Binomial Heaps

Leftist Binomial heaps

trees Actual Amortized
Insert O(log n) O(1) O(1)
Remove min (or O(log n) O(n) O(log n)
max)
Meld O(log n) O(1) O(1)
 Min Binomial Heap
• Collection of min trees.
2 1
4

4 5 9 7 3

6 7 6 8
5
5
8
6 9
9
 Node Structure
• Degree
 Number of children.
• Child
 Pointer to one of the node’s children.
 Null iff node has no child.
• Sibling
 Used for circular linked list of siblings.
• Data
 Binomial Heap Representation
• Circular linked list of min trees. A
2 1
4

4 5 9 7 3

6 7 6 8
5
5
8
6 9 Degree fields
9 not shown.
 Insert 10
• Add a new single-node min tree to the collection.
• Update min-element pointer if necessary.
A
2 1
4

4 5 9 7 3

10
6 7 6 8
5
5
8
6 9
9
 Meld
1 4
5 7

B 7 3
9
A
8

• Combine the 2 top-level circular lists.

• Set min-element pointer.
 Meld
1 4
5 7

C 7 3
9

8
 Remove Min

• Empty binomial heap => fail.

 Nonempty Binomial Heap
• Remove a min tree.
• Reinsert subtrees of removed min tree.
• Update binomial heap pointer. A
2 1
4

4 5 9 7 3

10
6 7 6 8
5
5
8
6 9
9
 Remove Min Tree
• Same as remove a node from a circular list.

a b dc d e

• No next node => empty after remove.

• Otherwise, copy next-node data and remove
next node.
 Reinsert Subtrees

1 4
5 7

7 3
9
A B

8 subtrees

• Combine the 2 top-level circular lists.

 Same as in meld operation.
 Update Binomial Heap Pointer

• Must examine roots of all min trees to

determine the min value.
 Complexity Of Remove Min

• Remove a min tree.

 O(1).
• Reinsert subtrees.
 O(1).
• Update binomial heap pointer.
 O(s), where s is the number of min trees in final top-level
circular list.
 s = O(n).
• Overall complexity of remove min is O(n).
 Enhanced Remove Min

• During reinsert of subtrees, pairwise combine

min trees whose roots have equal degree.
• This is essential to get stated amortized bounds
and so is the correct way to remove from a
Binomial heap.
• The simple remove min described earlier does
not result in the desired amortized complexity
and so is incorrect for Binomial heaps.
 Pairwise Combine
2 1
4

4 5 9 7 3

10
6 7 6 8
5
5
8
6 9
9

Examine the s = 7 trees in some order.

Determined by the 2 top-level circular lists.
 Pairwise Combine
2 4 10 5 6 1 5

4 9 9 3 9
5 7

6 7 6 8

8
4
3
tree table 2
1
0
Use a table to keep track of trees by degree.
 Pairwise Combine
2 4 10 5 6 1 5

4 9 9 3 9
5 7

6 7 6 8

8
4
3
tree table 2
1
0
 Pairwise Combine
2 4 10 5 6 1 5

4 9 9 3 9
5 7

6 7 6 8

8
4
3
tree table 2
1
Combine 2 min trees of degree 0. 0
Make the one with larger root a subtree of other.
 Pairwise Combine
2 4 5 6 1 5

4 9 10 9 3 9
5 7

6 7 6 8

8
4
3
tree table 2
1
Update tree table. 0
 Pairwise Combine
2 4 5 6 1 5

4 9 10 9 3 9
5 7

6 7 6 8

8
4
3
tree table 2
1
Combine 2 min trees of degree 1. 0
Make the one with larger root a subtree of other.
 Pairwise Combine
2 4 6 1 5

4 9 10 5 3 9
5 7

6 7 6 9 8

8
4
3
tree table 2
1
Update tree table. 0
 Pairwise Combine
2 4 6 1 5

4 9 10 5 3 9
5 7

6 7 6 9 8

8
4
3
tree table 2
1
0
 Pairwise Combine
2 4 6 1 5

4 9 10 5 3 9
5 7

6 7 6 9 8

8
4
3
tree table 2
1
Combine 2 min trees of degree 2. 0
Make the one with larger root a subtree of other.
 Pairwise Combine
2 6 1 5

4 9 3 9
5 4 7

6 7 6 8
10 5

8 9
4
3
tree table 2
1
Combine 2 min trees of degree 3. 0
Make the one with larger root a subtree of other.
 Pairwise Combine
1 6 5

4 7 3
2 9

9 10 5 8
4 5

6 7 9
6
4
3
8 tree table 2
1
0
Update tree table.
 Pairwise Combine
1 6 5

4 7 3
2 9

9 10 5 8
4 5

6 7 9
6
4
3
8 tree table 2
1
0
 Pairwise Combine
1 6 5

4 7 3
2 9

9 10 5 8
4 5

6 7 9
6
4
3
8 tree table 2
1
0
Create circular list of remaining trees.
 Complexity Of Remove Min

• Create and initialize tree table.

 O(MaxDegree).
 Done once only.
• Examine s min trees and pairwise combine.
 O(s).
• Collect remaining trees from tree table, reset table
entries to null, and set binomial heap pointer.
 O(MaxDegree).
• Overall complexity of remove min.
 O(MaxDegree + s).
 Binomial Trees

• Bk is degree k binomial tree.

B0
• Bk , k > 0, is:

…
B2 Bk-1
B0 B1
 Examples

B0 B1 B2 B3
 Number Of Nodes In Bk
• Nk = number of nodes in Bk.
B0 N0 = 1
• Bk , k > 0, is:

…
B2 Bk-1
B0 B1

• Nk = N0 + N1 + N2 + …+ Nk-1 + 1
= 2k.
 Equivalent Definition
• Bk , k > 0, is two Bk-1s.
• One of these is a subtree of the other.

B1 B2 B3
 Nk And MaxDegree

• N0 = 1
• Nk = 2Nk-1
= 2k.
• If we start with zero elements and perform
operations as described, then all trees in all
binomial heaps are binomial trees.
• So, MaxDegree = O(log n).