How do people usually scale out the elasticsearch cluster architecture?

I would like to know how people generally scale out the elasticsearch cluster.
I am bit lost with the scale out theory.

I just proposed a 3 node cluster each on 2 Datacenters wrapping up with the tribe node to one of our customer.

DC1
node1 mdi
node2 mdi
node3 mdi
node4 * tribe node

DC2
node1 mdi
node2 mdi
node3 mdi

I think next step should be that splitting the dedicated master and data node for future flexibility. But you could also just simply keep adding nodes and increase the "discovery.zen.minimum_master_nodes" setting.

I appreciate if someone could give me advice with the scale out theory.

That's the better long term plan, ie around 10 or more nodes.

1 Like

I see. Thanks.

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.