Setting up something that will scale

Hi Everyone,

Just looking for general advice. I am trying to setup a scale able instance on google cloud, where as needed I could automatically spin up new identical instances, and load balance requests towards it. Obviously not asking how to accomplish all this, but can anyone put me in the general direction of how to setup something like this

Generally;

  • have a set of nodes that always exist (ie a cluster that is always N nodes)
  • use auto expanding replicas if you have a read heavy use case
  • consider scaling client nodes dynamically instead of data nodes, to reduce shard movement

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