Final8.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Final8",
      "version": "0.3.2",
      "provenance": [],
      "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/github/4en1x/diploma/blob/master/Final8.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "metadata": {
        "id": "7veJyTIe7rFK",
        "colab_type": "code",
        "outputId": "e263176e-4d7f-4810-8272-480a6f839695",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 190
        }
      },
      "cell_type": "code",
      "source": [
        "!pip install git+git://github.com/andreinechaev/nvcc4jupyter.git\n",
        "%load_ext nvcc_plugin\n",
        "!cd /usr/src\n",
        "!ln -sfn linux-headers-2.6.35-28-generic linux"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting git+git://github.com/andreinechaev/nvcc4jupyter.git\n",
            "  Cloning git://github.com/andreinechaev/nvcc4jupyter.git to /tmp/pip-req-build-9a__qslt\n",
            "Requirement already satisfied (use --upgrade to upgrade): NVCCPlugin==0.0.2 from git+git://github.com/andreinechaev/nvcc4jupyter.git in /usr/local/lib/python3.6/dist-packages\n",
            "Building wheels for collected packages: NVCCPlugin\n",
            "  Building wheel for NVCCPlugin (setup.py) ... \u001b[?25ldone\n",
            "\u001b[?25h  Stored in directory: /tmp/pip-ephem-wheel-cache-431y_0h_/wheels/10/c2/05/ca241da37bff77d60d31a9174f988109c61ba989e4d4650516\n",
            "Successfully built NVCCPlugin\n",
            "directory /content/src already exists\n",
            "Out bin /content/result.out\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "eOojdQP47xCL",
        "colab_type": "code",
        "outputId": "b3d32966-f475-4b38-87f5-7d872d9d751f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        }
      },
      "cell_type": "code",
      "source": [
        "%%cu\n",
        "#include <cuda_runtime.h>\n",
        "#include \"device_launch_parameters.h\"\n",
        "#include <stdlib.h>\n",
        "#include <stdio.h>\n",
        "#include \"linux/kernel.h\"\n",
        "#include <iostream>\n",
        "#include <ctime>\n",
        "#include <iomanip>\n",
        "#include <fstream>\n",
        "#include <math.h>\n",
        "#include <cmath>\n",
        "\n",
        "using namespace std;\n",
        "\n",
        "#define ITERATIONS 2000000\n",
        "#define N 256\n",
        "\n",
        "#define e 2.7182818\n",
        "\n",
        "#define blockZSize 4\n",
        "\n",
        "#define blockXSize 16\n",
        "#define blockYSize 16\n",
        "\n",
        "#define tileXSize 4\n",
        "#define tileYSize 4\n",
        "\n",
        "#define extN (N + 2 * blockZSize)\n",
        "#define Dx 1\n",
        "#define h ((float) Dx / (N + 1))\n",
        "#define h2 powf(h, 2.0)\n",
        "\n",
        "__device__\n",
        "float f(int i, int j) {\n",
        "\treturn 2 * powf(e, i * h + j * h);\n",
        "}\n",
        "\n",
        "float u_t(float x) {\n",
        "\treturn powf(e, x);\n",
        "}\n",
        "\n",
        "float u_l(float x) {\n",
        "\treturn powf(e, x);\n",
        "}\n",
        "\n",
        "float u_r(float x) {\n",
        "\treturn powf(e, x + 1);\n",
        "}\n",
        "\n",
        "float u_d(float x) {\n",
        "\treturn powf(e, x + 1);\n",
        "}\n",
        "\n",
        "\n",
        "__device__\n",
        "void tile(float* matrix, int iGl, int jGl, bool isBlack) {\n",
        "\tint iBegin = iGl * blockXSize + threadIdx.x * tileXSize + blockZSize;\n",
        "\tint jBegin = jGl * blockYSize + threadIdx.y * tileYSize + blockZSize;\n",
        "\n",
        "  int N1 = tileXSize + blockZSize * 2;\n",
        "  int N2 = tileYSize + blockZSize * 2;\n",
        "  \n",
        "\tfloat helper[12 * 12];\n",
        "    \n",
        "  for(int i = 0; i < N1; i++) {\n",
        "\t\tfor(int j = 0; j < N2; j++) {\n",
        "\t\t\thelper[i * N1 + j] = matrix[extN * (iBegin - blockZSize + i) + jBegin - blockZSize + j];\n",
        "\t\t}\n",
        "\t}\n",
        "  \n",
        "  int iSplit = iBegin - 2 * blockZSize + 1;\n",
        "  int jSplit = jBegin - 2 * blockZSize + 1;\n",
        "    \n",
        "  for(int k = 1; k < blockZSize; k++) {\n",
        "\t\tfor (int i = k; i < k + tileXSize; i++) {\n",
        "       \n",
        "\t\t  for (int j = k; j < k + tileXSize; j += 1) {\n",
        "        int b = N1 * i + j;\n",
        "\t\t\t  helper[b] = 0.25 * (\n",
        "            helper[b + N1]\n",
        "          + helper[b - N1]\n",
        "          + helper[b + 1]\n",
        "\t\t\t\t  + helper[b - 1] \n",
        "          - h2 * f(iSplit + i, jSplit + j)\n",
        "        );\n",
        "\t\t  }\n",
        "\t  }\n",
        "\t}\n",
        "\n",
        "  //printf(\"iGl : %d, %d, %d, %d, %d, %d, %d \\n\", iGl, jGl, threadIdx.x, threadIdx.y, isBlack, iBegin, jBegin);\n",
        "\n",
        "\tfor (int i = iBegin; i < iBegin + tileXSize; i++) {\n",
        "\t\t\n",
        "\t\tfor (int j = jBegin; j < jBegin + tileYSize; j += 1) {\n",
        "      int helperIndex = N1 * (i - iBegin + blockZSize) + j - jBegin + blockZSize;\n",
        "\t\t\tmatrix[extN * i + j] = 0.25 * (\n",
        "          helper[helperIndex + N1]\n",
        "        + helper[helperIndex - N1]\n",
        "        + helper[helperIndex + 1]\n",
        "\t\t\t\t+ helper[helperIndex - 1] \n",
        "        - h2 * f(i - blockZSize + 1, j - blockZSize + 1)\n",
        "      );\n",
        "\t\t}\n",
        "\t}\n",
        "}\n",
        "\n",
        "__global__\n",
        "void calculateBlack(float* matrix) {\n",
        "\ttile(matrix, blockIdx.x, blockIdx.y, true);\n",
        "}\n",
        "\n",
        "__global__\n",
        "void calculateRed(float* matrix) {\n",
        "\ttile(matrix, blockIdx.x, blockIdx.y, false);\n",
        "}\n",
        "\n",
        "float func(int i, int j) {\n",
        "\treturn powf(e, i * h + j * h);\n",
        "}\n",
        "\n",
        "float* createMatrix(int row, int col) {\n",
        "\tfloat* matrix = new float[row * col];\n",
        "\n",
        "\tfor(int i = 0; i < row; i++) {\n",
        "\t\tfor(int j = 0; j < col; j++) {\n",
        "\t\t\tmatrix[i * row + j] = 0;\n",
        "\t\t}\n",
        "\t}\n",
        "\n",
        "  for(int k = 0; k < blockZSize; k++) {\n",
        "    for(int i = k; i < row - k; i++) {\n",
        "      matrix[k * row + i] = u_t((i - blockZSize + 1) * h);\n",
        "      matrix[(row - 1 - k) * row + i] = u_d((i - blockZSize + 1) * h);\n",
        "    }\n",
        "\t}\n",
        "\n",
        "  for(int k = 0; k < blockZSize; k++) {\n",
        "    for(int j = k; j < col - k; j++) {\n",
        "      matrix[j * col + k] = u_l((j - blockZSize + 1) * h);\n",
        "      matrix[j * col + row - 1 - k] = u_r((j - blockZSize + 1) * h);\n",
        "\t  }\n",
        "  }\n",
        "    \n",
        "\treturn matrix;\n",
        "}\n",
        "\n",
        "float calculateDifference(float* matrix1) {\n",
        "\tfloat tolerance = 0.0f;\n",
        "  for (int i = blockZSize; i < extN - blockZSize; i++) {\n",
        "\t\tfor (int j = blockZSize; j < extN - blockZSize; j++) {\n",
        "      float fCurrent = func(i - blockZSize + 1, j - blockZSize + 1);\n",
        "      if(abs(matrix1[i * extN + j] - fCurrent) > tolerance) {\n",
        "          tolerance = abs(matrix1[i * extN + j] - fCurrent);\n",
        "      } \n",
        "\t\t}\n",
        "\t}\n",
        "\treturn tolerance;\n",
        "}\n",
        "\n",
        "\n",
        "int main() {\n",
        "  ofstream myfile;\n",
        "  ofstream myfiletimes;\n",
        "  myfile.open(\"diff_8_4.txt\");\n",
        "  myfiletimes.open(\"diff_8_4_times.txt\");\n",
        "\tfloat* matrixHost = createMatrix(extN, extN);\n",
        "\tfloat* matrixDevice;\n",
        "\n",
        "\tint matrixSize = sizeof(float) * extN * extN;\n",
        "\n",
        "\tcudaMalloc((void**)&matrixDevice, matrixSize);\n",
        "\tcudaMemcpy(matrixDevice, matrixHost, matrixSize, cudaMemcpyHostToDevice);\n",
        "\n",
        "\tlong startTime = clock();\n",
        "\tdim3 blocks(N / blockXSize, N / blockYSize);\n",
        "\tdim3 threads(blockXSize / tileXSize, blockYSize / tileYSize);\n",
        "\n",
        "\tfor (size_t i = 0; i < ITERATIONS; i++) {\n",
        "\t\tcalculateBlack <<< blocks, threads >>>(matrixDevice);\n",
        "\t\tcudaDeviceSynchronize();\n",
        "\t\tcudaPeekAtLastError();\n",
        "\n",
        "    if(i%1000 == 0) {\n",
        "        cudaMemcpy(matrixHost, matrixDevice, matrixSize, cudaMemcpyDeviceToHost);\n",
        "        float diff = calculateDifference(matrixHost);\n",
        "        cout << \" Difference: \" << diff << \"iter: \" << i;\n",
        "        myfile << diff << \" \";\n",
        "        myfiletimes << clock() - startTime << \" \";\n",
        "        if(diff < 0.01) break;\n",
        "    }  \n",
        "\t}\n",
        "  \n",
        "\tstd::cout << \"Calculation take time: \" << clock() - startTime << std::endl;\n",
        "  myfile << clock() - startTime << \" \";\n",
        "    \n",
        "  cout << \" Difference: \" << calculateDifference(matrixHost);\n",
        "\tcudaError_t cudaResult = cudaMemcpy(matrixHost, matrixDevice, matrixSize, cudaMemcpyDeviceToHost);\n",
        "\n",
        "\tif (cudaResult != cudaSuccess) {\n",
        "\t\tstd::cout << \"Cuda error: \" << cudaResult << std::endl;\n",
        "\t}\n",
        "\n",
        "    myfile.close();\n",
        "    myfiletimes.close();\n",
        "\treturn 0;\n",
        "}"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "' Difference: 7.27506iter: 0 Difference: 4.03278iter: 1000 Difference: 3.28314iter: 2000 Difference: 2.81016iter: 3000 Difference: 2.4259iter: 4000 Difference: 2.08669iter: 5000 Difference: 1.78861iter: 6000 Difference: 1.52956iter: 7000 Difference: 1.30606iter: 8000 Difference: 1.11491iter: 9000 Difference: 0.951559iter: 10000 Difference: 0.811921iter: 11000 Difference: 0.692767iter: 12000 Difference: 0.591054iter: 13000 Difference: 0.50431iter: 14000 Difference: 0.430302iter: 15000 Difference: 0.367125iter: 16000 Difference: 0.31318iter: 17000 Difference: 0.267121iter: 18000 Difference: 0.227795iter: 19000 Difference: 0.194219iter: 20000 Difference: 0.165552iter: 21000 Difference: 0.141076iter: 22000 Difference: 0.12018iter: 23000 Difference: 0.102341iter: 24000 Difference: 0.0871098iter: 25000 Difference: 0.0741065iter: 26000 Difference: 0.0630043iter: 27000 Difference: 0.0535259iter: 28000 Difference: 0.045433iter: 29000 Difference: 0.0385265iter: 30000 Difference: 0.032629iter: 31000 Difference: 0.0276008iter: 32000 Difference: 0.0233111iter: 33000 Difference: 0.0196433iter: 34000 Difference: 0.0165231iter: 35000 Difference: 0.0138507iter: 36000 Difference: 0.0115759iter: 37000 Difference: 0.00963473iter: 38000Calculation take time: 8921858\\n Difference: 0.00963473'"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "metadata": {
        "id": "2WopvbTPF4jz",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 393
        },
        "outputId": "28b4c841-704d-42f6-b45c-6c4e571181fa"
      },
      "cell_type": "code",
      "source": [
        "# libraries and data\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import os\n",
        "\n",
        "k = 0\n",
        "\n",
        "c1 = list(map(lambda x: float(x), open(\"diff_8_1.txt\", 'r').read().split(' ')[:-2]))[k:]\n",
        "c2 = list(map(lambda x: float(x), open(\"diff_8_2.txt\", 'r').read().split(' ')[:-2]))[k:]\n",
        "c3 = list(map(lambda x: float(x), open(\"diff_8_3.txt\", 'r').read().split(' ')[:-2]))[k:]\n",
        "c4 = list(map(lambda x: float(x), open(\"diff_8_4.txt\", 'r').read().split(' ')[:-2]))[k:]\n",
        "\n",
        "t1 = list(map(lambda x: float(x) / 1000000, open(\"diff_8_1_times.txt\", 'r').read().split(' ')[:-1]))[k:]\n",
        "t2 = list(map(lambda x: float(x) / 1000000, open(\"diff_8_2_times.txt\", 'r').read().split(' ')[:-1]))[k:]\n",
        "t3 = list(map(lambda x: float(x) / 1000000, open(\"diff_8_3_times.txt\", 'r').read().split(' ')[:-1]))[k:]\n",
        "t4 = list(map(lambda x: float(x) / 1000000, open(\"diff_8_4_times.txt\", 'r').read().split(' ')[:-1]))[k:]\n",
        "\n",
        "\n",
        "\n",
        "# style\n",
        "plt.style.use('seaborn-darkgrid')\n",
        " \n",
        "# create a color palette\n",
        "palette = plt.get_cmap('Set1')\n",
        " \n",
        "# multiple line plot\n",
        "\n",
        "plt.plot(t1, c1,  marker='', color=palette(1), linewidth=1, alpha=0.9, label='zIndex = 1')  \n",
        "plt.plot(t2, c2,  marker='', color=palette(2), linewidth=1, alpha=0.9, label='zIndex = 2')  \n",
        "plt.plot(t3, c3,  marker='', color=palette(3), linewidth=1, alpha=0.9, label='zIndex = 3')  \n",
        "plt.plot(t4, c4,  marker='', color=palette(4), linewidth=1, alpha=0.9, label='zIndex = 4')  \n",
        "\n",
        "# Add legend\n",
        "plt.legend(loc=2, ncol=2)\n",
        "plt.xlim(left=0)\n",
        "# Add titles\n",
        "plt.title(\"Нормы на соседних итерациях, размер матрицы = 256\", loc='left', fontsize=12, fontweight=0, color='orange')\n",
        "plt.xlabel(\"Секунды\")\n",
        "plt.ylabel(\"Норма матрицы\")"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0, 0.5, 'Норма матрицы')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 34
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAecAAAFnCAYAAACcvYGMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nosJIOAshhBAV5v8DgC5UJFcawNEAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 576x396 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}