Suppose your entire civilization was running in a single Docker container. Being a SPOF, that's not ideal. Now, consider scaling out horizontally by running on

Suppose your entire civilization was running in a single Docker container. Being a SPOF, that's not ideal. Now, consider scaling out horizontally by running on a cluster of a dozen planets. That provides a bit more breathing room for your devops team. The website highscalability.com is a good resource on this topic of fault tolerance and availability. It turns out that a lot of what applies to keeping services like Google running applies to the entirety of the human race.