More people are picking up the drive story. I expect to hear rebuttals any time now from the big expensive disk players.
FWIW: we have been talking about this for a while. Lots of our partners have observed these things. MTBF is a great way to estimate things. The model appears to be broken, as it is not 5-20% off. But 5-10x off.
This is important. If you develop a theory, and it mispredicts something by a significant amount, you have, really, one option for your theory. Throw it out and develop a new one that better predicts the data.
This paper offers a tantalizing view of what a real theory could look like, by showing where the failures are. You can have a multi-modal distribution, with infant mortality of drives, as well as wear and tear mortality. Not to mention use-based mortality. All three distributions are different. They have very different characteristics.
But the interesting aspect remains that you cannot easily draw a hyperplane between SATA and the others on reliability. Which suggests that they really aren’t as different as some might wish you to believe.
Real data trounces FUD. You be the judge.