I'm not a big fan of putting machine learning into a file system. Usually you can't understand why a machine learning algorithm is doing what it's doing, and I would be worried about a production server suddenly having massively different performance because the cache learning algorithm started doing something differently. Interesting idea, but I would want to battle test it before I bought in.

It's impossible to tell what's going on nowdays with optimizing compilers, memory overcommitment, hardware branch predictions, power saving measures, etc.

Certainly a good idea, _if_ you have unlimited resources.

Without that, I think one should focus on getting the on disk structures right and the code manipulating them solid. That is enough of a challenge.

Isn't their lunch being eaten by intel optane?