java-class-load-performance.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I've started on something of an experiment to explore how I can improve Clojure start-up time. I was curious what impact compression level has on start-up latency for JVM applications. A simple clojure application was created `(+ 1 2 3)` and packaged into an uberjar. Four Zip compression levels were tested 0, 1, 6, and 9. The uberjar was unpacked and then repacked using the linux command line zip tool as follows:\n",
    "\n",
    "```\n",
    "zip -r -0 ../main_c0.jar *\n",
    "zip -r -1 ../main_c1.jar *\n",
    "zip -r -6 ../main_c6.jar *\n",
    "zip -r -9 ../main_c9.jar *\n",
    "```\n",
    "\n",
    "All applications, screen-locks, system sleep, and anti-virus were closed to minimise interference. A test was devised which executes the 4 uberjars 100 times each. I will refer to each uberjars execution set as a run. Every 10 executions a sleep of 30s is used to settle the system as well as writing an 8MB file to flush disk caches. See the test runner shell script below for exact details. The test was left to execute and the results from the first run were discarded as a \"warm-up\" cycle.\n",
    "\n",
    "## Hypothesis\n",
    "\n",
    "- H0 - compression level will not have an impact on application start-up time.\n",
    "- Ha - compression level will have an impact on application start-up time.\n",
    "- p-level - 0.1.\n",
    "\n",
    "## Results\n",
    "\n",
    "The results are below in ms for the execution time.\n",
    "\n",
    "### Run 2 Results\n",
    "\n",
    "| (ms)   | C0  | C1  | C6  | C9  |\n",
    "|--------|-----|-----|-----|-----|\n",
    "| 25PCTL | 570 | 616 | 613 | 613 |\n",
    "| 75PCTL | 600 | 655 | 643 | 643 |\n",
    "| Max    | 650 | 730 | 702 | 700 |\n",
    "\n",
    "Overall 0 compression appears to shave 50ms from load time. Furthermore for the sample size it appears to be statistically significant.\n",
    "\n",
    "## Next Steps\n",
    "\n",
    "Compressing with level 0 has an obvious tradeoff for the distribution of artefact over the internet. I'd be curious to see what impact a faster algorithm such as Snappy or LZ4 has on class load performance.\n",
    "\n",
    "For the next test I'll see what happens when the java byte code is loaded into a byte array and a custom class loader is used. This will be a baseline measurement to demonstrate whether I can write a ClassLoader which is faster than the default JAR loader.\n",
    "\n",
    "### Notes\n",
    "\n",
    "For future tests I would probably reduce the number of runs in the warm-up to 10 or 20 executions per jar rather than 100.\n",
    "\n",
    "### Test Runner\n",
    "\n",
    "```\n",
    "#!/bin/sh\n",
    "\n",
    "JARS=\"main_c0.jar main_c1.jar main_c6.jar main_c9.jar\"\n",
    "\n",
    "for JAR in $JARS; do\n",
    "\tB=0\n",
    "\tOUT=\"${JAR}-results-r1.txt\"\t\n",
    "\tSORTED=\"${JAR}-sorted-results-r1.txt\"\n",
    "\techo \"starting test ${OUT}\"\n",
    "\twhile [ $B -lt 10 ]; do\n",
    "\t\t# try to flush the disk cache\n",
    "\t\tdd if=/dev/zero of=/tmp/rando.txt bs=4096 count=2048\n",
    "\t\trm /tmp/rando.txt\n",
    "\t\tA=0\n",
    "\t\twhile [ $A -lt 10 ]; do\n",
    "\t\t\tjava -jar $JAR >> \"$OUT\"\n",
    "\t\t\tA=`expr $A + 1`\n",
    "\t\tdone\n",
    "\n",
    "\t\tsleep 30\n",
    "\t\tB=`expr $B + 1`\n",
    "\tdone\n",
    "\tcat $OUT | sort -n > $SORTED\n",
    "done\n",
    "\n",
    "for JAR in $JARS; do\n",
    "\tB=0\n",
    "\tOUT=\"${JAR}-results-r2.txt\"\t\n",
    "\tSORTED=\"${JAR}-sorted-results-r2.txt\"\n",
    "\techo \"starting test ${OUT}\"\n",
    "\twhile [ $B -lt 10 ]; do\n",
    "\t\t# try to flush the disk cache\n",
    "\t\tdd if=/dev/zero of=/tmp/rando.txt bs=4096 count=2048\n",
    "\t\trm /tmp/rando.txt\n",
    "\t\tA=0\n",
    "\t\twhile [ $A -lt 10 ]; do\n",
    "\t\t\tjava -jar $JAR >> \"$OUT\"\n",
    "\t\t\tA=`expr $A + 1`\n",
    "\t\tdone\n",
    "\n",
    "\t\tsleep 30\n",
    "\t\tB=`expr $B + 1`\n",
    "\tdone\n",
    "\tcat $OUT | sort -n > $SORTED\n",
    "done\n",
    "\n",
    "for JAR in $JARS; do\n",
    "\tB=0\n",
    "\tOUT=\"${JAR}-results-r3.txt\"\t\n",
    "\tSORTED=\"${JAR}-sorted-results-r3.txt\"\n",
    "\techo \"starting test ${OUT}\"\n",
    "\twhile [ $B -lt 10 ]; do\n",
    "\t\t# try to flush the disk cache\n",
    "\t\tdd if=/dev/zero of=/tmp/rando.txt bs=4096 count=2048\n",
    "\t\trm /tmp/rando.txt\n",
    "\t\tA=0\n",
    "\t\twhile [ $A -lt 10 ]; do\n",
    "\t\t\tjava -jar $JAR >> \"$OUT\"\n",
    "\t\t\tA=`expr $A + 1`\n",
    "\t\tdone\n",
    "\n",
    "\t\tsleep 30\n",
    "\t\tB=`expr $B + 1`\n",
    "\tdone\n",
    "\tcat $OUT | sort -n > $SORTED\n",
    "done\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import scipy.stats as stats\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import math\n",
    "\n",
    "rounds = 2 # ignore the first round\n",
    "levels = ['0', '1', '6', '9']\n",
    "runs = []\n",
    "for i in range(len(levels)*2):\n",
    "    col = i % len(levels)\n",
    "    run = 2 + math.floor(i/len(levels))\n",
    "    level = levels[col]\n",
    "    \n",
    "    filename = 'main_c{}.jar-results-r{}.txt'.format(level, run)\n",
    "    df = pd.read_csv(filename, header=None)\n",
    "    df.columns = [level]\n",
    "    if col == 0:\n",
    "        runs.append(df)\n",
    "    else:\n",
    "        runs[run-2] = pd.concat([runs[run-2], df], axis=1, join='inner')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "significant: True\n",
      "significant: True\n"
     ]
    }
   ],
   "source": [
    "second = runs[0]\n",
    "third = runs[1]\n",
    "\n",
    "result = stats.ttest_ind(second['0'].as_matrix(), second['1'].as_matrix(), equal_var = False)\n",
    "print(\"significant:\", result.pvalue < 0.01)\n",
    "result = stats.ttest_ind(third['0'].as_matrix(), third['1'].as_matrix(), equal_var = False)\n",
    "print(\"significant:\", result.pvalue < 0.01)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA6wAAAJJCAYAAACuxbrBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x115077588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>6</th>\n",
       "      <th>9</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.000000e+02</td>\n",
       "      <td>1.000000e+02</td>\n",
       "      <td>1.000000e+02</td>\n",
       "      <td>1.000000e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>5.881217e+08</td>\n",
       "      <td>6.383170e+08</td>\n",
       "      <td>6.315500e+08</td>\n",
       "      <td>6.359029e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.315387e+07</td>\n",
       "      <td>2.728576e+07</td>\n",
       "      <td>2.389348e+07</td>\n",
       "      <td>2.511420e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>5.482715e+08</td>\n",
       "      <td>5.976797e+08</td>\n",
       "      <td>5.886895e+08</td>\n",
       "      <td>5.992121e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.702272e+08</td>\n",
       "      <td>6.161399e+08</td>\n",
       "      <td>6.131123e+08</td>\n",
       "      <td>6.137505e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5.857689e+08</td>\n",
       "      <td>6.348973e+08</td>\n",
       "      <td>6.281904e+08</td>\n",
       "      <td>6.318848e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.007704e+08</td>\n",
       "      <td>6.557159e+08</td>\n",
       "      <td>6.432410e+08</td>\n",
       "      <td>6.535822e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>6.500378e+08</td>\n",
       "      <td>7.302775e+08</td>\n",
       "      <td>7.024601e+08</td>\n",
       "      <td>7.001149e+08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  0             1             6             9\n",
       "count  1.000000e+02  1.000000e+02  1.000000e+02  1.000000e+02\n",
       "mean   5.881217e+08  6.383170e+08  6.315500e+08  6.359029e+08\n",
       "std    2.315387e+07  2.728576e+07  2.389348e+07  2.511420e+07\n",
       "min    5.482715e+08  5.976797e+08  5.886895e+08  5.992121e+08\n",
       "25%    5.702272e+08  6.161399e+08  6.131123e+08  6.137505e+08\n",
       "50%    5.857689e+08  6.348973e+08  6.281904e+08  6.318848e+08\n",
       "75%    6.007704e+08  6.557159e+08  6.432410e+08  6.535822e+08\n",
       "max    6.500378e+08  7.302775e+08  7.024601e+08  7.001149e+08"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,10))\n",
    "second.boxplot()\n",
    "plt.show()\n",
    "second.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA6UAAAJJCAYAAABS0JELAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n0Mz/D8Wx1z8WXG4gAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x110e75828>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,10))\n",
    "plt.scatter(second.index, second['0'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>6</th>\n",
       "      <th>9</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.000000e+02</td>\n",
       "      <td>1.000000e+02</td>\n",
       "      <td>1.000000e+02</td>\n",
       "      <td>1.000000e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>5.930233e+08</td>\n",
       "      <td>6.396256e+08</td>\n",
       "      <td>6.400394e+08</td>\n",
       "      <td>6.367756e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.329890e+07</td>\n",
       "      <td>2.869792e+07</td>\n",
       "      <td>2.741453e+07</td>\n",
       "      <td>2.147787e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>5.492006e+08</td>\n",
       "      <td>5.981918e+08</td>\n",
       "      <td>5.975137e+08</td>\n",
       "      <td>5.967547e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.763274e+08</td>\n",
       "      <td>6.124102e+08</td>\n",
       "      <td>6.183124e+08</td>\n",
       "      <td>6.200072e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5.919177e+08</td>\n",
       "      <td>6.386069e+08</td>\n",
       "      <td>6.330692e+08</td>\n",
       "      <td>6.364510e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.073691e+08</td>\n",
       "      <td>6.594419e+08</td>\n",
       "      <td>6.559583e+08</td>\n",
       "      <td>6.516836e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>6.641747e+08</td>\n",
       "      <td>7.121491e+08</td>\n",
       "      <td>7.251308e+08</td>\n",
       "      <td>7.028247e+08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  0             1             6             9\n",
       "count  1.000000e+02  1.000000e+02  1.000000e+02  1.000000e+02\n",
       "mean   5.930233e+08  6.396256e+08  6.400394e+08  6.367756e+08\n",
       "std    2.329890e+07  2.869792e+07  2.741453e+07  2.147787e+07\n",
       "min    5.492006e+08  5.981918e+08  5.975137e+08  5.967547e+08\n",
       "25%    5.763274e+08  6.124102e+08  6.183124e+08  6.200072e+08\n",
       "50%    5.919177e+08  6.386069e+08  6.330692e+08  6.364510e+08\n",
       "75%    6.073691e+08  6.594419e+08  6.559583e+08  6.516836e+08\n",
       "max    6.641747e+08  7.121491e+08  7.251308e+08  7.028247e+08"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA6wAAAJJCAYAAACuxbrBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10d3cef60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,10))\n",
    "runs[1].boxplot()\n",
    "runs[1].describe()"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda env:arc]",
   "language": "python",
   "name": "conda-env-arc-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}