Phoenix Response Time Graphs

:heavy_check_mark: Very very low overhead (powered by DogSketch)

:heavy_check_mark: Cluster-wide performance charts (not just a single node!)

:heavy_check_mark: No external dependencies (runs 100% in-BEAM)

:heavy_check_mark: Accurate p50, p90, p99 and throughput

:heavy_check_mark: Linear and Log scale

Linear Scale

Log Scale

Caveat

To our knowledge nobody is running this in production yet. BUT, in theory, it should be able to track an obscene number of requests without slowing down your system. I suspect it will handle at least 100k requests per second per node with ease. Give it a try and let us know!

Installation

1. Add spotlight to your list of dependencies
2. Configure LiveView (if you haven't already)
3. Add spotlight to your Phoenix router

1. Add spotlight to your list of dependencies

Add to mix.exs:

def deps do
  [
    {:spotlight, "~> 0.1.0"}
  ]
end

Then run mix deps.get.

2. Configure LiveView (if you haven't already)

3. Add spotlight to your Phoenix router

# lib/my_app_web/router.ex
use MyAppWeb, :router
import SpotlightWeb.Router

...

scope "/" do
  pipe_through :browser
  spotlight("/spotlight")
end

We heavily recommend that you put Spotlight behind some kind of authentication before adding it to your production servers.