This was a very interesting post. I really enjoyed the take away around byte-seconds:

how do you measure overall resource consumption? The answer is to multiply time by memory, which gets you a single dimension measured in byte-seconds. […]relatively it makes a lot of sense: If you use twice as much memory as your competitor, you can make up for it by being twice as fast.

Embarrassingly, this never occurred to me as a way to think about multi-dimensional optimization problems. I’m definitely going to use this in the future.