From the course: AWS Essential Training for Developers

What is database as a service (DBaaS)? - Amazon Web Services (AWS) Tutorial

From the course: AWS Essential Training for Developers

Start my 1-month free trial Buy for my team

What is database as a service (DBaaS)?

“

- AWS has a lot of services for helping you manage databases, and this can be pretty overwhelming. Having a cloud vendor help you manage the servers and backups of your data is called Database as a Service, or DBaaS, and it lets you focus on the data itself and optimizing retrieval, versus configuring failover and managing database engine updates. If you are migrating an existing application into AWS, then your database technology has already been chosen for you, either by your development team or by the vendor who created your app, and so I would recommend looking at the AWS Data Migration Service, which can take your on-premise relational database and create a real-time in-sync clone of your database in the cloud so you can perform a smooth transition from your on-premise data center into AWS. Now, if you're building a new application or re-architecting an existing application, choosing a database provider can be complicated. Before you start having an internal war with your team about whether to use an Oracle database or a Postgres database. I would back up and look at the data needs for your whole application. What kind of data will you be storing? How much data do you currently have? How much data will you write to your database over the period of, I don't know, say, one hour? How much data will you read? Will you be running a bunch of large queries and reports? What kind of uptime do you need? Ask some higher-level questions first, and then design a database solution around those broad categories. If you're going to store a lot of historical data for reporting, I don't know, maybe you need a big data solution, but if you need real-time access to a current month of data, maybe you just store that in a relational database and then move the rest of it into a big data store later. If your relational database is going to ingest a large stream of data, maybe you need a queue that sits in front of the relational database that aggregates some of the data over, say, like a five-minute period, and then it only inserts the aggregated data into the relational database. If the data types that you're storing are object-based and you don't need a lot of reporting, maybe it's better to use a NoSQL database. Once you know the broad strokes and identify the larger pieces of how it's all going to tie together, then you research which relational database would be good for your application based on feature sets and budget and things like vendor support. There are not right and wrong answers here when choosing database vendors, but there's only good, better and best for your specific use case. But using the wrong type of database technology in the wrong way can really hurt your app going forward. Relational databases are not a good substitute for a big data store and vice versa. Even if you're familiar with database types, I would recommend you read the book "Designing-Data Intensive Applications." It doesn't favor one specific database engine, but it talks about the broad strokes and how the database technologies all differ from each other and when is the right time to employ each strategy. Most of my experiences with relational database engines, and this book really helped me fill in some gaps in my own knowledge. Getting this higher perspective on how to store and process data within your app will help you select between all of the various services that AWS offers for storing your data. Now, let's look at what AWS can do for us.

Contents