I'm having some problems figuring what is the role of a @Clustered MDB inside a JBOSS AS7 cluster using HornetQ.
After all, a lot of message clustering is done on the hornetQ side, meaning that if you send a lot of messages, they will be load balanced along the clustered queue. So if I have the same MDB (without the @Clustered annotation) deployed on each of the cluster nodes, it will receive the same amount of messages.
Or, because there is the possibility of a slow consumer on one of the nodes (because of poor hardware resources) and the MDBs on the other nodes have consumed their messages, they could take some messages form the slow node and this is when the @Clustered annotation becomes handy ?
Also, if I am using a live-backup configuration, and I have the same MDB deployed on both the live and backup server, it is necessary to make it @Clustered ?