Azure Cookbook

Chapter 4. Persisting Data

Almost every application needs to persist data in some form, and cloud applications are no exception. The data could be in one of the following forms:

• Binary or text files, such as images, videos, music, Apache Avro, CSV, or JSON formats

• Structured or semi-structured data stored in relational or NoSQL databases

In Chapter 3, we talked about Azure Storage, which can store unstructured data such as blob files (text, binary, etc.). In this chapter, we will talk about databases offered by Azure to persist relational and NoSQL data.

The first option is to go with IaaS database offerings. You can provision an Azure VM and install any database engine that you like, including Microsoft SQL Server, MySQL, or even MongoDB. This option gives you great flexibility because you own the underlying virtual machine; however, it requires a lot of administrative overhead. You are responsible for maintaining, securing, and patching the VM operating system and the database engine.

A second (and better) option, and the one we’ll focus on in this chapter, is to go with the Azure managed databases. These are PaaS (platform as a service) offerings, also called managed databases, which allow you to provision a ready-to-use database in minutes. Azure takes care of the underlying VM OS patching and security and many other administrative tasks such as scaling.

Azure SQL, Azure Cosmos DB, and Azure Database for PostgreSQL are a few Azure PaaS databases available to you. Having ...