In the case that your computation is massive, you may want to change your concurrency design. You can add a writer thread to be the only one who talks to database. Other threads write their manipulated data into a mutex-guarded shared data structure (say a queue). Then the writer thread can write data into database and free up the queue within time period, or triggered by an external event.
This will improve performance only if the bottle-neck of time is in data manipulation part of the application. Otherwise you may got no improvements, but a considerable overhead (The time spent waiting behind locks). Cheers
_______________________________________________ Interest mailing list Interest@qt-project.org http://lists.qt-project.org/mailman/listinfo/interest
