When analyzing trees, complexities are often O$($height of tree$).$ Given that B$-$trees can be used to produce trees of lower height than the equivalent BST$,$ explain why;

i$)$ B$-$trees are not a better choice than Red$-$Black trees when the tree is stored purely in memory.

ii$)$ B$-$trees are a better choice than Red$-$Black trees when the tree is stored on disk.