I wonder how this would compare to a B-Tree where you've explicitly loaded large branches of the tree into cpu cache and perform multiple comparisons in a singl

I wonder how this would compare to a B-Tree where you've explicitly loaded large branches of the tree into cpu cache and perform multiple comparisons in a single cpu cycle.