I would probably start with a configuration that is close to your ideal. A prototype, I suppose. Then you could be more specific about what you need. Something that would work but perhaps not ideally, for your first round, is to configure a synchronous replicated cache with a async cache loader, say the BerkeleyDB or JBPM. Add in some eviction. And you'd get something close to what you need. 3.2 will have better replication strategy. Having sync put and async removal is a feature maybe you could raise in JIRA? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4185811#... Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&a...