While I was at AWS I managed to catch a couple of internal presentations from James Hamilton where he went deep into their global network infrastructure. It’s something that’s way outside my wheelhouse but that I find fascinating nonetheless.
In particular I always found the discussion about dealing with failure of global infrastructure that’s sitting in the ocean floor particularly intriguing. One non-obvious (to me at least) aspect was what do you do if there’s a failure of the power feed equipment that’s driving the cable. In most cases the answer was “double the voltage until you can repair it properly”. He explains it far better than I ever could on his blog: https://perspectives.mvdirona.com/2017/01/cs-responder-trans...
In particular I always found the discussion about dealing with failure of global infrastructure that’s sitting in the ocean floor particularly intriguing. One non-obvious (to me at least) aspect was what do you do if there’s a failure of the power feed equipment that’s driving the cable. In most cases the answer was “double the voltage until you can repair it properly”. He explains it far better than I ever could on his blog: https://perspectives.mvdirona.com/2017/01/cs-responder-trans...