Have you tried comparing execution plans? Also, what is your SQL Server version?

Do you need a distinct set or all records from both tables? In the latter case, changing UNION to UNION ALL may also help.

>Hi
>I have a view (MyView) which is
>
>

SELECT * FROM DB1.dbo.tableA
>UNION 
>SELECT * FROM DB2.dbo.TableB
>

>
>I have SQL statement
>
>

>SELECT myView.* FROM MyView
>INNER JOIN DB2.dbo.TableZ ON myView.Key = TableZ.Key
>

>
>It is working fine, returned result in 2 secs
>
>However, if I amend it as below, it will take more than a minute to execute.
>
>

>SELECT myView.*, TableZ.FieldA FROM MyView
>INNER JOIN DB2.dbo.TableZ ON myView.Key = TableZ.Key
>

>
>After some trial and error, found that, if I remark the edit my view without UNION (select only from TableB), the performance is 2 secs as well. Why?
>
>Beside, I tried to add an index which contains Key + FieldA, but not helping at all.
>
>Please advice, why remove the UNION can be helpful? How to solve this matter with "right" way?
>
>Thank you