And? The article makes no mention of what each server is responsible for, how much redundancy is built in (this is AWS where nodes die on a whim), whether all o

And? The article makes no mention of what each server is responsible for, how much redundancy is built in (this is AWS where nodes die on a whim), whether all of these servers are used for production or if there is a spread across dev and staging environments, etc. When all of these factors are considered, 52 servers isn't a huge number.