DB Instance Types

The smallest management unit of RDS is DB instance. A DB instance is an isolated database environment on the cloud. Each DB instance can contain multiple user-created databases, and you can access a DB instance using the same tools and applications you might use to access a standalone database. You can easily create or modify DB instances using the management console or HTTPS-compliant application programming interfaces (APIs). RDS does not have limits on the number of running DB instances. Each DB instance has a unique identifier.

DB instances are classified into the following types.

**Table 1** DB instance types
DB Instance Type	Description	Notes
Single-node	A single-node architecture is less expensive than a primary/standby DB pair.	If a fault occurs on a single-node instance, the instance cannot recover in a timely manner.
Primary/Standby	An HA architecture. In a primary/standby pair, each instance has the same instance class. The primary and standby instances can be deployed in different AZs.	When a primary instance is being created, a standby instance is provisioned along with it to provide data redundancy. The standby instance is invisible to you after being created. If a failover occurs due to a primary instance failure, there is a brief interruption between your database client and the instance. The client needs to be able to reconnect to the instance. The default replication mode between the primary and standby instances is as follows: RDS for MySQL: Semi-synchronous
Read replica	A single-node or HA architecture	Read replicas include single-node read replicas and HA read replicas. Single-node read replicas: If you choose single-node read replicas, you are advised to buy more than one single-node read replica and enable database proxy. That way, if one read replica fails, the database proxy can route traffic to other read replicas. HA read replicas: If the physical server where a primary read replica is deployed fails, the standby read replica automatically takes over the workloads. When you purchase a read replica, select the same value for Table Name as the DB instance. If the replication between a read replica (single-node or HA) and the DB instance is abnormal, it can take a long time to rebuild and restore the read replica (depending on the data volume).