What is GraphOS?

We can do a lot just using the Sandbox Explorer—but a production API needs a whole suite of tools to keep everything running smoothly.

Enter Apollo GraphOS.

Apollo GraphOS is a complete cloud platform for building, managing, and scaling your GraphQL API. GraphOS provides a set of tools and services so that developers can focus on building better apps, faster. It consists of three main parts:

• GraphOS Studio - the web interface, where we can inspect detailed metrics and monitor performance
• Rover CLI - a command-line interface used locally or in a CI/CD pipeline
• Platform API - a GraphQL API which lets us perform GraphOS actions in our own application

Federation and the supergraph

By leveraging GraphOS for our API's growth, we'll be able to ship changes safely and quickly, and introduce new data sources and capabilities.

What we'll create today is a supergraph—it's our same GraphQL API, only now it's built to scale incrementally, following the principles of the federated architecture pattern. No monoliths here—as a federated graph, our supergraph is really the result of multiple smaller GraphQL services being composed together. These pieces are called subgraphs, and today we'll set one up as the foundation of a supergraph we can continue building piece-by-piece.

Following this pattern keeps our architecture lean, recombinable, and scalable, without losing the big picture of our API's complete capabilities. It makes it easier for teams to ship code AND collaborate, because individual subgraphs can follow their own development and deployment processes without compromising anyone else's workflow.

Introducing Studio

Today, we'll be working primarily out of Studio—it's our portal to the inner workings of our API, where we can set some controls and monitor performance. It's our handy web interface where we can visualize the pieces of our supergraph coming together as we build them.

Studio houses a number of pages, including the Explorer we worked with earlier, so querying our live supergraph can be done as soon as we've provisioned it. We'll also find insights such as operation and field-level metrics, client awareness, and a comprehensive guide to our supergraph's composed schema.