A query should be self explanatory. The performance optimization is secondary and only necessarily id there's performance problem.

select Table1.* T1 where exists (select 1 from Table2 T2 where T1.pk_fld = T2.pk_fld and T2.fld2 = '123')

>
>What is a better practice and more efficient when using a derived table in a SQL Select:
>
>

>select Table1.* T1 where exists (select * from Table2 T2 where T1.pk_fld = T2.pk_fld and T2.fld2 = '123')
>

>
>vs
>
>

>select Table1.* T1 where T1.pk_fld in (select pk_fld from Table2 T2 where T1.pk_fld = T2.pk_fld and T2.fld2 = '123')
>

>
>TIA