Services that actually matter don't use this model. Over an appropriate time period AWS availability is exemplary.
However do note. Avoid us-east-1 if at all possible. Understand that despite the documentation, service control planes do not operate in isolation between availability zones. If you need higher uptime you need to have cross regional deployments. When AWS says a service is global, it actually means us-east-1 :)
However do note. Avoid us-east-1 if at all possible. Understand that despite the documentation, service control planes do not operate in isolation between availability zones. If you need higher uptime you need to have cross regional deployments. When AWS says a service is global, it actually means us-east-1 :)