Using databases in GraphQL

GraphQL is a protocol for sending and receiving data. Apollo is one of the many libraries that you can use to implement that protocol. Neither GraphQL (in its specifications) nor Apollo work directly on the data layer. Where the data that you put into your response comes from, and where the data that you send with your request is saved, are up to the user to decide.

This logic indicates that the database and the services that you use do not matter to Apollo, as long as the data that you respond with matches the GraphQL schema.

As we are living in the Node.js ecosystem in this project and book, it would be fitting to use MongoDB. MongoDB offers a great client library for Node.js, and also uses JavaScript as its native choice of language for interactions and querying.

The general alternative to a database system like MongoDB is a typical SQL server with proven stability and enormous spreading. One case that I encounter more and more frequently involves systems and applications relying on older code bases and databases that need upgrades. A great way to accomplish this is to get an over-layering API level with GraphQL. In this scenario, the GraphQL server receives all requests, and, one by one, you can replace the existing code bases that the GraphQL server relies on. In these cases, it is helpful that GraphQL is database agnostic.

In this book, we will use SQL via Sequelize in order to see this feature in a real-world use case. For future purposes, it will also help you to handle problems with existing SQL-based systems.