This just highlights how large public cloud providers are not exempt from outages/service issues and that redundancy needs to be built in at application level and hosting then distributed over multiple service providers.