CP4151-ADVANCED DATA

STRUCTURES AND
ALGORITHMS
UNIT 2
 UNIT II
9

HIERARCHICAL DATA
STRUCTURES
Binary Search Trees: Basics – Querying a Binary search tree – Insertion
and Deletion- Red Black trees: Properties of Red-Black Trees –
Rotations – Insertion – Deletion -B-Trees: Definition of B -trees – Basic
operations on B-Trees – Deleting a key from a B-Tree- Heap – Heap
Implementation – Disjoint Sets - Fibonacci Heaps: structure –
Mergeable-heap operations- Decreasing a key and deleting a node-
Bounding the maximum degree.
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
 Binary Search Trees
► A Binary Search Tree is a binary tree with the following
properties: Given a node x in the tree
◮ if y is a node in the left subtree of x, then key[y] ≤ key[x].
◮ if y is a node in the right subtree of x, then key[x] ≤ key[y].

13
7 23

4 8 17

10
3 5 21
9
1

► For simplicity, we will assume that all keys are distinct.

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
 Binary Search Tree Operations
► Given a binary search tree, there are several operations we want to
perform.
◮ Insert an element
◮ Delete an element
◮ Search for an element
◮ Find the minimum/maximum element
◮ Find the successor/predecessor of a node.
► Once we see how these are done, it will be apparent that the complexity of
each of these is O(h), where h is the height of the tree.
► The insert and delete operations are the hardest to implement.
► Finding the minimum/maximum and searching are the easiest, so we will
start with these.

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
BST: Minimum/Maximum
► The minimum element is the left-most node.
► The maximum element is the right-most node of the tree.
► Here are implementations of these methods:
node Find_Min(x) { node Find_Max(x) {
while(x.left!=null) while(x.right!=null)
x=x.left; x=x.right;
return x; return x;
} }

13

7 Maximum 23

4 8 17

10
3 5 21
9
1 Minimum
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
BST: Searching
► Search finds the node with value k in the tree rooted at x.
node Search(x,k) {
while(x!=null && k !=x.key) {
if(k<x.key)
x=x.left; else
x=x.right;
}
return x;
}

Search(x,10) x 13 Search(x,16)

7 23

4 8 17

10 NIL
3 5 21
9
1
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
 BST: Successor/Predecessor
► Finding the Successor/Predecessor of a node is harder.
► To find the successor y of a node x (if it exists)
◮ If x has a nonempty right subtree, then y is the smallest
element in the tree rooted at x.right. Why?
◮ If x has an empty right subtree, then y is the lowest
ancestor of x whose left child is also an ancestor of x.
Clearly.
node Successor(x) {
if(x.right!=null)
return Find_Min(x.right);
y=p[x];
while(y!=null && x=y.right) {
x=y;
y=y.parent;
}
return y;
}

► The predecessor operation is symmetric to successor.

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
BST: Successor Argument
► So, why is it that if x has an empty right subtree, then y is
the lowest ancestor of x whose left child is also an
ancestor of x?
► Let’s look at it the other way.
► Let y be the lowest ancestor of x whose left child is also an
ancestor of x.
► What is the predecessor of y?
► Since y has a left child, it must be the largest element in
the tree rooted at y.left
► If x is not the largest element in the subtree rooted at
y.left, then some ancestor of x (in the subtree) is the left
child of its parent.
► But y, which is not in this subtree, is the lowest such node.
► Thus x is the predecessor of y, and y is the successor of x.
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
BST: Successor Examples 13

Successor(7)=8 7 23

4 Find_Min 8 17

10
3 5 21
9
1

2 13

Successor(10)=13 7 23
1

4 2 8 y 17

1
10 x
3 5 21
9
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
1
Engineering
 BST: Insertion
► First, search the tree until we find a node whose
appropriate child is null. Then insert the new node there.
► Below, T is the tree, and z the node we wish to insert.
Insert(T,z) {
node y=null;
x=T.root;
while(x!=null) {
y=x;
if(z.key<x.key)
x=x.left;
else
x=x.right;
}
z.parent=y;
if(y==null)
T.root=z;
else
if(z.key<y.key)
y.left=z;
else
y.right=z;
}
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
 BST: Insertion Example
Insert(T,z) 13

7 23

4 8 17

y
3 5 10 21

1 9
6 22
z

Insert(T,z) 13

7 23

4 8 17

y
3 10 21
5

1 6 9
18 22
z

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 BST: Deletion

► Deleting a node z is by far the most difficult operation.

► There are 3 cases to consider:
 ◮ If z has no children, just delete it.
 ◮ If z has one child, splice out z. That is, link z’s parent and child.
 ◮ If z has two children, splice out z’s successor y, and replace the contents of z

with the contents of y.

► The last case works because if z has 2 children, then its successor has no left child.
Why?
► Deletion is made worse by the fact that we have to worry about boundary
conditions
► To simplify things, we will first define a function called
 SpliceOut.

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 BST: Splice Out
► Any node with at most 1 child can be “spliced out”.
► Splicing out a node involves linking the parent and child of
a node.
SpliceOut(T,y) {
//Two children--can’t splice out.
if(y.left!=null && y.right!=null) return;

if(y.left!=null) x=y.left;
else if (y.right!=null) x=y.right;
else x=null;

if(x!=null) x.parent=y.parent;

//Set y’s parent’s child to y’s child

if(y.parent==null) x=T.root;
else {
if(y==y.parent.left) y.parent.left=x;
else y.parent.right=x;
}
}
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
 BST: SpliceOut Examples
13 13

7 23 SpliceOut(T,z) 7 23

4 8 17 4 8 17

10 10
3 5 z 21 3 21
9 9
1 1

13 13

7 23 7 23
SpliceOut(T,z)
4 8 z 17 4 10 17

10 9
3 5 21 3 5 21
9
1 1

13 z 7
SpliceOut(T,z)
7 4 8

4 8 10
3
10
3

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 BST: Deletion Algorithm
► Once we have defined the function SpliceOut, deletion
looks simple.
► Here is the algorithm to delete z from tree T .
Delete(T,z) {
if(z.left==null || z.right==null)
SpliceOut(T,z);
else {
y=Successor(z);
z.key=y.key;
SpliceOut(T,y);
}
}

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 BST: Deletion Examples 13 13

7 23 7 23
Delete(T,z)
4 8 17 4 8 17

10 10
3 5 21 3 5 21
z 9
1 1

13 13

7 23 7 23
Delete(T,z)
4 8 17 4 8 17

10 z 9
3 5 21 3 5 21
9
1 1

13 13
z
7 23 8 23
Delete(T,z)
4 10 17 4 10 17

9 12 9 12
3 21 3 21
y 11
8 11 1
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
1
Engineering
 BST: Time Complexity
► We stated earlier, and have now seen, that all of the BST
operations have time complexity O(h), where h is the
height of the tree.
► However, in the worst-case, the height of a BST is O(n),
where n is the number of nodes.
► In this case, the BST has gained us nothing.
► To prevent this worst-case behavior, we need to develop a
method which ensures that the height of a BST is kept to a
minimum.
► Red-Black Trees are binary search trees which have
height Θ(log n).

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
Red-Black Trees
► A red-black tree is a BST with the following properties:
◮ Each node is colored either red or black.
◮ If a node is red, both its children are black.
◮ Every simple path from a node to a descendent leaf has the
same number of black nodes.
13

6 23

3 8 30

1 5 7 10 27 41

4 9 11

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Red-Black Trees Fact and Terms
► The black-height of a node x is the number of black
nodes, not including x, on a path to any leaf.
► A red-black tree with n nodes has height at most
2 log(n + 1).
► Since red-black trees are binary search trees, all of the
operations that can be performed on binary search trees
can be performed on them.
► Furthermore, the time complexity will be the
same–O(h)–where h is the height.
► Unfortunately, insertion and deletion as defined for regular
binary search trees will not work for red-black trees. Why
not?
► Fortunately, insertion and deletion can both be modified so
that they work, and still have time complexity O(h).
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Insert and Delete in RB Trees
► Inserting a node into a red-black tree is not trivial.
11 11
Insert(9)
7 13 7 13

3 8 24 3 8 24

1 4 1 4 9

Insert(10) ?
?

11 11
7 13 7 13

3 8 24 3 8 24

1 4 9 1 4 9

10 10

► Similar things happen when we try to delete nodes.

► We will not discuss in depth these operations.
► We will discuss some of the concepts, however.
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
 Red-Black Tree Insertion: Method
► To insert a node x into a red-black tree, we do the
following:
◮ Insert x with the standard BST Insert.
◮ Color x red.
◮ If x’s parent is red, fix the tree.
► Notice that x’s children, null, are black.
► Since we colored x red, we have not changed the black
height.
► The only problem we have is (possibly) having a red node
with a red child.
► Fixing the tree involves re-coloring some of the nodes and
performing rotations.

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
Left- and Right-Rotations
► Rotations are best defined by an illustration:
Right-Rotate(T,y)
y x

x A y
C
Left-Rotate(T,x)
A B B C

► Here, the letters A, B, and C represent arbitrary subtrees.

They could even be empty.
► It is not too hard to see that the binary search tree property
will still hold after a rotation.

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Rotation Example
13
x
7 23

4 8 17

10
3 5
9

Right-Rotate(T,x)

13

4 23

7 17
3

5 8

10

9
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
 Rotation Example
13

4 23

7 17
3
z
5 8

10

Left-Rotate(T,z)

13

4 23

8 17
3
7 10

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of 5 9

Engineering
 Insertion Example
11 11
Insert(T,10) 7 13
7 13

3 8 24 3 8 24
x
1 4 9 1 4 9

10

Left-Rotation(T,x)

11 11
7 13 Recolor 7 13

3 9 24 3 9 24

1 4 8 10 1 4 8 10

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Red Black Tree Summary
► Red-black trees are binary search trees which have height
Θ(log n) guaranteed.
► The basic operations can all be implemented in time
O(log n).
► Although inserting and deleting nodes only requires time
O(log n), they are nonetheless not trivial to implement.
► A regular binary search tree does not guarantee time
complexity of O(log n), only O(h), where h can be as large
as n.
► Thus red-black trees are useful if one wants to guarantee
that the basic operations will take O(log n) time.

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 B-‐Trees

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Problem: Data base too big to fit memory
Disk reads are slow

Example: 1,000,000 records on disk

Binary search might take
20 disk reads

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Disk reads are done in blocks

Example: One block read can retrieve

100 records

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of Engineering

 1,000,000 Records

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Block 0 Block 1 Block 2 ............. …B.…o
l …ck…9…9.………………...………...…………. Block 9999

1,000,000 Records

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 10,000 Records

Block 0 Block 1 Block 2 ............. …B.…o

l …ck…9…9.………………...………...…………. Block 9999

1,000,000 Records

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Block
Block 1 ................................... Block 99
0

10,000 Records

Block 0 Block 1 Block 2 ............. …B.…o

l …ck…9…9.………………...………...…………. Block 9999

1,000,000 Records

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Block
0
100 Records

Block 0 Block 1 ................................... Block 99

10,000 Records

Block 0 Block 1 Block 2 ............. …B.…o

l …ck…9…9.………………...………...…………. Block 9999

1,000,000 Records

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 DEF: A B-‐Tree of order m is an m-‐way tree such that
1. All leaf nodes are at the same level.
2. All non-‐leaf nodes (except the root) have at most m
and at least m/2 children.
3. The number of keys is one less than the number of
children for non-‐leaf nodes and at most m-‐1 and at
least m/2 for leaf nodes.
4. The root may have as few as 2 children unless the
tree is the root alone.

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Example for m = 5
DEF: A B-‐Tree of order 5 is an 5-‐way tree such that
1. All leaf nodes are at the same level.
2. All non-‐leaf nodes (except the root) have at most 5
and at least 2 children.
3. The number of keys is one less than the number of
children for non-‐leaf nodes and at most 4 and at
least 2 for leaf nodes.
4. The root may have as few as 2 children unless the
tree is the root alone.

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of Engineering

 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

A B F G

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

A B F G K

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

A B G K

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

A B D G H K M

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

A B D G H J K M

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

F J

A B D G H K M

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

F J

A B D E G H I K M R S

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

F J

A B D E G H I K M R S X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

F J R

A B D E G H I K M S X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

F J R

A B C D E G H I K M S X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

C F J R

A B D E G H I K M S X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

C F J R

A B D E G H I K L M N S T U X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

C F J R

A B D E G H I K L M N P S T U X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

C F J M R

A B D E G H I K L N P S T U X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Creating a B-‐tree of order 5

AGFBKDHMJESIRXCLNTUP

C F M R

A B D E G H I K L N P S T U X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Deleting
Nodes
• Delete E from leaf node

C F M R

A B D E G H I K L N P S T U X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Deleting
Nodes
• Delete E

C F M R

A B D G H I K L N P S T U X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Deleting
Nodes
• Borrow from a neighbor

C G M R

A B D F H I K L N P S T U X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Deleting
Nodes
• Delete F -‐-‐-‐ but can’t borrow from a neighbor

C G M R

A B D H I K L N P S T U X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Deleting
Nodes
Combine and push the problem up one level

C M R

A B D G H I K L N P S T U X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Deleting
Nodes
Can’t borrow so combine

C J M R

A B D G H I K L N P S T U X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Deleting
Nodes
Delete M from non-‐leaf node
Note: immediate predecessor in non-‐leaf
Is always in a leaf.

C J M R

A B D G H I K L N P S T U X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Deleting
Nodes
Delete M from non-‐leaf node

Overwrite M with immediate predecessor

C J L R

A B D G H I K N P S T U X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
 Deleting
Nodes
Borrow from a neighbor

C I L R

A B D G H J K N P S T U X

Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of

Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering
Mrs.P.Joy Suganthy Bai,AP/CSE Grace College of
Engineering