Is there a guide or tutorial for this?
And what is the difference of having a multi-cluster vs multi-nodes?
TIA 
Multi cluster means the compute resources and data are segregated.
It's no different from setting up a single cluster really, just keep the discovery list and cluster name different 
No, because if you search an index on one cluster you cannot see an index in the other one.
On a single cluster with multiple nodes and multiple indices you can search across all the indices.
One reason you may want separate clusters apart from this data level "partitioning", is if you have a high volume user and a whole bunch of smaller users. It makes sense to put the high volume user onto their own cluster and then share another cluster with all the smaller ones.
1 Like
Thank you @warkolm.
Is there a minimun number of nodes ideal for a cluster?
I've seen in some articles/blogs that it should be 3 nodes in a cluster.
No less than 3 is best practise, it maintains quorum.
1 Like
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.