Since the shards all have a unique name within the cluster, there is no requirement as to the number of shards per worker.
To design for later scale out, it is good practice to begin with a larger number of shards than workers.
As an example, most queries to the fare cache involve all fares between multiple cities with durations for departure dates of a week to a month.
These queries are being made into a database table that is likely being updated at the time.
The sub-tables could be placed on smaller instances.
Citus changes the DDL for CREATE TABLE to allow for any number of shards to be placed on workers.Hipmunk is all about discovering a better travel experience, and that requires knowing the alternatives.So each night, Hipmunk conducts proactive searches of a little over 100 million possible trips, which are then placed in a database called the fare cache. What if you just want to spend the weekend on an island?Postgre SQL has an extension that does this automatically, it is called Citus.Citus interacts with the Postgre SQL query planner and executor to permit distributed tables over a cluster of servers.