个匹配项

如：查询 A表中没有和B表或C表相连的数据 

 

select A.id,A.reads,A.addtime 

from A 

where A.person_id not in (select user_id from B ) 

or A.worksite_id not in (select id from C) 

order by A.addtime desc

程序执行时间：40109.38毫秒 

 

select A.id,A.reads,A.addtime

from A 

where not exists (select id FROM B where B.user_id=A.person_id) 

or not exists (select id FROM C where C.id=A.worksite_id) 

order by Sendorder.addtime desc

程序执行时间：8531.25毫秒 

ms ssql 实例：

select 
subscriber_id
from 
NewsLetterSystem_CampaignCategorySubscriber ccs WITH ( NOLOCK )
where NOT EXISTS(
SELECT  distinct subscriber_id
FROM    NewsLetterSystem_OpenTracking ot WITH ( NOLOCK ) 
WHERE  ot.subscriber_id=ccs.subscriber_id and ot.campaign_id = 52801)
and ccs.campaign_id = 52801 and ccs.status_id = 1