What I'd like to do is to replicate data nodes within a cluster not only for backup/failover purposes but also for scale-out reasons. Is this possible? If not: why? If yes: how do I do it? Actually, it would be nice if someone could briefly explain how the data node replication technically works for my better understanding :)
MySQL Cluster automatically replicates data nodes, based on the number of replicas and the number of data nodes you specify. Your tables are automatically sharded across these nodes, and updates are replicated between them for HA.
Worth watching the short demo to see more: http://www.oracle.com/pls/ebn/swf_viewer.load?p_shows_id=11464419
For more detail - take a look at the 'Cluster overview' section of the docs: http://dev.mysql.com/doc/refman/5.5/en/mysql-cluster-basics.html