First off thank you all for you responses, they really are much appreciated.
@Dhiru Pandey : I will try and get some CPU profiles and jstack dumps, thanks for the suggestion. I've tried tuning the Thread Pool Size but it doesn't really seem to make much of a difference
@Scott Oaks: I will perfectly honest in saying I think that the DB is where the bottleneck is occurring. Every bean in the system extends one one bean, lets call it XXBean. So for instance Person extends XXBean and SomeOtherObject extends XXBean as well. Basically this means that every table in the entire system is joined to a single table which IMHO does not seem like a good idea and I personally have never seen a design like it before.
@km: Logging into the admin console is not the problem, a single request bringing down the rest of the server is. I laughed at your -server setting as that was only discovered a few months ago when we discovered our live site was running with the -client setting set.
You folks are probably rolling your eyes and saying that we shouldn't be allowed to control a wordpress blog and I might not disagree with you! The simple fact is that we are in this situation whether we like it or not, I'm willing to learn and so is the rest of the team.
Thanks again!
Simon
[Message sent by forum member 'scomley' (register_at_comley.org)]
http://forums.java.net/jive/thread.jspa?messageID=385800