Can anyone tell me if there are any peform gain between creating a cursor that then a tuple will consult in another table or create a query with a join between two tables would be more performative?
For example:
As a general rule, the most performative would be Join himself.
But you have to see how tables are created, indexes, quantity of data in tables and so on.
I finally performed a test and the PL / SQL code with cursor was 20x faster than a Join SQL in Oracle.
But before that in this same appointment I created an index where the course fell from 19000 (5hs) to 500 (5hs), I think I created a good index, but even then after changes the SQL code for a cursor In PL / SQL time dropped to 9mins.
Can there be any explanation for this? For the final result, I wanted to inform you that the result was 189,923 tuples