**Objective:** Write an algorithm to find out whether in a given array there exists or not two numbers whose sum is exactly equals to a given number. This problem has been asked in Amazon and Microsoft interviews.

**Input:** An array arrA[], A number k

**Output:** True or false based upon we have found any two numbers in array arrA[] whose sum is equal to k

**Approach:**

**Method 1: Using Binary Search**

- First sort the array using Merge Sort(To know about Merge Sort and its implementation Click Here)
- Now do the linear scan to the from the left array , say starting index i=0
- Calculate Rem_Sum = number – arrA[i]
- If Rem_Sum<0, move to the next element
- If Rem_Sum>0, Perform
**Binary Search**for Rem_Sum on the remaining elements on the right side.

Time Complexity – O(nlogn)

**Method 2: Using Both Ends**

- First sort the array using Merge Sort(To know about Merge Sort and its implementation Click Here)
- Start from the both ends of the array
- Add first (say ‘a’) and last element(say ‘b’) of the array say S
- If S > number , S = S-(last_element) + (element before that)
- If S < number , S = S – (first element) + (next element)
- If if S=number, return true
- Repeat step
- If iteration gets over and retrun false.

**Complete Code for Both Methods:**

Output: USING Both Ends -Sum of two numbers in array A is 269 ??? :false USING Binary Search -Sum of two numbers in array A is 269 ??? :false USING Both Ends -Sum of two numbers in array A is 8 ??? :true USING Binary Search -Sum of two numbers in array A is 8 ??? :true

I’d think that after executing line #22, if RemS is negative we should break out of that loop. That’s because arrA is sorted in ascending order, so all subsequent instances of RemS will be also negative.