SQL & PL/SQL

How does oracle 10g execute a correlated subquery?

I read some articles online & i am a little confused.

example:

select * from emp e
where e.deptno in (select d.deptno from dept d
where e.deptno = d.deptno);

My questions .......

1.In the above example, does oracle read the entire outer table first and then run the inner query using the rows returned by the outer query?

I read in some articles that they execute simultaneously.

How does this work?

2.Should the inner query have lesser amount of rows compared to the outer query for a good performance?

3.Can every correlated subquery be converted to a join and if so which one to use?


Truly appreciate any inputs on how oracle executes it at the backend.

Thanks in advance.