bài 1
#include <iostream>
#include <omp.h>
using namespace std;

int main() {
    int n = omp_get_num_procs();
    cout << "So CPU : " << n << endl;
    return 0;
}

bài 2
1.tuần tưj
#include <iostream>
#include <vector>
#include <cstdlib>
#include <ctime>
#include <chrono>
using namespace std;

int main() {
    int N, M;
    cout << "Nhap N: ";
    cin >> N;
    cout << "Nhap M: ";
    cin >> M;

    vector<vector<double>> A(N, vector<double>(M));
    vector<double> x(M), y(N);

    srand(time(NULL));

    for (int i = 0; i < N; i++)
        for (int j = 0; j < M; j++)
            A[i][j] = rand() % 10;

    for (int i = 0; i < M; i++)
        x[i] = rand() % 10;

    auto start = chrono::high_resolution_clock::now();

    for (int i = 0; i < N; i++) {
        y[i] = 0;
        for (int j = 0; j < M; j++)
            y[i] += A[i][j] * x[j];
    }

    auto end = chrono::high_resolution_clock::now();
    chrono::duration<double> elapsed = end - start;

    cout << "Thoi gian (chrono) = " << elapsed.count() << " s\n";
    return 0;
}

2. omp
#include <iostream>
#include <vector>
#include <cstdlib>
#include <ctime>
#include <omp.h>
using namespace std;

int main() {
    int N, M;
    cout << "Nhap N (so hang): ";
    cin >> N;
    cout << "Nhap M (so cot): ";
    cin >> M;

    vector<vector<double>> A(N, vector<double>(M));
    vector<double> x(M);
    vector<double> y(N);

    srand(time(NULL));

    for (int i = 0; i < N; i++)
        for (int j = 0; j < M; j++)
            A[i][j] = rand() % 10;

    for (int i = 0; i < M; i++)
        x[i] = rand() % 10;

    double start = omp_get_wtime();

    #pragma omp parallel for
    for (int i = 0; i < N; i++) {
        double sum = 0;
        for (int j = 0; j < M; j++)
            sum += A[i][j] * x[j];
        y[i] = sum;
    }

    double end = omp_get_wtime();
    cout << "Thoi gian (OpenMP) = " << (end - start) << " s\n";

    return 0;
}
3. mpi
#include <iostream>
#include <vector>
#include <cstdlib>
#include <ctime>
#include <mpi.h>
using namespace std;

int main(int argc, char** argv) {
    MPI_Init(&argc, &argv);

    int rank, size;
    MPI_Comm_rank(MPI_COMM_WORLD, &rank);
    MPI_Comm_size(MPI_COMM_WORLD, &size);

    int N, M;

    if (rank == 0) {
        cout << "Nhap N: ";
        cin >> N;
        cout << "Nhap M: ";
        cin >> M;
    }

    MPI_Bcast(&N, 1, MPI_INT, 0, MPI_COMM_WORLD);
    MPI_Bcast(&M, 1, MPI_INT, 0, MPI_COMM_WORLD);

    int localN = N / size;

    vector<double> x(M);
    vector<vector<double>> localA(localN, vector<double>(M));
    vector<double> localY(localN);

    if (rank == 0) {
        srand(time(NULL));
        vector<vector<double>> A(N, vector<double>(M));

        for (int i = 0; i < N; i++)
            for (int j = 0; j < M; j++)
                A[i][j] = rand() % 10;

        for (int i = 0; i < M; i++)
            x[i] = rand() % 10;

        // gửi A cho các process con
        for (int p = 1; p < size; p++)
            for (int i = 0; i < localN; i++)
                MPI_Send(A[p*localN + i].data(), M, MPI_DOUBLE, p, 0, MPI_COMM_WORLD);

        // copy phần rank 0
        for (int i = 0; i < localN; i++)
            for (int j = 0; j < M; j++)
                localA[i][j] = A[i][j];
    } 
    else {
        for (int i = 0; i < localN; i++)
            MPI_Recv(localA[i].data(), M, MPI_DOUBLE, 0, 0, MPI_COMM_WORLD, MPI_STATUS_IGNORE);
    }

    MPI_Bcast(x.data(), M, MPI_DOUBLE, 0, MPI_COMM_WORLD);

    MPI_Barrier(MPI_COMM_WORLD);
    double start = MPI_Wtime();

    for (int i = 0; i < localN; i++) {
        localY[i] = 0;
        for (int j = 0; j < M; j++)
            localY[i] += localA[i][j] * x[j];
    }

    vector<double> y;
    if (rank == 0) y.resize(N);

    MPI_Gather(localY.data(), localN, MPI_DOUBLE,
               y.data(), localN, MPI_DOUBLE,
               0, MPI_COMM_WORLD);

    MPI_Barrier(MPI_COMM_WORLD);
    double end = MPI_Wtime();

    if (rank == 0)
        cout << "Thoi gian (MPI) = " << (end - start) << " s\n";

    MPI_Finalize();
    return 0;
}
4. mpi + omp 
#include <iostream>
#include <vector>
#include <cstdlib>
#include <ctime>
#include <mpi.h>
#include <omp.h>
using namespace std;

int main(int argc, char** argv) {
    MPI_Init(&argc, &argv);

    int rank, size;
    MPI_Comm_rank(MPI_COMM_WORLD, &rank);
    MPI_Comm_size(MPI_COMM_WORLD, &size);

    int N, M;

    if (rank == 0) {
        cout << "Nhap N: ";
        cin >> N;
        cout << "Nhap M: ";
        cin >> M;
    }

    MPI_Bcast(&N, 1, MPI_INT, 0, MPI_COMM_WORLD);
    MPI_Bcast(&M, 1, MPI_INT, 0, MPI_COMM_WORLD);

    int localN = N / size;

    vector<double> x(M);
    vector<vector<double>> localA(localN, vector<double>(M));
    vector<double> localY(localN);

    if (rank == 0) {
        srand(time(NULL));
        vector<vector<double>> A(N, vector<double>(M));

        for (int i = 0; i < N; i++)
            for (int j = 0; j < M; j++)
                A[i][j] = rand() % 10;

        for (int i = 0; i < M; i++)
            x[i] = rand() % 10;

        for (int p = 1; p < size; p++)
            for (int i = 0; i < localN; i++)
                MPI_Send(A[p*localN + i].data(), M, MPI_DOUBLE, p, 0, MPI_COMM_WORLD);

        for (int i = 0; i < localN; i++)
            for (int j = 0; j < M; j++)
                localA[i][j] = A[i][j];
    }
    else {
        for (int i = 0; i < localN; i++)
            MPI_Recv(localA[i].data(), M, MPI_DOUBLE, 0, 0, MPI_COMM_WORLD, MPI_STATUS_IGNORE);
    }

    MPI_Bcast(x.data(), M, MPI_DOUBLE, 0, MPI_COMM_WORLD);

    MPI_Barrier(MPI_COMM_WORLD);
    double start = MPI_Wtime();

    #pragma omp parallel for
    for (int i = 0; i < localN; i++) {
        double sum = 0;
        for (int j = 0; j < M; j++)
            sum += localA[i][j] * x[j];
        localY[i] = sum;
    }

    vector<double> y;
    if (rank == 0) y.resize(N);

    MPI_Gather(localY.data(), localN, MPI_DOUBLE,
               y.data(), localN, MPI_DOUBLE,
               0, MPI_COMM_WORLD);

    MPI_Barrier(MPI_COMM_WORLD);
    double end = MPI_Wtime();

    if (rank == 0)
        cout << "Thoi gian (Hybrid MPI+OpenMP) = " << (end - start) << " s\n";

    MPI_Finalize();
    return 0;
}