I watched a free high quality video with Aaron Hillegass about Core Data vs Tokyo Cabinet. Besides that this guy is amazingly funny (really, if you want to laugh now, watch it!), he shows off Tokyo Cabinet beeing about 40x faster than Core Data.
I wonder if it's worth thinking about how to attach this to Core Data? Does that make any sense? Maybe as a custom atomic store or something like this?
It is not the underlying persistence layer that makes his BNR (note that he was demoing his library not Tokyo Cabinet) faster. His persistence engine is faster because it does a LOT less than Core Data does.
The end point of his view is that if you do not need all of the functionality that Core Data offers you might want to look at his engine. However his engine does not do 1/10th of what Core Data provides for you in terms of memory management, object modeling, etc.
BNRPersistence, while extremely interesting (I am looking into parts of it from one of my projects) is not a drop in replacement for Core Data and Tokyo Cabinet, while fascinating is not a drop in replacement for sqlite at this time.