Programming Google App Engine with Java

Chapter 11. Using Google Cloud SQL with App Engine

A sizable portion of this book is dedicated to Google Cloud Datastore, a schemaless persistent object store designed for applications that scale to arbitrary sizes. Those automatic scaling capabilities come at the expense of features common to relational databases such as MySQL, PostgreSQL, and Oracle Database. While many common tasks of web application programming are well suited, or even better suited, for a scalable datastore, some cases call for a real relational database, with normalized entries, real-time join queries, and enforced and migratable data schemas. And sometimes you just want to run third-party software that expects you to have a SQL database on hand.

For those cases, there’s Google Cloud SQL. A feature of Google Cloud Platform, Cloud SQL gives you a straight-up no-nonsense MySQL relational database, designed for ease of use from Cloud Platform runtime environments like App Engine. Your database lives on a Cloud SQL instance, a virtual machine of a particular size that runs for as long as you need the database to be available. Your app code connects to the instance to execute SQL statements and queries using a standard database interface. You can also configure an IP address for the instance and connect to it with any MySQL client.

Naturally, Cloud SQL is the opposite of Cloud Datastore when it comes to scaling. Each SQL instance has a large but limited capacity for concurrent connections, CPU, and data. If ...