Hacker News Reader

That day was a nightmare for a lot of people - it wasn't just S3 that went down, it was like all of US-EAST.

Luckily my company decided against multi-az for the cost savings so I spent all day firefighting.

Multi-AZ doesn't help when a whole region is down, unless you're referring to multi-region AZs (e.g us-east-1a and us-west-1a)

I have to think they're talking about the latter.

So what’s the cost breakdown? Did they make the right decision?

For one day of his time and probably a small part of a day of diminished service, most likely.