Databases in the Google Cloud

Spoiled for Choice

Even in the cloud age, a classic IT task is operating databases. In addition to SQL relational databases, non-relational databases are making inroads into the corporate landscape. Today, companies likely use more than just one kind of database.

In principle, cloud databases can be operated in the same way as databases in an organization's data center. The admin team simply installs the database as a virtual machine. Administration in this case is no different from local administration. Aspects such as high availability, patching, backup and restore, disaster recovery, and scaling remain the administrator's responsibility.

To make it as easy as possible for companies to operate databases, cloud providers also offer databases as managed services, which means the customer is, by and large, only responsible for the application and the database logic but does not need to worry about operations issues.

Choosing a Database

Google is one of the cloud pioneers: Google Cloud Platform (GCP) is one of the three largest cloud providers alongside Amazon Web Services (AWS) and Microsoft Azure. As such, it offers a comprehensive portfolio of database technologies, but choosing the right database is not always easy and needs to be carefully considered. A catalog of criteria can help:

The access method is important: Legacy applications such as billing software use SQL to access the database. The database needs to have an appropriate relational schema, support highly available operation, and, of course, maintain a consistent dataset at all times.
Access techniques can differ: Relational databases rely on SQL commands (e.g., SELECT, INSERT, UPDATE, DELETE) and usually support transactions. In contrast, NoSQL databases often use a REST interface, and object stores use files in the form of objects. Alternatively, a database can

...

Use Express-Checkout link below to read the full article (PDF).