Jhongesell/gpu_numba.ipynb

## gpu_numba.ipynb
{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "GPU_Numba",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "machine_shape": "hm",
      "authorship_tag": "ABX9TyOUAXg+UxAuVdEAxxl2r72f",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/mapolinario94/8176e1faf62923d3c670a99e996dcf4a/gpu_numba.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FGYEXyy3A6qH",
        "colab_type": "text"
      },
      "source": [
        "# PROGRAMACIÓN DE GPU CON NUMBA"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "N5gYK7Z17s3a",
        "colab_type": "text"
      },
      "source": [
        "---\n",
        "Este *notebook* es parte del *webinar* [**\"Python: Acelerando aplicaciones numéricas con Numba\"**](https://www.facebook.com/Capacita.INICTEL.UNI/videos/672245813319700/) expuesto por [Marco Apolinario](https://www.linkedin.com/in/marco-apolinario) el 29 de abril del 2020, y organizado por el Instituto Nacional de Investigación y Capacitación de Telecomunicaciones de la Universidad Nacional de Ingeniería (INICTEL-UNI).\n",
        "\n",
        "---"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vgainhOE-Mxh",
        "colab_type": "text"
      },
      "source": [
        "En este *notebook* se presentan dos ejemplos sobre el uso del lenguaje de programación Python y la librería Numba para programar una GPU (*Graphics Processing Unit*). El primer ejemplo consiste en la multiplicación de dos matrices, donde se muestra de forma resumida el procedimiento para la multiplicación seguido por la implementación en Python. El segundo ejemplo trata sobre la propagación de señales acústicas en un medio bidimensional homogéneo, lo que se simula usando el método de diferencias finitas. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MSkryXfL08Tr",
        "colab_type": "code",
        "outputId": "2dc20331-086a-40c8-f955-d7415db827ee",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 302
        }
      },
      "source": [
        "! nvidia-smi"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Thu Apr 30 00:18:26 2020       \n",
            "+-----------------------------------------------------------------------------+\n",
            "| NVIDIA-SMI 440.64.00    Driver Version: 418.67       CUDA Version: 10.1     |\n",
            "|-------------------------------+----------------------+----------------------+\n",
            "| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n",
            "| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n",
            "|===============================+======================+======================|\n",
            "|   0  Tesla K80           Off  | 00000000:00:04.0 Off |                    0 |\n",
            "| N/A   69C    P8    36W / 149W |      0MiB / 11441MiB |      0%      Default |\n",
            "+-------------------------------+----------------------+----------------------+\n",
            "                                                                               \n",
            "+-----------------------------------------------------------------------------+\n",
            "| Processes:                                                       GPU Memory |\n",
            "|  GPU       PID   Type   Process name                             Usage      |\n",
            "|=============================================================================|\n",
            "|  No running processes found                                                 |\n",
            "+-----------------------------------------------------------------------------+\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AFOKm27bZuki",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from numba import cuda\n",
        "import timeit\n",
        "import numpy as np\n",
        "from IPython.display import clear_output\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QYLcqw3hBC-j",
        "colab_type": "text"
      },
      "source": [
        "## Ejemplo 1: Multiplicación Matricial\n",
        "\n",
        "Dadas las matrices $A$ y $B$:\n",
        "\n",
        "$$\n",
        "A = \n",
        "\\begin{bmatrix}\n",
        "a_{11} & a_{12}\\\\\n",
        "a_{21} & a_{22}\n",
        "\\end{bmatrix}\n",
        ",\n",
        "B = \n",
        "\\begin{bmatrix}\n",
        "b_{11} & b_{12}\\\\\n",
        "b_{21} & b_{22}\n",
        "\\end{bmatrix}\n",
        "$$\n",
        "\n",
        "La multiplicacion matricial seria:\n",
        "\n",
        "$$\n",
        "C = A*B = \\begin{bmatrix}\n",
        "a_{11} & a_{12}\\\\\n",
        "a_{21} & a_{22}\n",
        "\\end{bmatrix} *\\begin{bmatrix}\n",
        "b_{11} & b_{12}\\\\\n",
        "b_{21} & b_{22}\n",
        "\\end{bmatrix}\n",
        "= \n",
        "\\begin{bmatrix}\n",
        "a_{11}b_{11}+a_{12}b_{21} & a_{11}b_{12}+a_{12}b_{22} \\\\\n",
        "a_{21}b_{11}+a_{22}b_{21} & a_{21}b_{12}+a_{22}b_{22}\n",
        "\\end{bmatrix}\n",
        "=\n",
        "\\begin{bmatrix}\n",
        "c_{11} & c_{12}\\\\\n",
        "c_{21} & c_{22}\n",
        "\\end{bmatrix}\n",
        "$$\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VOwwLpCeuC4F",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "@cuda.jit\n",
        "def matmul(A, B, C):\n",
        "    i, j = cuda.grid(2)\n",
        "    \n",
        "    if i < C.shape[0] and j < C.shape[1]:\n",
        "        tmp = 0\n",
        "        for k in range(A.shape[1]):\n",
        "            tmp += A[i, k] * B[k, j]\n",
        "        C[i, j] = tmp\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JUp6yOFmuJv8",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "dim = 1600\n",
        "A = np.ones([dim, dim])\n",
        "B = np.ones([dim, dim])\n",
        "C = np.zeros([dim, dim])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "v1xUCI9xui82",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "A_arr = cuda.to_device(A)\n",
        "B_arr = cuda.to_device(B)\n",
        "C_arr = cuda.to_device(C)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ww60a2kclhzC",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "threadsperblock = (16, 16)\n",
        "blockspergrid_x = np.int(C.shape[0] / threadsperblock[0]) + 1\n",
        "blockspergrid_y = np.int(C.shape[1] / threadsperblock[1]) + 1\n",
        "blockspergrid = (blockspergrid_x, blockspergrid_y)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ude6pxwA7kzh",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "matmul[blockspergrid, threadsperblock](A_arr, B_arr, C_arr)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Zr5_zW6EudAM",
        "colab_type": "code",
        "outputId": "c7bed908-86f6-4f37-aeb6-d2464d26dff6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 134
        }
      },
      "source": [
        "C"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0., 0., 0., ..., 0., 0., 0.],\n",
              "       [0., 0., 0., ..., 0., 0., 0.],\n",
              "       [0., 0., 0., ..., 0., 0., 0.],\n",
              "       ...,\n",
              "       [0., 0., 0., ..., 0., 0., 0.],\n",
              "       [0., 0., 0., ..., 0., 0., 0.],\n",
              "       [0., 0., 0., ..., 0., 0., 0.]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HhY0MRnCae7t",
        "colab_type": "code",
        "outputId": "a12c2181-5443-45e3-f462-aa3b054616b6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 134
        }
      },
      "source": [
        "C_arr.copy_to_host(C)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[1600., 1600., 1600., ..., 1600., 1600., 1600.],\n",
              "       [1600., 1600., 1600., ..., 1600., 1600., 1600.],\n",
              "       [1600., 1600., 1600., ..., 1600., 1600., 1600.],\n",
              "       ...,\n",
              "       [1600., 1600., 1600., ..., 1600., 1600., 1600.],\n",
              "       [1600., 1600., 1600., ..., 1600., 1600., 1600.],\n",
              "       [1600., 1600., 1600., ..., 1600., 1600., 1600.]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f7km_QUUvCWJ",
        "colab_type": "code",
        "outputId": "72b0f6dc-3376-4696-ed50-f5c24f21fc19",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 134
        }
      },
      "source": [
        "C"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[1600., 1600., 1600., ..., 1600., 1600., 1600.],\n",
              "       [1600., 1600., 1600., ..., 1600., 1600., 1600.],\n",
              "       [1600., 1600., 1600., ..., 1600., 1600., 1600.],\n",
              "       ...,\n",
              "       [1600., 1600., 1600., ..., 1600., 1600., 1600.],\n",
              "       [1600., 1600., 1600., ..., 1600., 1600., 1600.],\n",
              "       [1600., 1600., 1600., ..., 1600., 1600., 1600.]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MOW2x0WiAtAB",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def matmul_gpu():\n",
        "  matmul[blockspergrid, threadsperblock](A_arr,B_arr,C_arr)\n",
        "\n",
        "def matmul_cpu():\n",
        "  np.matmul(A, B)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2C3Y4--GB2b_",
        "colab_type": "code",
        "outputId": "dd76fd9c-0375-4554-ffe1-5be9c29815da",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 67
        }
      },
      "source": [
        "gpu_time = timeit.timeit(matmul_gpu, number=1)\n",
        "cpu_time = timeit.timeit(matmul_cpu, number=1)\n",
        "print(\"GPU time: {0}\".format(gpu_time))\n",
        "print(\"CPU time: {0}\".format(cpu_time))\n",
        "print(\"CPU/GPU: {0}\".format(cpu_time/gpu_time))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "GPU time: 0.0011548869999842282\n",
            "CPU time: 0.14470406600003116\n",
            "CPU/GPU: 125.29716413987457\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aARplLbgAuKg",
        "colab_type": "text"
      },
      "source": [
        "## Ejemplo 2: Propagación de señales acusticas\n",
        "Usando el metodo de diferencias finitas (FDM), se simulara la propagacion de una señal acustica mediante la ecuacion de onda siguiente:\n",
        "\n",
        "$$ \\partial_{tt} p(x,y,t) = c(x,y)^2 [\\partial_{xx} p(x,y,t)+ \\partial_{yy} p(x,y,t)] + s(x, y, t)\n",
        "$$\n",
        "\n",
        "Donde:\n",
        "- $p$ es la presion\n",
        "- $c$ es la velocidad de la onda en el medio\n",
        "- $x$ e $y$ son las dimensiones espaciales\n",
        "- $t$ es la dimension temporal.\n",
        "\n",
        "El primer paso para hallar una solución a la ecuacion diferencial es discretizar las variables involucradas. En este caso la presion, $p$, se discretiza con la siguiente expresion:\n",
        "\n",
        "$$\n",
        "p_{i, j}^n = p(i dx,j dy,n dt)\n",
        "$$\n",
        "\n",
        "Donde:\n",
        "- $dt$ es el tamaño del intervalo de tiempo en el cual se segmenta la dimension temporal\n",
        "- $dx$ e $dy$ son las longitudes de los espaciamientos en los que se segmentan las dimensiones espaciales\n",
        "- $i$ y $j$ son los indices de los ejes espaciales\n",
        "- $n$ es el indice del eje temporal.\n",
        "\n",
        "\n",
        "De la misma forma, se realiza las aproximaciones para las derivadas temporales y espaciales. \n",
        "\n",
        "$$\n",
        "\\partial_{tt} p \\approx \\frac{p_{i, j}^{n+1} - 2 p_{i, j}^n + p_{i, j}^{n-1}}{dt^2}\n",
        "$$\n",
        "\n",
        "$$\n",
        "\\partial_{xx} p \\approx \\frac{p_{i+1, j}^{n} - 2 p_{i, j}^n + p_{i-1, j}^{n}}{dx^2}\n",
        "$$\n",
        "\n",
        "$$\n",
        "\\partial_{yy} p \\approx \\frac{p_{i, j+1}^{n} - 2 p_{i, j}^n + p_{i, j-1}^{n}}{dy^2}\n",
        "$$\n",
        "\n",
        "Para hallar la solucion de la ecuacion diferencial, se reemplazan las aproximaciones en la ecuacion original y se despeja la variable $p_{i, j}^{n+1}$. Considerando que $dx$ es igual a $dy$, la expresion final es:\n",
        "\n",
        "$$\n",
        "p_{i, j}^{n+1} = [c_{i,j}\\frac{dt}{dx}]^2[p_{i+1, j}^{n} - 2 p_{i, j}^n + p_{i-1, j}^{n}+ p_{i, j+1}^{n} - 2 p_{i, j}^n + p_{i, j-1}^{n}] + 2p_{i,j} - p_{i,j}^{n-1} + s_{i,j}^{n}dt^{2}\n",
        "$$\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5BNPugg6Ch1n",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "@cuda.jit\n",
        "def FDM_forward(p_new, p, p_old, c, source, t, dx_arr, dt_arr):\n",
        "  # Definimos el tamaño de los pasos en el tiempo y espacio\n",
        "  dt = dt_arr[0]\n",
        "  dx = dx_arr[0]\n",
        "\n",
        "  # Posicion de la fuente\n",
        "  xs = 512\n",
        "  ys = 512\n",
        "\n",
        "  # Obtenemos la posicion global del hilo dentro de la grilla\n",
        "  i, j = cuda.grid(2)\n",
        "\n",
        "  # Condiciones de frontera\n",
        "  if i > 0 and i < (p_new.shape[0]-1) and j > 0 and j < (p_new.shape[1]-1):\n",
        "    # Calculo de las derivadas parciales\n",
        "    pxx = (p[i+1,j]+p[i-1,j] - 2*p[i,j])/dx**2\n",
        "    pyy = (p[i,j+1]+p[i,j-1] - 2*p[i,j])/dx**2\n",
        "\n",
        "    # Propagación de la presion\n",
        "    p_new[i,j] = (dt*c[i,j])**2*(pxx + pyy) + 2*p[i,j] - p_old[i,j]\n",
        "\n",
        "    \n",
        "    if i == xs and j == ys:\n",
        "      p_new[i,j] += source[t[0]]*dt**2\n",
        "  cuda.syncthreads()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "51ylFGYCxLQK",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "@cuda.jit\n",
        "def copy_p(p_new, p_old):\n",
        "  i, j = cuda.grid(2)\n",
        "  if i < p_new.shape[0] and j < p_new.shape[1]:\n",
        "    p_old[i,j] = p_new[i,j]\n",
        "  cuda.syncthreads()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OybkqRmdxPvs",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def source_function(nt, f_0, dt):\n",
        "  src = np.empty(nt + 1)\n",
        "  for it in range(nt):\n",
        "    src[it] = np.exp(-f_0 ** 2 * ((it - 100) * dt) ** 2)\n",
        "  src = np.diff(src) / dt\n",
        "  src[nt - 1] = 0\n",
        "  return src"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9qgBKViM8rM7",
        "colab_type": "code",
        "outputId": "9cbe1c52-d033-4304-e580-881cee856490",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        }
      },
      "source": [
        "src = source_function(500, 150, 0.0007)\n",
        "plt.plot(src)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7ffb05f9b128>]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fzgHdEXHxTy3",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def simulation_gpu(plot=False):\n",
        "  # Definimos las variables iniciales\n",
        "  dim = 1024\n",
        "  c_max = 3000\n",
        "  p = np.zeros([dim, dim])\n",
        "  p_old = np.zeros([dim, dim])\n",
        "  p_new = np.zeros([dim, dim])\n",
        "  c = c_max*np.ones([dim, dim])\n",
        "  f_0 = 150\n",
        "  dx = (c_max/f_0)/5\n",
        "  dt = 0.5*dx/c_max\n",
        "  nt = 5000\n",
        "  source = source_function(nt, f_0, dt)\n",
        "\n",
        "  if plot:\n",
        "    fig_img = plt.imshow(p, vmin=-100, vmax=100, cmap=\"seismic\")\n",
        "\n",
        "\n",
        "  # Definimos el numero de hilos por bloque y el numero de bloques totales\n",
        "  threadsperblock = (16, 16)\n",
        "  blockspergrid_x = np.int(p.shape[0] / threadsperblock[0] + 1)\n",
        "  blockspergrid_y = np.int(p.shape[1] / threadsperblock[1] + 1)\n",
        "  blockspergrid = (blockspergrid_x, blockspergrid_y)\n",
        "\n",
        "  # Se copian las variables del HOST al DEVICE\n",
        "  p_arr = cuda.to_device(p)\n",
        "  pold_arr = cuda.to_device(p_old)\n",
        "  pnew_arr = cuda.to_device(p_new)\n",
        "  c_arr = cuda.to_device(c)\n",
        "  s_arr = cuda.to_device(source)\n",
        "  dt_arr = cuda.to_device(dt)\n",
        "  dx_arr = cuda.to_device(dx)\n",
        "\n",
        "  for k in range(nt):\n",
        "    t_arr = cuda.to_device(k)\n",
        "    FDM_forward[blockspergrid, threadsperblock](pnew_arr, p_arr, pold_arr, c_arr, s_arr, t_arr, dx_arr, dt_arr)\n",
        "    copy_p[blockspergrid, threadsperblock](p_arr, pold_arr)\n",
        "    copy_p[blockspergrid, threadsperblock](pnew_arr, p_arr)\n",
        "    \n",
        "    if k % 100 == 0 and plot:\n",
        "      p_arr.copy_to_host(p)\n",
        "      plt.imshow(p, vmin=-p.max()/2, vmax=p.max()/2, cmap=\"seismic\")\n",
        "      plt.title(\"simulation: {0} -- pmax: {1}\".format(k, p.max()))\n",
        "      plt.show()\n",
        "      #plt.draw()\n",
        "      plt.pause(0.05)\n",
        "      clear_output(wait=True)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jCZ33gsFUTq6",
        "colab_type": "code",
        "outputId": "53781e36-da4c-46a3-e83a-e1a6a72dc99e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        }
      },
      "source": [
        "simulation_gpu(True)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "U06rFi81UqJ9",
        "colab_type": "code",
        "outputId": "349890fd-e003-4c38-abac-dda75d5b5a10",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "print(timeit.timeit(simulation_gpu, number=10)/10)3"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "9.515541317600015\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}