Heej,
I'm working on a project where I should persist a lot of POJO files in to an Oracle database. The problem is that for each POJO Hibernate creates a new insert JDBC request. When persisting over 45.000.000 objects, this will take a really long time. I have to save data to a database and that data will be spread over 15 different tables. So for each set of data, I have to do 15 different insert statements. When I have to persist 3.000.000 data sets, 45.000.000 insert statements are needed.
When I was discussing this issue with a colleague, he suggested using CSV files instead of inserting each POJO with an insert statement. Now I'm wondering, is there an easy way of letting Hibernate write for each table a CSV file?
Or maybe someone does know a better solution?