First, planning for AWS outage is pointless. Unless you provide service of national security or something, your customers are going to understand that when there's global internet outage your service doesn't work either. The cost of maintaining a working failover across multiple cloud providers is just too high compared to potential benefits. It's astonishing that so few eningeers understand the fact that maintaining a technically beautiful solution costs time and money, which might not make a justified business case.
Second, preparing for the disappearance of AWS is even more silly. The chance that it will happen are orders of magnitude smaller than the chance that the cost of preparing for such an event will kill your business.
Let me ask you: how do you prepare your website for the complete collapse of western society? Will you be able to adapt your business model to post-apocalyptic world where it's only cockroaches?
>Let me ask you: how do you prepare your website for the complete collapse of western society?
That's the main topic that going through my mind lately, if you replace "my website" with "Wikimedia movement".
We need a far better social, juridical and technical architecture regarding resilience as hostil agendas are in the rise at all level agaisnt sourced trackable global volunteer community knowledge bases.
Second, preparing for the disappearance of AWS is even more silly. The chance that it will happen are orders of magnitude smaller than the chance that the cost of preparing for such an event will kill your business.
Let me ask you: how do you prepare your website for the complete collapse of western society? Will you be able to adapt your business model to post-apocalyptic world where it's only cockroaches?