We are evaluating Hibernate to use it in our J2EE environment.

Our application should be capable of running in any J2EE compliant application server and against any RDBMS in the market (atleast the ones which hibernate claims to support).

My current understanding is (apart from other advantages of Hibernate) the main difference is by default entity beans support clustering and hibernate does not. Please let me know whether my understanding is correct.

Please let me know whether Hibernate supports and works well in the clustering environment of weblogic, websphere, jboss etc.

Thanks for your response,
goHibernate

you can do clustering with hibernate.

We use loadbalancing with sticky session session, but there are ways to use failover too. Actually it's not really an hibernate problem, you have to be sure your httpSession is serialisable...

About second level cache, JBossTreeCache is able to do the job in a clustered environment.

I think there is a lot of nonsense written about the supposed "clusterability" of entity beans.

Really, clusterability is a feature of the remote invokation layer, ie. of the session bean facade, or servlet. Not of the entity bean, which is in practice called locally.

And most existing EJB containers have at most a very crude second-level clustered cache implementation - Hibernate is much more sophisticated in this respect.

What do you think your CMp container is doing for you in terms of clustering?