3-way Merge Sort
Merge sort involves recursively splitting the array into 2 parts, sorting and finally merging them. A variant of merge sort is called 3-way merge sort where instead of splitting the array into 2 parts we split it into 3 parts.
Merge sort recursively breaks down the arrays to subarrays of size half. Similarly, 3-way Merge sort breaks down the arrays to subarrays of size one third.
Input: arr = [45, -2, -45, 78, 30, -42, 10, 19, 73, 93]
Output: [-45, -42, -2, 10, 19, 30, 45, 73, 78, 93]
Input: arr = [23, -19]
Output: [-19, 23]
After 3 way merge sort: -45 -42 -2 10 19 30 45 73 78 93
How the above code works?
Here, we first copy the contents of data array to another array called fArray. Then, sort the array by finding midpoints that divide the array into 3 parts and called sort function on each array respectively. The base case of recursion is when size of array is 1 and it returns from the function. Then merging of arrays starts and finally the sorted array will be in fArray which is copied back to gArray.
Time Complexity: In case of 2-way Merge sort we get the equation: T(n) = 2T(n/2) + O(n)
Similarly, in case of 3-way Merge sort we get the equation: T(n) = 3T(n/3) + O(n)
By solving it using Master method, we get its complexity as O(n log 3n).
Although time complexity looks less compared to 2 way merge sort, the time taken actually may become higher because number of comparisons in merge function go higher. Please refer Why is Binary Search preferred over Ternary Search? for details.
Auxiliary Space Complexity: The space complexity of 3-way merge sort is same as 2-way merge sort: O(n)
Similar article: 3 way Quick Sort
This article is contributed by Pavan Gopal Rayapati. If you like GeeksforGeeks and would like to contribute, you can also write an article using write.geeksforgeeks.org or mail your article to email@example.com.