Thoughts on env vars as I spend some time with Fly...

I actually like that Fly separates env vars and secrets as distinct concepts. Secrets get extra protection, env vars are for config.

What I don't like: Env vars require a code change. And for secrets, I actually had to *destroy* my machines after I unset a couple.

Heroku might be the only platform where you can change an env var and it just restarts. Just works, and fast.

Easy to overlook until you don't have it.

I love how Northflank's pricing page lets you choose the timeframe for the metered pricing.

All metered pricing should have this.

I just updated our PaaS Pricing Comparison Calculator:

- Added Amazon ECS Fargate
- Updated pricing for Railway
- Factored CPU time-slicing for Fly shared machines

Heroku still stands out as the most expensive, but the others are very competitive!

On @searls latest podcast he mentioned "Heroku" and "price", so I can't help but respond. I made the post public since I think it's a really useful pattern.

tldr: Got a high-memory job that pushes you to perf dynos? Isolate it and scale to zero.

Nearly 10 years (!!) of autoscaling, some patterns never change: noisy neighbors, too many queues, scaling workers to zero (and saving 💰), and the chaos of downscaling too far.

Stability is boring, repeatable choices.

Here's what we've learned 👉 https://judoscale.com/blog/autoscaling-insights-what-nearly-a-decade-of-autoscaling-your-apps-has-revealed-to-us