GraphQL Overview: Why It Replaces REST and How to Implement It

Background

REST has been the dominant web‑service architecture for over 20 years because of its simplicity, scalability and JSON‑based data exchange, but its coarse‑grained endpoints often lead to redundant APIs, excessive data transfer, and increased workload for both front‑end and back‑end developers when requirements change.

GraphQL Introduction

GraphQL, originally developed by Facebook, is a query language and runtime that offers a more efficient, powerful and flexible alternative to REST, allowing clients to request exactly the data they need.

Benefits of GraphQL over REST

Unlike REST’s multiple endpoints, GraphQL exposes a single endpoint, supports declarative data fetching, reduces network traffic, is transport‑layer and database‑agnostic, and enables rapid client‑side feature expansion without server changes.

GraphQL Thinking Model

Design a data model (schema), let the front‑end declare the required fields, and let the GraphQL server automatically assemble and return the precise data.

Execution Logic

GraphQL can be introduced without discarding existing REST services; it can act as a middle layer between front‑end and back‑end, handling queries, mutations and subscriptions.

Basic Architecture

The client communicates only with the GraphQL layer, which in turn connects to various data sources (databases, micro‑services, etc.), allowing on‑demand data composition.

GraphQL Features Summary

Declarative data fetching (precise responses)

Single GraphQL endpoint per service

Transport‑layer and database‑agnostic

Supported Data Operations

Query – read data

Mutation – create, update, delete

Subscription – real‑time updates via WebSocket

Core Concept: Schema

The schema defines object types, fields and their relationships, similar to classes in OOP, and drives automatic data resolution.

Scalar Types

Built‑in scalars include String, Int, Float, Boolean, Enum and ID; custom scalars can also be defined.

Object Types

Object types group related fields and can model complex relationships (one‑to‑many, etc.).

Type Modifiers

Modifiers such as List ([Type]), Non‑Null (Type!), List‑Non‑Null ([Type]!), and Non‑Null List of Non‑Null ([Type!]!) allow fine‑grained control over field requirements.

Other Types

Interface – shared fields across implementing object types

Union – a type that can be one of several object types without common fields

Input – used for mutations, defined with the input keyword

Server Implementation

GraphQL servers exist for many languages, including Java, JavaScript, Go, Python, C#, Rust, Kotlin, Swift, etc.

Client Implementation

Clients are available for C#/.NET, Java/Android, JavaScript, Swift/iOS, Flutter, Python, and many others.

GraphQL Services and Tools

Popular managed services include Apollo Engine, Graphcool, Tipe, AWS AppSync and Hasura. Useful tools are GraphiQL, GraphQL Language Service and quicktype for type generation.

Resources

https://github.com/chentsulin/awesome-graphql

For more learning, visit GraphQL.cn and explore the awesome‑graphql community.