Multiplication of Matrix using threads

Multiplication of matrix does take time surely. Time complexity of matrix multiplication is O(n^3) using normal matrix multiplication. And Strassen algorithm improves it and its time complexity is O(n^(2.8074)).

But, Is there any way to improve the performance of matrix multiplication using the normal method.
Multi-threading can be done to improve it. In multi-threading, instead of utilizing a single core of your processor, we utilizes all or more core to solve the problem.

We create different threads, each thread evaluating some part of matrix multiplication.
Depending upon the number of cores your processor has, you can create the number of threads required. Although you can create as many threads as you need, a better way is to create each thread for one core.

In second approach,we create a separate thread for each element in resultant matrix. Using pthread_exit() we return computed value from each thread which is collected by pthread_join(). This approach does not make use of any global variables.

Examples:

Input : 
Matrix A
 1 0 0
 0 1 0
 0 0 1

Matrix B
 2 3 2
 4 5 1
 7 8 6

Output : Multiplication of A and B
2 3 2
4 5 1
7 8 6

Recommended: Please try your approach on {IDE} first, before moving on to the solution.

NOTE* It is advised to execute the program in linux based system
Compile in linux using following code:

g++ -pthread program_name.cpp

// CPP Program to multiply two matrix using pthreads 
#include <bits/stdc++.h> 
using namespace std; 
  
// maximum size of matrix 
#define MAX 4 
  
// maximum number of threads 
#define MAX_THREAD 4 
  
int matA[MAX][MAX]; 
int matB[MAX][MAX]; 
int matC[MAX][MAX]; 
int step_i = 0; 
  
void* multi(void* arg) 
{ 
    int core = step_i++; 
  
    // Each thread computes 1/4th of matrix multiplication 
    for (int i = core * MAX / 4; i < (core + 1) * MAX / 4; i++)  
        for (int j = 0; j < MAX; j++)  
            for (int k = 0; k < MAX; k++)  
                matC[i][j] += matA[i][k] * matB[k][j]; 
} 
  
// Driver Code 
int main() 
{ 
    // Generating random values in matA and matB 
    for (int i = 0; i < MAX; i++) { 
        for (int j = 0; j < MAX; j++) { 
            matA[i][j] = rand() % 10; 
            matB[i][j] = rand() % 10; 
        } 
    } 
  
    // Displaying matA 
    cout << endl 
         << "Matrix A" << endl; 
    for (int i = 0; i < MAX; i++) { 
        for (int j = 0; j < MAX; j++)  
            cout << matA[i][j] << " "; 
        cout << endl; 
    } 
  
    // Displaying matB 
    cout << endl 
         << "Matrix B" << endl; 
    for (int i = 0; i < MAX; i++) { 
        for (int j = 0; j < MAX; j++)  
            cout << matB[i][j] << " ";         
        cout << endl; 
    } 
  
    // declaring four threads 
    pthread_t threads[MAX_THREAD]; 
  
    // Creating four threads, each evaluating its own part 
    for (int i = 0; i < MAX_THREAD; i++) { 
        int* p; 
        pthread_create(&threads[i], NULL, multi, (void*)(p)); 
    } 
  
    // joining and waiting for all threads to complete 
    for (int i = 0; i < MAX_THREAD; i++)  
        pthread_join(threads[i], NULL);     
  
    // Displaying the result matrix 
    cout << endl 
         << "Multiplication of A and B" << endl; 
    for (int i = 0; i < MAX; i++) { 
        for (int j = 0; j < MAX; j++)  
            cout << matC[i][j] << " ";         
        cout << endl; 
    } 
    return 0; 
} 

Output:

Matrix A
3 7 3 6 
9 2 0 3 
0 2 1 7 
2 2 7 9 

Matrix B
6 5 5 2 
1 7 9 6 
6 6 8 9 
0 3 5 2 

Multiplication of A and B
43 100 132 87 
56 68 78 36 
8 41 61 35 
56 93 129 97

An approach without using global variables:

NOTE* It is advised to execute the program in linux based system
Compile in linux using following code:

g++ -pthread program_name.cpp

// C Program to multiply two matrix using pthreads without  
// use of global variables  
#include<stdio.h> 
#include<pthread.h> 
#include<unistd.h> 
#include<stdlib.h> 
#define MAX 4 
  
  
//Each thread computes single element in the resultant matrix 
void *mult(void* arg) 
{ 
    int *data = (int *)arg; 
    int k = 0, i = 0; 
      
    int x = data[0]; 
    for (i = 1; i <= x; i++) 
           k += data[i]*data[i+x]; 
      
    int *p = (int*)malloc(sizeof(int)); 
         *p = k; 
      
//Used to terminate a thread and the return value is passed as a pointer 
    pthread_exit(p); 
} 
  
//Driver code 
int main() 
{ 
  
    int matA[MAX][MAX];  
    int matB[MAX][MAX];  
      
      
    int r1=MAX,c1=MAX,r2=MAX,c2=MAX,i,j,k; 
  
  
    // Generating random values in matA 
    for (i = 0; i < r1; i++)  
            for (j = 0; j < c1; j++)  
                   matA[i][j] = rand() % 10;  
            
        // Generating random values in matB  
    for (i = 0; i < r1; i++)  
            for (j = 0; j < c1; j++)  
                   matB[i][j] = rand() % 10;  
     
    // Displaying matA          
    for (i = 0; i < r1; i++){ 
        for(j = 0; j < c1; j++) 
            printf("%d ",matA[i][j]); 
        printf("\n"); 
    } 
              
    // Displaying matB                 
    for (i = 0; i < r2; i++){ 
        for(j = 0; j < c2; j++) 
            printf("%d ",matB[i][j]); 
        printf("\n");     
    } 
      
      
    int max = r1*c2; 
      
      
    //declaring array of threads of size r1*c2         
    pthread_t *threads; 
    threads = (pthread_t*)malloc(max*sizeof(pthread_t)); 
      
    int count = 0; 
    int* data = NULL; 
    for (i = 0; i < r1; i++) 
        for (j = 0; j < c2; j++) 
               { 
                 
               //storing row and column elements in data  
            data = (int *)malloc((20)*sizeof(int)); 
            data[0] = c1; 
      
            for (k = 0; k < c1; k++) 
                data[k+1] = matA[i][k]; 
      
            for (k = 0; k < r2; k++) 
                data[k+c1+1] = matB[k][j]; 
               
             //creating threads 
                pthread_create(&threads[count++], NULL,  
                               mult, (void*)(data)); 
                  
                    } 
      
    printf("RESULTANT MATRIX IS :- \n"); 
    for (i = 0; i < max; i++)  
    { 
      void *k; 
        
      //Joining all threads and collecting return value  
      pthread_join(threads[i], &k); 
             
            
          int *p = (int *)k; 
      printf("%d ",*p); 
      if ((i + 1) % c2 == 0) 
          printf("\n"); 
    } 
  
      
  
  return 0; 
} 

Output:

Matrix A
3 7 3 6 
9 2 0 3 
0 2 1 7 
2 2 7 9

Matrix B
6 5 5 2 
1 7 9 6 
6 6 8 9 
0 3 5 2 

Multiplication of A and B
43 100 132 87 
56 68 78 36 
8 41 61 35 
56 93 129 97

Attention reader! Don’t stop learning now. Get hold of all the important DSA concepts with the DSA Self Paced Course at a student-friendly price and become industry ready.

My Personal Notes

Recommended: Please try your approach on {IDE} first, before moving on to the solution.

Recommended Posts: