The following command was tested in matrix:

This would be the following output:

As you can see the input used was 200,000. The time indicated in the output is in milliseconds so it took 166100 milliseconds to complete. A much larger input would result in the program running for a very long time.

{| class="wikitable mw-collapsible mw-collapsed"

21.90 0.07 0.07 600021 0.00 0.00 int* std::__copy_move<false, true, std::random_access_iterator_tag>::__copy_m<int>(int const*, int const*, int*)

From the flat profile,the hotspots for this program would be the function merge function whih was called 200,006 times and thee int* std::__copy__move. The merge function would take up 52.3% of the time

----

=== Assignment 2 ===

We decided to choose the neural network code to paralellize. The reason being it gives us a good learning opportunity and more importantly that we can potentially speed up the execution time of the program by using CUDA.

=== Assignment 3 ===