Running a RethinkDB proxy node
We have our cluster consisting of one SEED Server and a RethinkDB instance to support the cluster. The SEED server is acting as master server (however, there is no such master/child concept in RethinkDB; we are phrasing this for understanding) to perform routing of queries and run various kinds of business logic to determine where the data is stored and where to point the request. This in turn increases the load and dependency on one RethinkDB instance. Refer to the following figure; the proxy node acts as the node to distribute the query across instances in the cluster:
RethinkDB allows us to create a proxy node to solve such problems. The proxy node will act like a query redirector and will not store any data. We can create a proxy node in our cluster and join our other instances to it to form the cluster. The RethinkDB proxy node will:
Reduce CPU load by doing some query processing itself
Reduce intra-cluster traffic because it knows exactly where to point...