finds.dev← search

// the find

igorkasyanchuk/rails_performance

★ 1,238 · Ruby · MIT · updated Jun 2026

Monitor performance of you Rails applications (self-hosted and free)

A Rails engine that gives you a self-hosted APM dashboard — p50/p90/p99 response times, slow requests, Sidekiq/DelayedJob monitoring, and CPU/memory tracking — all backed by Redis with no external service required. It's aimed at indie developers and small teams who want visibility into a production Rails app without paying for New Relic or Datadog.

Redis-only storage is a smart call — no extra database, expiry is built-in, and data never leaves your infrastructure. The coverage is wider than you'd expect: Grape APIs, Rake tasks, Sidekiq, DelayedJob, and custom `measure` blocks all work out of the box. The Kamal deployment event hook is a practical touch — annotating deploys on the timeline is exactly what you need when debugging a latency spike. The gem bundles its own CSS/JS rather than fighting your asset pipeline, which sidesteps a whole category of integration headaches.

All aggregation runs in-memory from raw Redis keys, which the README itself warns about — on high-traffic apps this will get slow or blow through memory. There's no persistence beyond the configurable window (default 4 hours), so you can't go back further to investigate an incident from last week. The TODO list is long and includes some basics (stacktraces for 500s, better small-screen UI, export to CSV) that feel like they should be done by now for a gem this old. Multi-server support exists but is rudimentary — you get per-host metrics with no aggregated view, so a cluster is harder to reason about than a single instance.

View on GitHub → Homepage ↗

// want more like this?

We dig through GitHub every week and send a few repos picked for what you actually care about — each with an honest take like this one.

Get finds in your inbox → Search again →