[Verse 1]
Building balanced trees, watch the order grow
Every node's got keys, children down below
When insertion comes, find the proper leaf
Pack it in but if it's full, prepare for some relief
Split that overflowing node right down the middle line
Promote the median up, reorganize and realign
Parent catches the promoted key with open arms
If parent overflows too, cascade up the charms
[Chorus]
B-tree magic, split and climb
Median up every time
Half stay low, half move right
Keep the balance burning bright
Overflow, split in two
Bubble up, push it through
Merge when sparse, borrow near
B-tree rules crystal clear
[Verse 2]
Deletion starts with finding where the target hides
If it's in a leaf, just pull it out and slide
But internal nodes need replacement strategy
Find inorder predecessor, swap identity
When a node gets too empty, time to rearrange
Borrow from a sibling if they've got keys to exchange
Rotate through the parent, keep the order intact
But if siblings are minimal, it's time to compact
[Chorus]
B-tree magic, split and climb
Median up every time
Half stay low, half move right
Keep the balance burning bright
Overflow, split in two
Bubble up, push it through
Merge when sparse, borrow near
B-tree rules crystal clear
[Bridge]
Minimum degree defines the lower bound
Half full nodes keep the structure sound
Root can be emptier, special case exception
Every other node maintains the same conception
[Verse 3]
Merge those underflowing nodes with siblings tight
Pull the separator down from parent height
Three become one in this combining dance
If parent empties out, give the merged child a chance
Properties preserved through every operation
Logarithmic height maintains performance optimization
Disk-friendly structure, minimize the reads
B-trees handle massive data storage needs
[Outro]
From root to leaf the order never breaks
Every split and merge, careful steps it takes
Balanced growth forever, that's the B-tree way
Efficient data access, every single day