GraphQL Benchmarks

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

• Introduction
• Quick Start
• Benchmark Results
• Architecture
  • K6
  • 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

Throughput (Higher is better)	Latency (Lower is better)
{{ posts { id userId title user { id name email }}}}
{ posts { title }}
{greet}

Query	Server	Requests/sec	Latency (ms)	Relative
1	{ posts { id userId title user { id name email }}}
	Tailcall	30,004.80	3.32	111.96x
	async-graphql	1,862.55	53.71	6.95x
	Caliban	1,535.48	65.27	5.73x
	Hasura	1,474.01	67.57	5.50x
	GraphQL JIT	1,325.51	75.13	4.95x
	Gqlgen	763.01	130.02	2.85x
	Netflix DGS	362.91	160.04	1.35x
	Apollo GraphQL	268.00	367.08	1.00x
2	{ posts { title }}
	Tailcall	61,811.50	1.61	44.69x
	async-graphql	9,412.32	10.76	6.81x
	Caliban	9,057.38	11.39	6.55x
	Hasura	2,437.90	40.99	1.76x
	Gqlgen	2,176.64	47.66	1.57x
	Apollo GraphQL	1,750.92	57.02	1.27x
	Netflix DGS	1,598.51	69.51	1.16x
	GraphQL JIT	1,383.03	72.19	1.00x
3	{ greet }
	Caliban	68,200.20	1.07	26.86x
	Tailcall	63,728.50	1.58	25.10x
	async-graphql	51,025.70	1.98	20.09x
	Gqlgen	47,338.30	5.10	18.64x
	Netflix DGS	8,138.82	15.05	3.20x
	Apollo GraphQL	7,989.19	12.73	3.15x
	GraphQL JIT	5,136.79	19.44	2.02x
	Hasura	2,539.46	39.31	1.00x

Architecture

A client (k6) 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.

K6

k6 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!