We're having some SQL bottlenecks, I guess it's due to transaction isolation.
We're gonna try debugging step-by-step our offending code and checking the database profiler to detect which resources are being blocked at a given time.
How would you diagnostic SQL bottlenecks on Java EE?
I'd use a quite classical method: measure the time used by each query and write in a debug file the timings, followed by the query. Then, run the application a few times, in normal usage conditions, and see the queries that take more time.
About timing in Java: How do I time a method's execution in Java?
Also notice that some SQL queries do take more time than others because of the quantity of data: for example, "SELECT" statements are common bottlenecks. A SELECT * FROM userdata statement takes much more time than a SELECT name, surname FROM userdata one, especially when you have lots of stored data.
