0% found this document useful (0 votes)

91 views

B Trees Slides

The document provides an overview of B-trees, which are balanced tree data structures used to access data stored on slow storage media like disks. It describes key properties of B-trees including their minimum degree, height restrictions, and how searching, adding, and removing values are performed. Values can only be added or removed from leaf nodes, and internal nodes must maintain a minimum number of child pointers. The document explains techniques for balancing the tree during operations like splitting full nodes, pushing values between nodes, and rotating nodes.

Uploaded by

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

91 views

B Trees Slides

Uploaded by

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 124

B-trees

Robert Horvick
SOFTWARE ENGINEER

@bubbafat www.roberthorvick.com
B-tree Overview
- Minimal degree

Overview - Height

Searching
Adding
- Splitting Nodes

Removing
- Push down
- Rotation
B-tree
A sorted, balanced, tree structure typically used to
access data on slow mediums such as disk or tape
drives.
B-tree Properties

3 6 10

1 2 4 5 7 8 9 11 12
B-tree Properties
1

3 6 10
2 3 4 5

1 2 4 5 7 8 9 11 12
B-tree Properties

3 6 10

1 2 4 5 7 8 9 11 12
B-tree Properties

3 6 10

1 2 4 5 7 8 9 11 12
B-tree Properties

3 6 10

1 2 4 5 7 8 9 11 12

Smaller Larger
B-tree Properties

3 6 10

1 2 4 5 7 8 9 11 12
B-tree Properties

3 6

1 2 4 5 7 8 9
B-tree Properties

3 6 10 12

… … … … …
Nodes will always have one
more child than values
Minimal Degree
The minimum number of children that every non-root
node must have. Represented as the variable “T”.
Minimal Degree

T Each non-root must contain at least T (minimal degree) children

T-1 Every non-root node will have at least T-1 values

2T Each node in the B-tree will contain a maximum of 2*T children

2T-1 Every node will have at most 2*T-1 values

Minimal Node
A non-root node that has T-1 values and T children.
Full Node
A node that has 2*T-1 values and 2*T children.
T = 3 (Valid)

3 6 10

1 2 4 5 7 8 9 11 12
T = 3 (Invalid)

3 6 10

1 4 5 7 8 9 11 12
T = 3 (Invalid)

3 6 10

1 4 5 7 8 9 11 12
T = 3 (Invalid)

3 6

1 2 4 5 7 8 9 10 11 12
T = 3 (Invalid)

3 6

1 2 4 5 7 8 9 10 11 12
Height
The number of edges between the root and the leaf
nodes. All B-tree leaf nodes must have the same height.
Height (Valid)

3 6 10

1 2 4 5 7 8 9 11 12
Height (Valid)

3 6 10
1

1 2 4 5 7 8 9 11 12
Height (Invalid)

3 6 15
1

1 2 4 5 7 10 14 16 17
2

8 9 11 12 13
Height (Invalid)

3 6 15
1

1 2 4 5 7 10 14 16 17
2

8 9 11 12 13
Searching
Searching

3 6 10

1 2 4 5 7 8 9 11 12

Smaller Larger
Searching

3 6 10

1 2 4 5 7 8 9 11 12
Searching

3 6 10

1 2 4 5 7 8 9 11 12
Searching

3 6 10

1 2 4 5 7 8 9 11 12
Searching

3 6 10

1 2 4 5 7 8 9 11 12
Searching

3 6 10

1 2 4 5 7 8 9 11 12
Searching

3 6 10

1 2 4 5 7 8 9 11 12
Searching

3 6 10

1 2 4 5 7 8 9 11 12
Searching

3 6 10

1 2 4 5 7 8 9 11 12
Searching

3 6 10

1 2 4 5 7 8 9 11 12
bool search(node, valueToFind)

foreach value in node

if valueToFind == value
return true

if valueToFind < value

return search(<left child>, valueToFind)
end

return search(<last child>, valueToFind)

end

Searching
Compare each value in the current node with the sought value, and then
recursively check every child’s value.
Searching node values can
be performed using a
binary search
Adding Values
Add Rules

Values can only be added to leaf nodes

Full nodes are split before the algorithm enters them (ensuring leaf
nodes are never full before adding an item)
add(T value) {
if Empty
Root = new BTreeNode(value)
else {
if Root.Full
SplitRootNode(Root)
InsertNonFull(Root, value)
}

Count++
}

Adding Data to the B-tree (Root Node)

If the root node is full then split the root node. Add the value to the non-full root
node.
InsertNonFull(BTreeNode node, T value) {
if node.Leaf
AddValueToLeaf (node, value)
else {
BTreeNode child = node.ChildForValue(value)

if (child.Full)
SplitChildNode(node)

InsertNonFull(child, value)
}
}

Adding Data to the Tree (Non-full Node)

If the non-full node is a leaf node then add the value, otherwise find the
appropriate child, splitting it if full, and then insert the node into the non-full
child node.
Adding Values
Adding Values

3
Adding Values

3 5
Adding Values

3 4 5
Adding Values

1 3 4 5
Adding Values

1 2 3 4 5
Adding Values

1 2 3 4 5 6
Adding Values

1 2 3 4 5

6
Adding Values

1 2 3 4 5
Adding Values

1 2 4 5
Adding Values

3
1

1 2 4 5
Splitting the root is the only
way that the B-tree height
increases
Adding Values

1 2 4 5
Adding Values

1 2 4 5
Adding a Value to a Leaf Node (T=3)

1 2 4 5 6
Root Node Splitting
Splitting two child nodes out of the root node to prevent
the root node from exceeding the maximum number of
values allowed by the B-tree degree.
Adding a Value to a Full Leaf (T=3)

1 2 4 5 6 7 8
Adding a Value to a Full Leaf (T=3)

1 2 4 5 6 7 8 9
Adding a Value to a Full Leaf (T=3)

1 2 4 5 6 7 8
Adding a Value to a Full Leaf (T=3)

1 2 6

4 5 7 8
Adding a Value to a Full Leaf (T=3)

3
1

1 2 6
2

4 5 7 8
Adding a Value to a Full Leaf (T=3)

1 2 4 5 6 7 8
Adding a Value to a Full Leaf (T=3)

3 6

1 2 4 5 7 8
Adding a Value to a Full Leaf (T=3)

3 6

1 2 4 5 7 8
Adding a Value to a Full Leaf (T=3)

3 6

1 2 4 5 7 8
Adding a Value to a Full Leaf (T=3)

3 6

1 2 4 5 7 8
Adding a Value to a Full Leaf (T=3)

3 6

1 2 4 5 7 8 9
Node Splitting
Splitting a child value in half by pulling the middle value
up to the parent.
Removing Values
Remove Rules

Values can only be removed from leaf nodes

Ensure that all nodes visited during the remove process have T values
before entering them
Balancing Operations