For the high availability of your application, it is critical to keep your database up and running all of the time. As horizontal scaling is not a straightforward option in the relational database, it creates additional challenges. To achieve high database availability, you can have a standby replica of the master database instance, as shown in the following diagram:
As shown in the preceding diagram, if the master instance goes down, your application server switches over to standby instance. Read replica takes the load off of master to handle latency. Master and standby are located in different availability zones, so your application will still be up even when an entire availability zone is down. This architecture also helps to achieve zero downtime which may be caused during the database maintenance window. When a master instance is down for maintenance, the application can failover to secondary standby instance...
