#08 - Tree Indexes: B+Trees (CMU Intro to Database Systems)

[@MohammedAl-Jawaheri]

I'm a bit confused. at 34:50 it seems like we split 16 (as opposed to 15) because 16 was the key that was going to cause the split, but when we launched it up, we decided to take the middle node as the one to split up (13). why isn't it a consistent rule? when 16 was inserted why didn't we say 13 14 15 16 17 we'll split up 15 Edit: NVM so it's the smallest key in the right node in the leaf for partitioning purposes and then later always the middle one (before insertion of the new node) ?

[@rohandvivedi]

Can some one help me with this: How does borrowing key from the siblings work, when we are dealing with variable length keys? Because now there is a possibiltiy that borrowing just 1 key may not suffice. And, also the new parent separator entry may make the parent node overfull or underfull.

[@luatle7100]

1:01:48 hope that satisfies your question

[@RN-er7mz]

Nice slides

[@ankursinghprime]

Nice Intro