Hi Steve, thank you for taking the time to look at this. The test checks the amount of entries after inserting them. So the amount of entries is the same - with or without the change in SequenceStyleGenerator.java. But with the change it is much faster. Just leaving out ids cannot increase performance, right? Leaving out records can increase performance but that is not the case, is it?
I want the test to write 50 records at once (bulk insert) using sequences with only one round trip to the database and another round trip to get the sequence ids. I did not see any test in hibernate checking this. For hsqldb at least I know how to trace database accesses. Unfortunately I am not familiar with the used h2 database.
I think you are right that I do not fully get the point.

At least what I expect is that for each 50 records two round trips to the database should be enough using sequence. Am I wrong?
Thank you so far!