As group replication is built on top of existing replication framework, let's revisit how replication works in MySQL traditionally. MySQL has provided a replication feature in its management service and utilities. It is usable with all kinds of storage engines, despite their differences. We configure a topology of two nodes to replicate the data, from one node to the other node. The one that accepts client connections is called a master. The one to which data is being copied is called a slave.
The following diagram shows the MySQL Database Replication Overview:
As shown in the preceding diagram, the clients connect to a single database instance called Master, where the Master is responsible for running all transactions. It is also responsible for recording all of the events that can change the state of MySQL database (DML...