Whenever you are planning to consume cloud computing resources, there is always a balance that needs to be achieved between the requirement of maintaining control of the environment and instantiation, and having the ability to focus on the task at hand. This is true for the need to use a relational database. In AWS you have a choice: you can either deploy your own EC2 instance with a database engine on it to maintain control of the configuration and the underlying operating system, or you can choose to use an RDS instance that can provide the same functionality and deliver a database service with minimal management overhead.

The decision to use RDS will be driven by the business driver of cloud adoption. Sometimes the business driver behind the use of cloud services is simplifying the management; in this case it is always easier to use the RDS service instead of managing your own server. In other cases, the business driver is increased flexibility and automation; again, RDS comes out on top as the flexibility of being able to deploy new RDS instances straight out of the SDK and managing the RDS infrastructure straight out of the code can be a big benefit.