* Program re-ordering for improved L2 cache hit rate. * Automatic performance tuning. # Motivations # Matrix multiplications are a key building block of most modern high-performance computing systems.

def simpson( ... ): # Composite Simpson integration for int(f(x),x=a..b) on n intervals. xs = np.linspace( ... ) # Array xs holds the interval boundaries as well as the mid points. I = 0.0 # ...