I assume you have seen many casual instances of cost overrun in that time. I'm sure you've also seen instances where an extra $10k flies out the door to AWS and people think "no big deal, that one was on us." This world doesn't have to exist. Even if AWS has a policy of always refunding people for a big oopsie, the fact that you have seen so many big ones suggests that you have also seen a lot of little ones.
By the way, there is nothing stopping AWS from reversing their trend of issuing refunds for big mistakes. "It hasn't happened in the past" isn't a great way to argue "it won't happen in the future."
Yes and I’ve also seen bad on prem build outs, bad hires, bad initiatives, proof of concepts that didn’t go anywhere, etc
Sure. The issues with AWS could all be solved with decent billing software, though. 15 years in there isn't a good excuse for this state of the world except that it's profitable.
You can set up billing alerts to trigger actions that stop things when they trigger. The easiest way is to take permissions away from the roles you create.
They give you the tools. It’s up to you to use them. If that’s too difficult, use the AWS LightSail services where you are charged a fixed price and you don’t have to worry about overages or the new free tier
https://aws.amazon.com/free/
Because despite what everyone here is saying, before July of this year, there was no such thing as a free tier of AWS, there was a free tier of some of their services
Instead of lightsail, I'll use digital ocean. It's a cheaper way to get undifferentiated VMs.
If it is so easy to set up these automations, Amazon can easily set up this automation for you. Ask yourself why they don't.
$1 cheaper per month for the lowest performance VM on both?