I agree that it is half-tuned. However, If I am not mistaken OpenJPA strives for "superior performance out of the box" (http://www.mail-archive.com/[email protected]/msg02064.html). Anyway, a similar comparison was done before, with somewhat disappointing results for OpenJPA which caused a lot of discussion (http://www.theserverside.com/news/thread.tss?thread_id=53142).
May be, instead of spending time on these individual tests, it is a good idea to set up some performance tests for some typical CRUD scenarios in order to test whether the performance is still in line with perception with each new release (eg. x% overhead compared to direct JDBC calls?). Needless to say that small changes can sometimes have a unexpected impact on performance. Moreover, publishing the results on the website like Eclipse does, could also show the areas of improvement regarding performance next to the other improvements. regards, Christiaan -- View this message in context: http://openjpa.208410.n2.nabble.com/Performance-comparison-tp5321131p5341570.html Sent from the OpenJPA Users mailing list archive at Nabble.com.
