Automatic web dyno scaling designed for Rails apps Starting at $0/mo.
Anytime a dyno sits idle waiting for a request, you’re wasting money. In a perfect world, your web dynos would be constantly busy, but never too busy to serve the next request. Of course this isn’t a perfect world, but Rails Autoscale can get you close, ensuring that when dynos sit idle too long, they scale down automatically.
Rails Autoscale is the only Heroku autoscaling solution that uses request queueing — the time between Heroku accepting a request and your application beginning processing — to trigger scaling events instead of response time or other unreliable metrics. The reality is that in many scenarios, adding dynos will not make your application faster. Adding dynos only helps when existing dynos are overloading and causing requests to queue.
Since request queueing information is only available at the application layer, there’s no single language-agnostic way to uncover this data. We’re Rails developers, so naturally we built the autoscaling solution we wish existed.
We could give you loads of knobs and switches, or we could give you a couple settings with sensible defaults. We prefer the latter. Tell us your dyno limit – we’ll never scale your app beyond this – and tweak the queue time threshold, and you’re done.
This isn’t rocket science. When too many requests are queued, your app scales up to meet the demand. When dynos are sitting idle for too long, your app scales down. See your recent activity in one simple chart. Adjust accordingly.
The available application locations for this add-on are shown below, and depend on whether the application is deployed to a Common Runtime region or Private Space. Learn More
|Region||Available||Installable in Space|