相关疑难解决方法(0)

(正在使用SQL Server 2012)

我找到了一些关于查询优化的主题,并将EXISTS与COUNT进行比较,但我找不到这个确切的问题.

我有一个看起来像这样的查询:

select * from
tblAccount as acc
join tblUser as user on acc.AccountId = user.AccountId
join tblAddress as addr on acc.AccountId = addr.AccountId
... **a few more joins**
where acc.AccountId in (
    select * accountid from
    (select accountid, count(*) from tblUser
    where flag = 1
    group by accountId) as tbl where c != 1

此查询立即运行(尽管db非常大,大约70Gb).

当我将查询包装在EXISTS中时,如下所示:

if exists
(
  **Exact same query as above**
)
begin
RAISERROR('Account found without exactly one flagged user.', 16, 1);
end …