GraphQL is an open-source query language

GraphQL is an open-source query language and runtime for APIs running over HTTP/2 designed to retrieve structured data from servers.

First released by Facebook in 2015, GraphQL has since been adopted by numerous organizations including Microsoft, Google, Apple, Netflix, Snapchat, Pinterest, Atlassian, Dropbox, Instagram, and Yelp.

Main key concepts and features of GraphQL:

• Efficient Data Fetching: With GraphQL, clients can request precisely the data they need and get it in a single request, eliminating over-fetching or under-fetching of data. This reduces network round trips and improves performance.
• Declarative Data Fetching: Clients specify the structure of the data they require using a GraphQL query. The query is shaped like the desired response, allowing clients to request nested and related data in a single request.
• Strong Typing System: GraphQL has a type system that defines the shape of the data available in the API. This enables clients to know the structure of the data and provides better tooling for development.
• Real-time Updates: GraphQL supports subscriptions, which allow clients to receive real-time updates when specific data changes on the server. This is useful for building applications that require live data updates, such as chat apps or real-time dashboards.
• Versioning and Backward Compatibility: GraphQL avoids versioning issues often encountered with REST APIs by providing strong backward compatibility. Clients can evolve their data requirements without breaking existing functionality.
• Introspection: GraphQL APIs are self-describing, meaning clients can query the API itself to discover the available types, fields, and operations. This makes it easier to explore and interact with GraphQL APIs.
• Ecosystem and Tooling: GraphQL has a growing ecosystem with various tools, libraries, and frameworks that facilitate its implementation in different programming languages and platforms.

Clients send queries written in the GraphQL format to the API endpoint of connected databases, and the database returns only the necessary fields and related objects which the client requested. Unlike REST APIs, all GraphQL requests must contain a type definition schema to be able to parse results.

Query definitions may reference different types than those queried previously with one connection, leading to significant benefits when working with rich object models containing cyclical references or non-flat hierarchies.