I am looking for the best solution for having multiple publish instances. I tried shared nothing and shared datastore configurations.
Is there any advantages or disadvantages of having 2 or more publish instances without cluster setup? In such a configuration how can I start a new publish instance? I mean how I should replicate the data from the author when I start a new publish instance (probably from backup), what are the best practices for this. How to solve reverse replication issues, so while I starting a new instance, other publish instance might get some new user generated data which must be replicated to the new publish instance too. What is your experience in this topic?
Thanks in advance!