GraphQL Benchmarks

Explore and compare the performance of the fastest GraphQL frameworks through our comprehensive benchmarks.

• Introduction
• Quick Start
• Benchmark Results
  • Throughput (Higher is better)
  • Latency (Lower is better)
• Architecture
  • WRK
  • GraphQL
  • Nginx
  • Jsonplaceholder
• GraphQL Schema
• Contribute

Introduction

This document presents a comparative analysis of several renowned GraphQL frameworks. Dive deep into the performance metrics, and get insights into their throughput and latency.

NOTE: This is a work in progress suite of benchmarks, and we would appreciate help from the community to add more frameworks or tune the existing ones for better performance.

Quick Start

Get started with the benchmarks:

1. Click on this link to set up on GitHub Codespaces.
2. Once set up in Codespaces, initiate the benchmark tests:

./setup.sh
./run_benchmarks.sh

Benchmark Results

Query	Server	Requests/sec	Latency (ms)	Relative
1	{ posts { id userId title user { id name email }}}
	Tailcall	21,130.70	4.73	177.93x
	GraphQL JIT	1,062.15	93.55	8.94x
	async-graphql	947.19	104.78	7.98x
	Caliban	738.85	136.07	6.22x
	Gqlgen	387.34	254.36	3.26x
	Netflix DGS	186.92	517.28	1.57x
	Apollo GraphQL	132.21	695.16	1.11x
	Hasura	118.76	774.49	1.00x
2	{ posts { title }}
	Tailcall	33,165.00	3.02	71.28x
	async-graphql	5,065.24	20.01	10.89x
	Caliban	4,754.92	21.56	10.22x
	GraphQL JIT	1,135.46	87.91	2.44x
	Gqlgen	1,101.39	99.69	2.37x
	Apollo GraphQL	905.12	110.95	1.95x
	Netflix DGS	807.80	124.74	1.74x
	Hasura	465.28	231.46	1.00x
3	{ greet }
	Tailcall	39,700.30	2.53	27.30x
	Caliban	32,756.20	3.08	22.53x
	Gqlgen	24,512.60	8.40	16.86x
	async-graphql	23,673.50	4.22	16.28x
	GraphQL JIT	4,568.26	21.84	3.14x
	Netflix DGS	4,183.74	28.25	2.88x
	Apollo GraphQL	4,066.61	27.50	2.80x
	Hasura	1,454.17	68.45	1.00x

1. {posts {title body user {name}}}

Throughput (Higher is better)

Latency (Lower is better)

2. {posts {title body}}

Throughput (Higher is better)

Latency (Lower is better)

3. {greet}

Throughput (Higher is better)

Latency (Lower is better)

Architecture

A client (wrk) sends requests to a GraphQL server to fetch post titles. The GraphQL server, in turn, retrieves data from an external source, jsonplaceholder.typicode.com, routed through the nginx reverse proxy.

WRK

wrk serves as our test client, sending GraphQL requests at a high rate.

GraphQL

Our tested GraphQL server. We evaluated various implementations, ensuring no caching on the GraphQL server side.

Nginx

A reverse-proxy that caches every response, mitigating rate-limiting and reducing network uncertainties.

Jsonplaceholder

The primary upstream service forming the base for our GraphQL API. We query its /posts API via the GraphQL server.

GraphQL Schema

Inspect the generated GraphQL schema employed for the benchmarks:

schema {
  query: Query
}

type Query {
  posts: [Post]
}

type Post {
  id: Int!
  userId: Int!
  title: String!
  body: String!
  user: User
}

type User {
  id: Int!
  name: String!
  username: String!
  email: String!
  phone: String
  website: String
}

Contribute

Your insights are invaluable! Test these benchmarks, share feedback, or contribute by adding more GraphQL frameworks or refining existing ones. Open an issue or a pull request, and let's build a robust benchmarking resource together!