> pretty close proxy for how much computation is happening.
[citation needed]. See the vastly different power budget and cost of AWS graviton ARM vs x86 compute. Looking even at power use directly is only going to give a very low precision proxy for aggregate compute, with water usage even more indirect.
Looking at power use directly and making some educated guesses about average FLOPs/watt is probably the most effective way to estimate aggregate compute.
Even at Amazon I wouldn't be surprised if it's the primary way they do it, and I would be interested in some research. I'm trying to think of other ways, and accurately aggregating CPU/GPU load seems virtually impossible to do in a very rigorous way at that scale.
And yes, as an outsider you might have trouble knowing the relative distribution of ARM/x86, but that's just another number you want to obtain to improve your estimate.
Counterpoint: you have no factual basis for believing anything about the energy used by various CPUs in EC2, none of which are publicly available parts.
