Although I didn’t see new attacks coming, I did push for partitioning caches on Schneier’s blog or untrusted processes on physically-different CPU’s + RAM sticks years ago. Got that last part from Lockdown prototype. There was apparently some research on randomization, too, I wasn’t aware of. I didn’t bother to look since someone’s always trying to use randomization.
I did a quick Google for recent papers on either to give an example of these kinds of techniques that might be retargeted for current problem. Interestingly, I found this one which was a peer review attacking an architecture with good results compared to how that normally works out.