What about redundancy and DR ?
I would have thought that such an organization would want to ensure 100% uptime whatever happened. Putting everything into one basket, even a basket internally redundant and failovered and whatnot such as Google (or AWS for that matter), is still only one basket.
Cloud has demonstrated that there is no vendor that can boast 100% uptime over time. Okay, some are worse than others (eh, Microsoft?), but cloud is not easy and no one is safe from unscheduled downtime.
So, is this a demonstration that one cannot easily switch from one cloud vendor to another, or is it just ignorance of proper DR protocol ?