How to parallelize a for loop with OpenMP?

Question

C++ / C#

Help parallelize with OpenMP

for (i=0; i<N-1; i++) {
    for (k=i+1; k <= N-1; k++)
      for (j=i+1; j <= N; j++)
        A(k,j) = A(k,j)-A(k,i)*A(i,j)/A(i,i);
  }

  X[N-1] = A(N-1,N)/A(N-1,N-1);
  for (j=N-2; j>=0; j--) {
    for (k=0; k <= j; k++)
      A(k,N) = A(k,N)-A(k,j+1)*X[j+1];
    X[j]=A(j,N)/A(j,j);
  }

Answer 1

What is A(i, j)?
It looks like you have a Gaussian algorithm there, and it should be an array.
The outer loop over i cannot be parallelized in Gauss, but row subtraction can.
Add this before the loops for j and k:
#pragma omp parallel for collapse(2)
In the last two loops, the outer loop cannot be parallelized either, because the result of subsequent calculations depends on the previous iterations. But before the inner loop, feel free to stick #pragma omp parallel for.