The assignment is develop to improve parallel programming skills by implementing an algorithm on a GPGPU device using CUDA. The task is to implement the computation of the power of a n-by-n square matrix A described in the PDF file attached.
Since your assignment will be automatically marked, it is crucial that you follow our instructions carefully. Your output will need to be in exactly the right format. To assist with this, we have made available some sample test cases. The sample test cases are by no means exhaustive. You will need to test your code more thoroughly by thinking carefully about the specifications and writing your own tests.
Note: The program needs to pass all the test cases available and pass the performance bench marks. The programmer needs to test the program on the server to match the expected input/output. Addtional reward will be given if the program gets a mark of 90-100% (This is usually obtain by writing own test cases and optimizing the program so it runs faster)
This is a second year assignment.