Item: Amazon Relational Database Service (RDS)
Rating: 10
Author: Rahul Chhajed

Use Cases and Deployment Scope

We have been using Amazon RDS as our production database. Even with all our products we have been using RDS. RDS is a highly managed database and easy to setup other things like monitoring and auto scaling of database. A managed database reduces a lot of the headache of managing, maintaining, and other tasks related to database.

Pros and Cons

Setting up and managing a new RDS is just a matter of few clicks.
Maintaining different version of backup is just like a matter of turning on and off.
Automatic minor update of database.
Best restore:- You can even restore to 1 second before time.

RDS has everything that can be configured, but when you need efficient configuration we need little bit more knowledge about Dev Ops things.
Although restoration is available, when you need to do it it's pretty tricky. A new instance will be created and you will face some downtime.
Understanding which configuration and pricing would be suitable for startups as well [as] enterprise needs more reading and research.

Return on Investment

We don't need to spend on a dedicated resource for managing database as it's a self managed cloud database.
The alerts on high CPU usage and memory are also great to keep an eye on your database.
The scaling becomes an headache, but with RDS automatic scaling you need not worry about memory and CPU.

Alternatives Considered

Amazon Redshift and Amazon DynamoDB

People use both RDS and Redshift and both allow you to use your traditional database over cloud. But both RDS and Redshift have their own different usages. RDS is particularly suit[ed] for Online Transaction processing systems ( OLTP) whereas, Redshift is used for analytics and reporting purposes. Although Redshift [is] known as a relational database, it lacks to enforce unique key constraints. DynamoDB is also great database but the problem is with doing migration or using multiple database then you need something which has a more wider community who keeps on developing wrappers for ORM with different databases.

Key Insights

Do you think Amazon Relational Database Service (RDS) delivers good value for the price?

Yes

Are you happy with Amazon Relational Database Service (RDS)'s feature set?

Yes

Did Amazon Relational Database Service (RDS) live up to sales and marketing promises?

Yes

Did implementation of Amazon Relational Database Service (RDS) go as expected?

Yes

Would you buy Amazon Relational Database Service (RDS) again?

Yes

Other Software Used

Kibana, Elasticsearch, Slack, Sentry, Apache Airflow

Likelihood to Recommend

RDS is well suited for faster deployments and when you don't have a dedicated Database Administrator to manage your database environment. With RDS it will be so easy to manage your production and development environment [with] a few clicks and a little bit of research. Both enterprise and startups can use RDS with a reasonable pricing plan. If you [are] already using some other cloud platform, then the migration is pretty simple to RDS also.

Users and Roles

20 - All of our micro services uses RDS as their database. Even any service build for sales, marketing or engineering department. Their database is RDS as its the most reliable. Even for our analytics team they also connect RDS into metabase or any other analytics tool and don't have any problem in using that as compared to other databases.

Support Headcount Required

3 - - We have dev-ops team and member of engineering team who can look into RDS whenever required. But we don't need any specific team to monitor RDS as its already a managed service. And the configuration and other things are easy to change any one with little knowledge is able to change thinks with their needs.

Business Processes Supported

Using Database for development, production and also testing environment.
Adding analytics data and then pulling it in different tools for Data science.
Master- slave usage of database with multiple region availability.

Innovative Uses

In a secured way yeah we needed to deploy an analytical tool on our server and it should have limited accessibility so we used managed RDS
Needed some database where we don't need to have a professional DBA
Needed to setup a automated backup for database with a robust recovery.

Future Planned Uses

For building in house Analytics dashboard
For storing and querying data science related stuff.
Storing our old data we might use RDS as backup db

Likelihood to Renew

We do renew our use of Amazon Relational Database Service. We don't have any problems faced with RDS in place. RDS has taken away lot of overhead of hosting database, managing the database and keeping a team just to manage database. Even the backup, security and recovery another overhead that has been taken away by RDS. So, we will keep on using RDS.

Products Replaced

Key Differentiators

Product Features
Product Usability
Product Reputation
Prior Experience with the Product

Single most factor was easy integration and past experience. Creation and connecting with the database is so seamless that its a good to go database anyday. If I talk about past experience then also its great database not much headache with recovery or configuration even scaling up the database. Because recovery, and scaling is very hard in databases.

Evaluation Lessons Learned

Our selection process still depends on some of the factors which are same as previous. Database integration should be easy, reliable and community support if any incident occurs. And it should be popular not like new engineers doesn't even have any idea about what we are using. Connectors and wrappers should be available for easy integration with any language or frameworks.

RDS as your Database Administrator

Overall Satisfaction with Amazon Relational Database Service (RDS)