This work is part of the High Performance Computer Architecture class at the Federal University of São Carlos, Brazil. Thus, it was done in Portuguese.
A explicação completa da implementação pode ser lida no relatório.
Método | N = 10 | N = 100 | N = 1000 | N = 10000 |
---|---|---|---|---|
CPU | 0.080960 | 0.591018 | 5.403639 | 52.253612 |
GPU | 0.005314 | 0.010240 | 0.068769 | 12.089377 |
GPU + arrayfun | 0.000647 | 0.000682 | 0.000740 | 0.002888 |