low-level/dgemm/eval/juwels/dgemm-juwels.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# DGEMM on JUWELS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import matplotlib.ticker as mtick\n",
    "%matplotlib inline\n",
    "sns.set(\n",
    "    palette=sns.color_palette(\"Paired\"),\n",
    "    rc={\n",
    "        \"figure.figsize\": (24, 10)\n",
    "    },\n",
    "    color_codes=False\n",
    ")\n",
    "from scipy.stats import linregress\n",
    "SYSTEM=\"juwels\"\n",
    "DATADIR=\"../../data/\"+SYSTEM+\"/\"\n",
    "plt.rcParams['figure.figsize'] = [24, 10]  # Bug; needed to be set twice"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.rcParams['figure.figsize'] = [24, 10]  # Bug; needed to be set twice"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "frequency = 2.7\n",
    "turbofrequency = 2.7  # running with --disable-turbomode\n",
    "sockets = 2\n",
    "corespersocket = 24\n",
    "flopspercycle = 32\n",
    "\n",
    "singlecorepeak = 1 * turbofrequency * flopspercycle\n",
    "theoreticalpeak= sockets * corespersocket * frequency * flopspercycle\n",
    "\n",
    "def theoricalPeak(core):\n",
    "    # Turbo Bost on one core\n",
    "    if core == 1:\n",
    "        return singlecorepeak\n",
    "    else:\n",
    "        return core * frequency * flopspercycle"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2021-02-25"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>jobid</th>\n",
       "      <th>systemname</th>\n",
       "      <th>numlib</th>\n",
       "      <th>error_code</th>\n",
       "      <th>runtime[s]</th>\n",
       "      <th>nodes</th>\n",
       "      <th>taskspernode</th>\n",
       "      <th>threadspertask</th>\n",
       "      <th>NLAST</th>\n",
       "      <th>time[s]</th>\n",
       "      <th>GFLOPS</th>\n",
       "      <th>GFLOPSpcore</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3418330</td>\n",
       "      <td>juwels</td>\n",
       "      <td>BLIS/0.8</td>\n",
       "      <td>0</td>\n",
       "      <td>136.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>8000</td>\n",
       "      <td>13.2800</td>\n",
       "      <td>75.69422</td>\n",
       "      <td>77.1107</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3418339</td>\n",
       "      <td>juwels</td>\n",
       "      <td>BLIS/0.8</td>\n",
       "      <td>0</td>\n",
       "      <td>139.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>19000</td>\n",
       "      <td>9.8512</td>\n",
       "      <td>1398.33130</td>\n",
       "      <td>58.0217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3418343</td>\n",
       "      <td>juwels</td>\n",
       "      <td>BLIS/0.8</td>\n",
       "      <td>0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>48</td>\n",
       "      <td>24000</td>\n",
       "      <td>10.6860</td>\n",
       "      <td>2502.58886</td>\n",
       "      <td>53.9042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3418364</td>\n",
       "      <td>juwels</td>\n",
       "      <td>OpenBLAS/0.3.12</td>\n",
       "      <td>0</td>\n",
       "      <td>129.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>8000</td>\n",
       "      <td>12.4530</td>\n",
       "      <td>80.71640</td>\n",
       "      <td>82.2322</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3418365</td>\n",
       "      <td>juwels</td>\n",
       "      <td>OpenBLAS/0.3.12</td>\n",
       "      <td>0</td>\n",
       "      <td>150.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>19000</td>\n",
       "      <td>10.6840</td>\n",
       "      <td>1072.96148</td>\n",
       "      <td>53.4989</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3418373</td>\n",
       "      <td>juwels</td>\n",
       "      <td>OpenBLAS/0.3.12</td>\n",
       "      <td>0</td>\n",
       "      <td>197.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>48</td>\n",
       "      <td>24000</td>\n",
       "      <td>12.8530</td>\n",
       "      <td>1835.95638</td>\n",
       "      <td>44.8136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>3418346</td>\n",
       "      <td>juwels</td>\n",
       "      <td>imkl/2020.4.304</td>\n",
       "      <td>0</td>\n",
       "      <td>132.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>8000</td>\n",
       "      <td>12.8320</td>\n",
       "      <td>77.56504</td>\n",
       "      <td>79.7997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>3418351</td>\n",
       "      <td>juwels</td>\n",
       "      <td>imkl/2020.4.304</td>\n",
       "      <td>0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>19000</td>\n",
       "      <td>8.6335</td>\n",
       "      <td>1590.26064</td>\n",
       "      <td>66.2051</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>3418357</td>\n",
       "      <td>juwels</td>\n",
       "      <td>imkl/2020.4.304</td>\n",
       "      <td>0</td>\n",
       "      <td>165.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>48</td>\n",
       "      <td>24000</td>\n",
       "      <td>9.1166</td>\n",
       "      <td>3086.01368</td>\n",
       "      <td>63.1812</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     jobid systemname           numlib  error_code  runtime[s]  nodes  \\\n",
       "0  3418330     juwels         BLIS/0.8           0       136.0      1   \n",
       "1  3418339     juwels         BLIS/0.8           0       139.0      1   \n",
       "2  3418343     juwels         BLIS/0.8           0       180.0      1   \n",
       "3  3418364     juwels  OpenBLAS/0.3.12           0       129.0      1   \n",
       "4  3418365     juwels  OpenBLAS/0.3.12           0       150.0      1   \n",
       "5  3418373     juwels  OpenBLAS/0.3.12           0       197.0      1   \n",
       "6  3418346     juwels  imkl/2020.4.304           0       132.0      1   \n",
       "7  3418351     juwels  imkl/2020.4.304           0       125.0      1   \n",
       "8  3418357     juwels  imkl/2020.4.304           0       165.0      1   \n",
       "\n",
       "   taskspernode  threadspertask  NLAST  time[s]      GFLOPS  GFLOPSpcore  \n",
       "0             1               1   8000  13.2800    75.69422      77.1107  \n",
       "1             1              24  19000   9.8512  1398.33130      58.0217  \n",
       "2             1              48  24000  10.6860  2502.58886      53.9042  \n",
       "3             1               1   8000  12.4530    80.71640      82.2322  \n",
       "4             1              24  19000  10.6840  1072.96148      53.4989  \n",
       "5             1              48  24000  12.8530  1835.95638      44.8136  \n",
       "6             1               1   8000  12.8320    77.56504      79.7997  \n",
       "7             1              24  19000   8.6335  1590.26064      66.2051  \n",
       "8             1              48  24000   9.1166  3086.01368      63.1812  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "INPUT_FILE=\"skylake_2021-02-25.csv\"\n",
    "df = pd.read_csv(DATADIR+INPUT_FILE)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1728x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.pivot_table(columns=[\"numlib\"], index=[\"threadspertask\",\"NLAST\"], values=\"GFLOPS\").plot(kind=\"bar\", width=0.5, logy=False, title=\"DGEMM on JUWELS.\",ylim=(0,theoreticalpeak));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.axis.XTick at 0x7f9c1a5c5850>,\n",
       " <matplotlib.axis.XTick at 0x7f9c1a20f670>,\n",
       " <matplotlib.axis.XTick at 0x7f9c1aabfb50>]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1008 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "sns.set(palette=sns.color_palette(\"Accent\"))\n",
    "plt.rcParams['figure.figsize'] = [20, 14]\n",
    "table=df.pivot_table(columns=[\"numlib\"],index=[\"threadspertask\"], values=[\"GFLOPSpcore\"])\n",
    "plot=table.plot(kind=\"line\", logy=False, title=\"DGEMM on JUWELS: GFLOPS per core\");\n",
    "plot.set_ylim(0,singlecorepeak)\n",
    "plot.set_xticks([1,24,48])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1008 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['theoricalPeak'] = [theoricalPeak(int(x)) for x in df[['threadspertask']].values]\n",
    "df['Efficiency'] = df.GFLOPS / df.theoricalPeak * 100\n",
    "\n",
    "ax = df.pivot_table(columns=[\"numlib\"], index=[\"threadspertask\",\"NLAST\"], values=\"Efficiency\").plot(kind=\"bar\", width=0.5, logy=False, title=\"DGEMM Efficiency on JUWELS.\",ylim=(0,100));\n",
    "ax.yaxis.set_major_formatter(mtick.PercentFormatter())"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}