This is just marketing at work. from TFA: The real impact of the system may come in the application of these methods to aircraft or automobiles, which use control systems to react to inputs from the environment in order to achieve optimal safety and performance. Examples include traction control in cars and stabilization systems in jet fighters. “If you have sensors feeding in data to the reduced order model system, then it could solve the equation corresponding to the input data, and indicate the appropriate response in real-time based on the calculations you performed on a supercomputer,” This is how things work already: control systems on a jet fighters do not solve a CFD problem to know how to control the plane, they have a built in model (yeah, "reduced order", if you want to call it this way) that approximate the actual behavior of the plane. Doing it on a smartphone is useless. Furthermore the article has no details on how the error bounds are claculated.

You are right, fluid dynamics simulations parallelize beautifully, but once you start increasing the number of cores the communication between machines will slow things down. And you can have implicit, time accurate, temporal schemes for which the stability condition is (theoretically) CFL less than infinity. But it's clear that if you want to resolve turbulence time scales (and length scales) on a complicated case with relatively high Reynolds number a 100 million processor machine may not be enough.

I propose this: http://images3.wikia.nocookie.net/nonciclopedia/images/5/5d/Pullman_shuttle.jpg

That's why it is always better to do calculations by hand. Considering the big mass of the object you can discard drag effects. This leaves you with a uniform acceleration situation. Consideing a=9.81 m/s^2 and an height of 324 meters, this gives an impact speed of 80 m/s = 288 Km/h. This is obviously independent of the mass (http://en.wikipedia.org/wiki/Inertial_mass#Equivalence_of_inertial_and_gravitational_masses) of the object since we discard drag.