Hi, can you add this library? https://github.com/kartikk221/hyper-express

https://github.com/kartikk221/hyper-express/blob/master/docs/Benchmarks.md

I tried pogo a week ago and I'm curious about its results.

Can you try pogo and update the list, please.

Thank you!

I think it'd be good to see results for Fastify, Express and Fastro, alongside a basic Deno and Node server, just to give even broader results, and it'd be nice to compare them with basic Node and Deno

So in a nutshell, maybe more could be added:

• Fastify
• Express
• Node
• Fastro
• Deno

If it helps speed up development time (should you agree with this suggestion), here's the code we've used in Drashland:

// node
const http = require("http");
const port = process.argv[2] || "1447";
console.log("port", port);
http
  .Server((req, res) => {
    res.end("Hello World!");
  })
  .listen(port);

console.log("server started");

// deno
import { serve } from "https://deno.land/[email protected]/http/server.ts";
const s = serve({ port: 1447 });
console.log("http://localhost:1447/");
for await (const req of s) {
  req.respond({ body: "Hello World\n" });
}

console.log("server started");

// fastify
const fastify = require('fastify')()

// Declare a route
fastify.get('/', (request, reply) => {
  return "Hello World!";
})

// Run the server!
fastify.listen(1447);

console.log("server started");

// fastro
import { Fastro } from "https://deno.land/x/[email protected]/mod.ts";

const server = new Fastro();

server.get("/", (req) => req.send("Hello World!"));

server.listen({
  port: 1447
});

console.log("server started");

// express
const express = require('express')
const app = express()
const port = 1447

app.get('/', (req, res) => res.send('Hello World!'))

app.listen(port);

console.log("server started");

(Credit to @crookse for making these benchmarks though, we're going to favour your repository over ours)

Add another column for average requests per second all frameworks can handle

Funding

• You can sponsor this specific effort via a Polar.sh pledge below
• We receive the pledge once the issue is completed & verified

That way we can speed up benchmark CI by ~3 min

For many of the results on the homepage, you are benchmarking autocannon. When the HTTP client is slower than the HTTP server, you are benchmarking the HTTP client. That's why the numbers get so close towards the top -- they're within the margin of error for autocannon.

Deno and Bun's numbers will be more accurate (higher) using a tool that is better at keeping up. wrk is also not quite fast enough.

oha would be a more accurate choice, though http_load_test would be best.

I have benchmarked Deno a few times on my machine and many benchmarks Deno is 2x slower than Bun
Maybe you can try use another benchmarking tool like wrk or bombardier to see the result.

Funding

• You can sponsor this specific effort via a Polar.sh pledge below
• We receive the pledge once the issue is completed & verified

const oha = await Deno.spawn("oha", { args: [...args, url], stdin: "null", stdout: "piped", stderr: "null", });

My fault, and my PR's keep getting in a weird state - don't suppose someone could add the fix for me? the port in node/bare.js needs to be changed to 8000, eg: const port = process.argv[2] || "8000"

uWebSocket is also fast HTTP Server

I would like to know of my new router that it is faster than Hono, could you add it? it's a little faster if you add the host to "hasName"

import fun from "https://deno.land/x/endofunctor/fun.ts";

// I use the default server
// import { serve } from "https://deno.land/[email protected]/http/server.ts";
await Deno.serve(
  fun()([
    {
      path: "/",
      f: (_) => "hello world",
    },
  ]),
  { port: 8000 },
);

please let me know if you have any questions

Can you please add cheetah to the benchmarks?

The repository:
https://github.com/azurystudio/cheetah

The results can be more useful if you also include your system specifications.

Funding

• You can sponsor this specific effort via a Polar.sh pledge below
• We receive the pledge once the issue is completed & verified

Saw the ci failed for fastro, then it was removed but it's still failing

I could probably work on this later, to fix the initial errors and add it back in

Drash released a new version

Just responding in plain text "Hello, Bench!" is meaningless in real-world scenarios.

Please considering add some more complex scenarios like parsing query strings, parsing cookies, parsing JSON from request bodies, serializing JSON to response bodies, or even interacting with databases.

Will try open a pr tonight

They're cool and make stuff easier to read without memorizing a bunch of different values from the tables

Check the results for pogo in the README, everything is showing 0. Can confirm this is also present before my add drash back PR (just in case i might have broken something)

Would you be open to benching our open source deno + oak server at https://github.com/pmprosociety/authcompanion as a friendly one time request? That would mean a lot. I can showcase in readme and link to bench repo, if OK.

How about adding h3 to the bench?

import { createServer } from "node:http";
import { createApp, eventHandler, toNodeListener } from "h3";

const app = createApp();
app.use(
  "/",
  eventHandler(() => "Hello, Bench!")
);

createServer(toNodeListener(app)).listen(8000);

Funding

• You can sponsor this specific effort via a Polar.sh pledge below
• We receive the pledge once the issue is completed & verified