IBM builds 120 petabyte cluster out of 200,000 hard drives
extremetech.com
extremetech.com
Maybe Larry Ellison was right about thin clients/NCs, just way too early.
If we look at history , there were many exponentially improving technologies(for example speed of flight or height of buildings). But the trend stopped when it became to expensive to continue developing the technology.There is also a case to be made for physical barriers to a technology.
With chip manufacturing plants costing $3 billion , and only 3 chip leading edge manufacturers still operating today, and the element size of modern chips is a few atoms (and element size of an atom seems like a hard natural barrier), you could see moore's law stopping.
And even if we can build this storage device, will people want it in 2040 ?
Why does it need to be put in Apple terms? How about we put it in Liquid Audio terms?