Array in data structure

 

What is Array ? OR Array Definition

Array in data structure is collection of items having same data type stored which are stored in contagious memory allocation. During array definition we should also keep in mind that array in data structure is a user define data type. Array in programming is a best data structure.
 
Implementation of array in programming is important because of following reasons
  1. a) Most assembly language have concept of array.
  2. b) From an array any other data structure we can be built.

Why do we need array?

In general if we store any data it store at any random location. Data at any 
random location is very tough to retrieve that’s why we need data structure.
 

Example of an Array

Consider three variable
 
int  a=10;
 
int b=20;
 
int c=30;
 
In array contain three elements we can write
 
int  a[3]={10,20,30};
 
Where 1000, 1002 and 1004 are the addresses where these variable are stored.

Application of array in data structure

Different application of array in data structure are as follows:
  • Array is used to sort the elements.
  • Array can be used to perform the matrix operations.
  • Array is used to implement the cpu scheduling.
  • Array can be used in recursive function.

Properties of an Array

There are following properties of an array.
  • Each element is to the same size.
  • Element are stored contagious, with the first element stored at the smallest memory address.
  • Starting address of array is called base address.
  • By default array is start from 0 and the element number is an address.

Memory as an Array

Memory can also be view as an array . Memory is either Byte addressable or word addressable
  • Byte Addressing: if each byte has a unique address, we have byte addressing.
  • Word Addressing: if each word is given unique address, but the byte within a single word cannot be distinguished.

Types of array in data structure

In general array may be of following types of array in data structure.
  • One Dimension array (1-D)
  • Two dimensional array (2-D)
  • Lower Triangular matrix array
  • Upper Triangular matrix array
  • Sparse matrix array .
Some other types of array in data structure are Tridiagonal matrix array, Z matrix array and Toeplitz matrix array.
 
Order of an Array
 
An array can be ordered in two way
  • Row major order
  • Column major order
Consider the matrix as shown in following figure.
 
 
Row major array for this matrix is
 
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
 
Column major order is
 
1 5 9 13 2 6 10 14 3 7 11 15 4 8 12 16

(i) One dimensional array

One dimensional array in data structure or Single-Dimensional array can be think as the” list of variables which have similar data types” and each variable can be distinctly accessed by specifying its index in square brackets preceded by the name of that array
 
int  a=10;
 
int b=20;
 
int c=30;
 
In array contain three elements we can write
 
int  a[3]={10,20,30};
 
 
Location of One D Array
 
Consider the array declaration : A[Lb……………..Ub], base address is BA, C-size of an array and Lb and Ub are the lower and upper bound of an array. Then address of the ith element can be find out using following formulas
 
                                     Loc(a[i]) = BA + (i – Lb) * C
 
Example: Suppose following array with its base address 1000 is given then address of the element at a[4] will be 1008,  
 
int a [5] = {10,20,30,40,50}
    Loc (a4) = BA + ( i – Lb) * c
                   = 1000 + ( 4 – 0 )*2
                   = 1000 + 8
                   = 1008
 
(ii) Two dimensional array
 
What is two dimensional array?
 
A two dimensional array in data structure is an array which is stored in the form of the row-column matrix, where the first index indicate the row and second index indicates the column. 
 
The second or the rightmost index of an array changes very fast as compared to first or left-most index while accessing the elements of an array.
Example – int a [4] [4] . This declaration represents an array which consist of 4 rows and 4 column.
 
Location of 2 D array in Row Major Order
 
int a[4] [4]=   A[Lb1………Ub1,Lb2………Ub2]
  Base Address =BA=1000
 
   S- size of any element
 
   No of rows- Ub1-Lb1+1
 
   No of columns- Ub2-Lb2+1
 
    LOC (aij)= BA+ [(i-lb1)*  no of column + (j- lb2)] * S
 
Example: Suppose if we want to find out the address of the element 32 then it will be calculated by above formula as follow:
    loc (a32) = BA+[(i-lb1)*column+(j-lb2)] * s
                    = 1000+[(3-1)*4+(2-1)]*2
                    = 1000+[8+1]*2
                    = 1000+18
                    = 1018
   So  the location of a32 in 2-d array,RMO form is 1018.
 
Location of 2 D Array in Column Major Order
 
int a [4] [4]
 
   a(Lb1…….Ub1,Lb2……Ub2)
 
   CMO
 
   Base address
 
   S- size of any element
 
   no of rows- Ub1-Lb1+1
 
   no of columns- Ub2-Lb2+1
 
         Loc(aij)= BA + [(J-lb2)*rows + (i-Lb1)] * S
 
Example
 
a[1…….4,1…….4]
 
     CMO
 
     BA- 1000
 
     Loc(a34) = BA +[(J-lb2) * no of rows + (I-Lb1)] *s
                     = 1000+[(4-1)*4+(3-1)]*2
                     = 1000+[12+2]*2
                     = 1000+28
                     = 1028
   So that the location of a34 in 2-d array, CMO form is 1028.
(iii) Lower triangular matrix 
 A[Lb1……Ub1,Lb2…….Ub2]
 
   LTM

 

 
   Row major Order for this lower triangular matrix is
 
    1 2 6 3 4 8 4 7 5 9 5 3 2 6 3
   
Base address
 
S- size of array
 
Loc (aij) = BA + [((i – Lb1)( i- Lb1+1)/2) + (j – Lb2)] * S
 A[lb1……..ub1,lb2……..ub2]
 
   LTM
 
   Column Major Order for this lower triangular matrix is
 
   1 2 3 4 5 6 4 7 3 8 5 2 9 6 3
   
Base address
 
size of array
 
loc(ai,j) = BA + [(col)(col+1)/2 –(ub2-j+1)(ub2-j+1+1)/2 + (i-j)] * s
 
(iv) Upper Triangular Matrix
 
   A[Lb1……Ub1,Lb2……Ub2]
 
    UTM

    Column Major Order
 
    Base address
 
    S- size of array
 
 LOC (aij) = BA + [((j – lb2)(j-lb2+1)/2) + (i-lb1)] * S
  • A[lb1…..ub1,lb2……ub2]
    UTM
   RMO
 
    Base address
 
    Size of array
 
LOC(ai,j) = BA + [(ub1-lb1+1)(ub1-lb1+1+1)/2 – (ub1-i+1)(ub1-i+1+1)/2 + (j-i)] * s
 

What is Sparse Matrix ?

Sparse matrix is a matrix which has most of the its elements of as 0 value, then it is called a sparse matrix.

Why to use Sparse Matrix instead of simple matrix ?

There are following reasons to use the sparse matrix
  • Storage: There are lesser non-zero elements than zeros and thus lesser memory can be used to store only those elements.
  • Computing time:Computing time can be saved by logically designing a data structure traversing only non-zero elements.
                            
In array this matrix can be represented as

Limitations of Array

There are following limitation of an array:
  • The dimension of an array is determined the moment the array is created, and cannot be changed later on.
  • An array is a static data structure. After declaring an array it is impossible to change its size. thus sometime memory spaces are misused.
  • Each element of array are of same data type as well as same size.we can not work with elements of different data type.
  • In an array the task of insertion and deletion is not easy because the elements are stored in contiguous memory location.
  • Array is a static data structure thus the number of elements can be stored in it are somehow fixed.
 

Leave a Reply

Your email address will not be published. Required fields are marked *

x