https://unix.stackexchange.com/questions/190350/mmu-less-ker...
676 karma · joined September 10, 2008
https://unix.stackexchange.com/questions/190350/mmu-less-ker...
I wonder if the Larrabee approach would work here: many simple RISC-V cores running in parallel, with wide vector units and a little specialized hardware (eg texture sampling).
I believe there's an exception for controlling remote devices when you need to make sure some rando doesn't brick your pricy gear.
Every time a bit of information is erased, a certain amount of energy must be dissipated as heat. A reversible computer doesn't (have to) erase information.
Reversible computing is energy-aware computing taken to its logical extreme.
If you want your CPU to run fast, you have to pipeline instructions and you need caches that can run at full CPU speed. It's easy to design a CPU without all the complexity, but it will be slow. No one knows how to do both simple and fast. It's probably impossible.
One data center * 24 backup generators * 3 MW each = 72 MW per data center. Four of those = 288 MW > 1/4GW.
Google has more than three data centers.
If you tried to use water ice as a standard, you'd find that its mass was changing all the time as the ice sublimated away, or humidity condensed onto its surface. It would be impossible to make precise measurements.
In the future, when a major feature like this is removed, it would be nice to see a banner across the top for people who will be affected.
100+ meter diameter radio telescopes pointed at the earth. That's the same size as the largest steerable ground-based radio telescopes. Makes you wonder what they can hear.
Of course, they would have to provide support to the support providers, but that would be a much smaller, more knowledgeable population.
I don't know anything solid about this, but I would very much like to know.