Week 1

The code for week 1 can be found in this Google Colab notebook. Make sure that you are using the "T4 GPU" runtime when running the cells

!pip install pycuda

import pycuda.driver as cuda

import pycuda.autoinit

from pycuda.compiler import SourceModule

import numpy as np

from time import perf_counter

module = SourceModule('''

__global__ void multiply(float *dest, float *a, float *b) {

 int i = blockIdx.x * blockDim.x + threadIdx.x;

 dest[i] = a[i] * b[i];

}

''')

multiply = module.get_function("multiply")

a = np.random.randn(1000000).astype(np.float32)

b = np.random.randn(1000000).astype(np.float32)

dest = np.zeros_like(a)

t0 = perf_counter()

multiply(cuda.Out(dest), cuda.In(a), cuda.In(b), block=(1000,1,1), grid=(1000,1,1))

t1 = perf_counter()

print(t1 - t0)

dest2 = np.zeros_like(a)

t0 = perf_counter()

for i in range(N):

 dest2[i] = a[i] * b[i]

t1 = perf_counter()

print(t1 - t0)