Today, computers constantly shuttle data back and forth among faster and slower memories. The systems keep frequently used data close to the processor and then move it to slower and more permanent storage when it is no longer needed for the ongoing calculations.
In this approach, the microprocessor is in the center of the computing universe, but in terms of energy costs, moving the information, first to be computed upon and then stored, dwarfs the energy used in the actual computing operation. ...
One reason is computing’s enormous energy appetite. A 10-petaflop supercomputer — scheduled to be built by I.B.M. next year — will consume 15 megawatts of power, roughly the electricity consumed by a city of 15,000 homes.
March 1, 2011
Here's an article about HP's plan for next generation computing built around memristors (dreamed up by Amy Chua's dad 40 years ago). The problem with the current architecture:
There are perhaps analogies here to the evolution of human intelligence. For millions of years, our predecessors' brains got larger, peaking with the Neanderthals. That suggests that bigger brains made us fitter.
But that's a brute force solution.The usual argument is that big brains require too much food. (They also make us fall over more.) Now, it could be that as human population increased after the invention of agriculture, there were enough mutations to create more intelligence per cubic centimeter, just as there have been with computer chips.
That's certainly true, but I suspect the concomitant problem of not just getting enough energy in, but of getting enough energy out of the skull, of heat dispersal, also became a problem with this trend toward bigger brains. With a roughly spherical shape, as volume goes up, so does the volume to surface ratio.
Unfortunately, I've never seen anybody who actually knows what they are talking about consider this question of brains shedding heat.
It could be that there is an ideal latitude at which the cost of keeping the brain warm is balanced by the cost of keeping the brain cool at lowest overall cost. In 1911, Yale Professor of Geography Ellsworth Huntington conducted a study of climate's effect on human achievement. He concluded that the ideal climate was roughly that of New Haven, Connecticut. In a recent article, Malcolm Gladwell had great fun with that: here we are, 100 years later, and we can see what a biased moron Huntington was! Proving how much things have changed in 100 years, Malcolm's article appeared in that glossy, ad-packed magazine, The Lagoser.