It’s always interesting to see just how much of the common Internet runs on Amazon’s AWS. When AWS goes down - granted, not often at all - you quickly find out that half the websites you rely on everyday is no longer accessible. How can I function at work when Reddit doesn't load? Nobody can.
That goes to show how important AWS is, and how it really should have absolutely zero downtime. The backups and failsafes should have their own backups and failsafes.
Qualtrics - the online survey company - was completely non-functional during the AWS outage two days ago. (I know this because it’s a service we use at work.) Qualtric’s IT people must have the easiest troubleshooting job in the world: throw hands up, blame Amazon. There’s not much to do when the contracted third-party server your service runs on is acting up. Calling AWS support isn’t going to make them go any faster in fixing the problem. I don’t email Squarespace help whenever their service goes down (more often than I’d prefer, honestly). There's only the wait.
Good news for me, my livelihood is not dependent on this website. Bad news for Qualtrics, when your core service goes down for much of a work day, that’s a lot of lost revenue, never mind angry paying customers. Perhaps the company’s surely high-deductible insurance plan cover such events? If the cut is deep enough, I’d even think about suing Amazon. Word on the street is, Jeff Bezos has plenty of millions to spare.
Can you imagine your smart home devices stopped working on Monday because the backend is AWS? No doubt Internet-of-things make life convenient, but if a server outage somewhere causes me to be locked outside of my home, that’s not going to work.
Right next door.