Item: Amazon Relational Database Service (RDS)
Rating: 9
Author: Eric Humphrey

Use Cases and Deployment Scope

We use Amazon ealational Database Service as a backend for a handful of internal applications. The advantages of using a PaaS RDBMS are many. Amazon handles backups, patching and upgrades. And since we use multi availability zone deployments, uptime is not a concern either. Are there downsides? Sure. We would like to see the option to keep backups longer than 30 days. There are workarounds, but we would rather it be built into the product. Another potential downside is the lack of control over upgrades. While this is almost always a good thing, there are cases where a legacy schema may depend on an older version.

All in all Amazon Relational Database Service is a great service. No complaints!

Pros and Cons

Automatic patching
Automatic backup
Mult-az HA deployment

Extend the retention time for backups
Provide an option to freeze a db version

Return on Investment

No downtime for patching is a huge plus. This allows our applications to stay online, providing better customer service.
Automated patching helps us maintain a secure baseline with no human effort.
Multi availability zone deployment provides better performance.

Alternatives Considered

Amazon Relational Database Service is the only PaaS we have tried.

Other Software Used

JIRA Software, Atlassian Confluence, Salesforce Communities

Likelihood to Recommend

If your application relies on legacy database engine features, you may not want to use Amazon Relational Database Service. If you are under regulatory (or business, for that matter) requirements to retain database backups for longer than 30 days, are you capable of writing a workaround? Are database dumps sufficient, or do you need file level backups?

RDS is a great PaaS database solution

Overall Satisfaction with Amazon Relational Database Service