Practical Parallelism in C++: MPI Gaussian Elimination Naive

In this video we look at a naive Gaussian Elimination implementation but in MPI this time!

[mario]

Your channel is amazing! I really hope the youtube algorithm picks you up soon!

[prateekranjan]

HI Nick. Awesome video. Quick question : Why are scale and pivot of "int" datatype and not float ? Since they both refer to the elements of A, shouldn't they be float as well ?

[yuriikvasiuk]

It's a little bit confusing. You are calling MPI_Scatter(matrix, N*num_rows ...), but what if N is not evenly divisible by size?

[selometas]

float *matrix = new float[N*N];
init_matrix(matrix, N ); // can you explain about this init_matrix?

[selometas]

//Get start time, have a question here what about MPI_barrier before t_start?

[TheKeyToMusicOfficial]

how many processors do you have on the machine you're testing on?