Apache Geode By Example - Partitioned & Replicated Regions
For the third episode of
Apache Geode By Example, I cover Apache Geode's two primary region types: partitioned and replicated. Think of regions like tables in a relational database. Like a table, a region is an abstraction that allows you to model your data.
Partitioned regions - a partitioned region spans the entire distributed system (each server node) and divides the data into buckets. Partitioned regions are best for large data sets, partitioned event listeners, and write heavy operations.
Replicated regions - depending on the level of replication, a server node will hold the primary copy of the data in a given region and then copy the data to replica regions on other servers. Best for read heavy applications with small data sets.
Checkout the Apache Geode docs for more specifics.