As discussed in WFLY-2298, the @Clustered annotation of AS7 has 2 orthogonal meanings: 1. If applied to a @Stateful EJB, then a specific cache will be used (as defined by the ejb subsystem) 2. If applied to a @Stateless or @Stateful EJB, remote EJB clients will have a dynamic view of the EJB deployment topology (i.e. the nodes on which a given bean is deployed). WFLY-2298 already aims to make the clusterability of a SFSB a property of the environment. This is effectively the case already, since @Clustered is ignored if the environment does not support HA. The omission of @Clustered means that the state should not replicate, whatever the environment - however, EJB 3.2 already provides a mechanism to disable replication/passivation via @Stateful(passivationCapable=false). Thus, reason #1 is no longer relevant. Reason #2 isn't really necessary either. @Clustered is ignored if, as mentioned above, the server profile does not support it, but also if the ejb client is not configured with a specific cluster name. Therefore, this behavior too can be made the default for HA profiles - and the client can dictate whether or not it is interested in deployment topology updates.