Row-Major Cache Locality in 2D Arrays

Snippet

Row-Major Cache Locality in 2D Arrays

In C++, multi-dimensional arrays are stored in row-major order. Accessing elements row by row (i then j) is significantly faster than column by column because it utilizes the CPU cache more effectively by accessing contiguous memory addresses.

snippet.cpp

cpp

#include <iostream>
 
const int SIZE = 1000;
 
int main() {
    int matrix[SIZE][SIZE] = {0};
    long long total = 0;
 
    // Row-major traversal is faster due to spatial locality
    for (int i = 0; i < SIZE; ++i) {
        for (int j = 0; j < SIZE; ++j) {
            total += matrix[i][j];
        }
    }
 
    std::cout << "Sum: " << total << std::endl;
    return 0;
}

Breakdown

int matrix[SIZE][SIZE] = {0};

Allocates a contiguous block of memory representing a 2D array on the stack.

total += matrix[i][j];

Accesses elements in the order they are stored in memory, maximizing cache hits.

Previous snippet Next snippet

More Intermediate C++ exercises →

From your library

Row-Major Cache Locality in 2D Arrays

Related