It saves computational time by spreading operations over different CPUs— I didn’t clearly understand this line?
