While setting up quite an elaborate Fredhopper infrastructure, after reading the documentation and this question, I come to the conclusion that it's not really possible to have more than one indexing server per cluster, apart from using Fredhopper's cascading deployment model.
What I want to achieve is that we basically have two indexer servers which get updated simultaneously with a Tridion publish action. I can imagine index server 1 replicating data to index server 2, but as far as I can see, this is not supported out of the box without extra replication jobs.
Is doing this kind of replication even possible? If it is, how should this be built?