Skip to content

Benchmark of a simple GraphQL server on popular spec implementations

License

Notifications You must be signed in to change notification settings

tsegismont/graphql-server-benchmark

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

69 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

GraphQL server benchmark

Benchmark of a simple GraphQL server on popular spec implementations.

Lastest results

The benchmark

The benchmark consists in a simple blogging GraphQL server:

  • fetching json (author profiles) from a remote HTTP server

  • reading rows (posts and comments) from a Postgres database

GraphQL Server Banchmark
Schema file
type Post {
    author: Author!
    title: String!
    content: String!
    comments: [Comment!]
}

type Comment {
    author: Author!
    post: Post!
    content: String!
}

type Author {
    firstName: String!
    lastName: String!
    bio: String!
    posts: [Post!]
    comments: [Comment!]
}

type Query {
    posts: [Post!]
    author(id: Int!): Author
}

The backend HTTP service is setup with 10 authors.

The Postgres database is setup with:

  • 2 posts per author

  • 2 comments per post:

    • 1 comment from another author

    • 1 comment (reply) from the post author

Each framework is tested with different queries.

Guidelines

The server must:

  • accept the BACKEND_HOST environment variable for configuration

  • accept the POSTGRES_HOST environment variable for configuration

The server may:

  • cache author profiles per-query

  • use HTTP and Postgres pipelining

The server must not:

  • cache all posts and comments

  • cache all author profiles

Test environment

  • 3 quad-core servers Intel Xeon e5-2637v3 128G RAM (injector, server, backend)

  • 1 quad-core server Intel Xeon e5-2637v3 256G RAM (postgres)

  • 10 GbE switch

Contributing

You can add an implementation in the implementations directory.

Make sure the name describes the software stack. For example, Java/vertx-graphql-java means the implementation:

  • is written in the Java language

  • uses Vert.x as the web server

  • uses GraphQL-Java engine

Each implementation must provide:

  • a Dockerfile (build and run)

  • a set-env.sh file that defines the FRAMEWORK_NAME implementation (must be unique across implementations)

For testing purposes, you can build and run with Docker:

About

Benchmark of a simple GraphQL server on popular spec implementations

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published