chutten/e10s_experiment.ipynb Secret

## e10s_experiment.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### e10s-beta46-noapz: MEMORY_TOTAL analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Unable to parse whitelist (/home/hadoop/anaconda2/lib/python2.7/site-packages/moztelemetry/bucket-whitelist.json). Assuming all histograms are acceptable.\n",
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/hadoop/anaconda2/lib/python2.7/site-packages/matplotlib/font_manager.py:273: UserWarning: Matplotlib is building the font cache using fc-list. This may take a moment.\n",
      "  warnings.warn('Matplotlib is building the font cache using fc-list. This may take a moment.')\n"
     ]
    }
   ],
   "source": [
    "import ujson as json\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import plotly.plotly as py\n",
    "import IPython\n",
    "\n",
    "from __future__ import division\n",
    "from moztelemetry.spark import get_pings, get_one_ping_per_client, get_pings_properties\n",
    "from montecarlino import grouped_permutation_test\n",
    "\n",
    "%pylab inline\n",
    "IPython.core.pylabtools.figsize(16, 7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "64"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sc.defaultParallelism"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def chi2_distance(xs, ys, eps = 1e-10, normalize = True):\n",
    "    histA = xs.sum(axis=0)\n",
    "    histB = ys.sum(axis=0)\n",
    "    \n",
    "    if normalize:\n",
    "        histA = histA/histA.sum()\n",
    "        histB = histB/histB.sum()\n",
    "    \n",
    "    d = 0.5 * np.sum([((a - b) ** 2) / (a + b + eps)\n",
    "        for (a, b) in zip(histA, histB)])\n",
    "\n",
    "    return d\n",
    "\n",
    "def median_diff(xs, ys):\n",
    "    return np.median(xs) - np.median(ys)\n",
    "\n",
    "def compare_histogram(histogram, e10s, none10s):\n",
    "    # Normalize individual histograms\n",
    "    e10s = e10s.map(lambda x: x/x.sum())\n",
    "    none10s = none10s.map(lambda x: x/x.sum())\n",
    "    \n",
    "    e10s = e10s.map(lambda x: x[x.index > 75 * 1024]).map(lambda x: x[x.index < 2048 * 1024])\n",
    "    none10s = none10s.map(lambda x: x[x.index > 75 * 1024]).map(lambda x: x[x.index < 2048 * 1024])\n",
    "    \n",
    "    pvalue = grouped_permutation_test(chi2_distance, [e10s, none10s], num_samples=100)\n",
    "    \n",
    "    eTotal = e10s.sum()\n",
    "    nTotal = none10s.sum()\n",
    "        \n",
    "    eTotal = 100*eTotal/eTotal.sum()\n",
    "    nTotal = 100*nTotal/nTotal.sum()\n",
    "        \n",
    "    fig = plt.figure()\n",
    "    fig.subplots_adjust(hspace=0.3)\n",
    "        \n",
    "    ax = fig.add_subplot(1, 1, 1)\n",
    "    ax2 = ax.twinx()\n",
    "    width = 0.4\n",
    "    ylim = max(eTotal.max(), nTotal.max())\n",
    "        \n",
    "    eTotal.plot(kind=\"bar\", alpha=0.5, color=\"green\", label=\"e10s\", ax=ax, width=width, position=0, ylim=(0, ylim + 1))\n",
    "    nTotal.plot(kind=\"bar\", alpha=0.5, color=\"blue\", label=\"non e10s\", ax=ax2, width=width, position=1, grid=False, ylim=ax.get_ylim())\n",
    "        \n",
    "    ax.legend(ax.get_legend_handles_labels()[0] + ax2.get_legend_handles_labels()[0],\n",
    "              [\"e10s ({} samples\".format(len(e10s)), \"non e10s ({} samples)\".format(len(none10s))])\n",
    "\n",
    "    # If there are more than 100 labels, hide every other one so we can still read them\n",
    "    if len(ax.get_xticklabels()) > 100:\n",
    "        for label in ax.get_xticklabels()[::2]:\n",
    "            label.set_visible(False)\n",
    "    \n",
    "    plt.title(histogram)\n",
    "    plt.xlabel(histogram)\n",
    "    plt.ylabel(\"Frequency %\")\n",
    "    plt.show()\n",
    "        \n",
    "    print \"The probability that the distributions for {} are differing by chance is {:.2f}.\".format(histogram, pvalue)\n",
    "\n",
    "def normalize_uptime_hour(frame):\n",
    "    frame = frame[frame[\"payload/simpleMeasurements/uptime\"] > 0]\n",
    "    frame = 60 * frame.apply(lambda x: x/frame[\"payload/simpleMeasurements/uptime\"]) # Metric per hour\n",
    "    frame.drop('payload/simpleMeasurements/uptime', axis=1, inplace=True)\n",
    "    return frame\n",
    "    \n",
    "def compare_count_histograms(pings, *histograms_names):\n",
    "    properties = histograms_names + (\"payload/simpleMeasurements/uptime\", \"e10s\")\n",
    "\n",
    "    frame = pd.DataFrame(get_pings_properties(pings, properties).collect())\n",
    "\n",
    "    e10s = frame[frame[\"e10s\"] == True]\n",
    "    e10s = normalize_uptime_hour(e10s)\n",
    "    \n",
    "    none10s = frame[frame[\"e10s\"] == False]\n",
    "    none10s = normalize_uptime_hour(none10s)\n",
    "    \n",
    "    for histogram in e10s.columns:\n",
    "        if histogram == \"e10s\" or histogram.endswith(\"_parent\") or histogram.endswith(\"_children\"):\n",
    "            continue\n",
    "            \n",
    "        compare_scalars(histogram + \" per hour\", e10s[histogram].dropna(), none10s[histogram].dropna())\n",
    "\n",
    "        \n",
    "def compare_histograms(pings, *histogram_names):\n",
    "    frame = pd.DataFrame(get_pings_properties(pings, histogram_names + (\"e10s\",) , with_processes=True).collect())\n",
    "    e10s = frame[frame[\"e10s\"] == True]\n",
    "    none10s = frame[frame[\"e10s\"] == False]\n",
    "    \n",
    "    for histogram in none10s.columns:\n",
    "        if histogram == \"e10s\" or histogram.endswith(\"_parent\") or histogram.endswith(\"_children\"):\n",
    "            continue\n",
    "            \n",
    "        has_children = np.sum(e10s[histogram + \"_children\"].notnull()) > 0\n",
    "        has_parent = np.sum(e10s[histogram + \"_parent\"].notnull()) > 0\n",
    "        \n",
    "        if has_children and has_parent:\n",
    "            compare_histogram(histogram + \" (parent + children)\", e10s[histogram].dropna(), none10s[histogram].dropna())\n",
    "\n",
    "        if has_parent:\n",
    "            compare_histogram(histogram + \" (parent)\", e10s[histogram + \"_parent\"].dropna(), none10s[histogram].dropna())\n",
    "\n",
    "        if has_children:\n",
    "            compare_histogram(histogram + \" (children)\", e10s[histogram + \"_children\"].dropna(), none10s[histogram].dropna())\n",
    "\n",
    "def compare_scalars(metric, *groups):\n",
    "    print \"Median difference in {} is {:.2f}, ({:.2f}, {:.2f}).\".format(metric,\n",
    "                                                                        median_diff(*groups), \n",
    "                                                                        np.median(groups[0]),\n",
    "                                                                        np.median(groups[1]))\n",
    "    print \"The probability of this effect being purely by chance is {:.2f}.\". \\\n",
    "        format(grouped_permutation_test(median_diff, groups, num_samples=10000))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Get e10s and non-e10s partitions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dataset = sqlContext.read.load(\"s3://telemetry-parquet/e10s_experiment/e10s_beta46_cohorts/v20160405\", \"parquet\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "What are the branches, and how many clients do we have in each branch?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Row(e10sCohort=u'unknown'),\n",
       " Row(e10sCohort=u'optedOut'),\n",
       " Row(e10sCohort=u'unsupportedChannel'),\n",
       " Row(e10sCohort=u'optedIn'),\n",
       " Row(e10sCohort=u'control'),\n",
       " Row(e10sCohort=u'disqualified'),\n",
       " Row(e10sCohort=u'test')]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.select(\"e10sCohort\").distinct().take(50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "691340"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.filter(dataset[\"e10sCohort\"] == \"test\").count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "690409"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.filter(dataset[\"e10sCohort\"] == \"control\").count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Sample by clientId; `sampled` is a small sample suitable for most measures, while `big_sampled` is a bigger sample used for when the small sample isn't statistically significant enough (such as for the slow script measures):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "sampled = dataset.filter(dataset.sampleId <= 6).filter((dataset.e10sCohort == \"test\") | (dataset.e10sCohort == \"control\"))\n",
    "big_sampled = dataset.filter(dataset.sampleId <= 50).filter((dataset.e10sCohort == \"test\") | (dataset.e10sCohort == \"control\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(97323, 704789)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sampled.count(), big_sampled.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "How many clients have a mismatching e10s cohort?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def e10s_status_mismatch(row):\n",
    "    branch_status = True if row.e10sCohort == \"test\" else False\n",
    "    e10sEnabled = json.loads(row.settings)[\"e10sEnabled\"]\n",
    "    return (row.e10sCohort, branch_status != e10sEnabled)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(u'control', 0), (u'test', 1549)]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sampled.rdd.map(e10s_status_mismatch).reduceByKey(lambda x, y: x + y).collect()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Transform Dataframe to RDD of pings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def row_2_ping(row):\n",
    "    ping = {\"payload\": {\"simpleMeasurements\": json.loads(row.simpleMeasurements) if row.simpleMeasurements else {},\n",
    "                        \"histograms\": json.loads(row.histograms) if row.histograms else {},\n",
    "                        \"keyedHistograms\": json.loads(row.keyedHistograms) if row.keyedHistograms else {},\n",
    "                        \"childPayloads\": json.loads(row.childPayloads) if row.childPayloads else {},\n",
    "                        \"threadHangStats\": json.loads(row.threadHangStats)} if row.threadHangStats else {},\n",
    "           \"e10s\": True if row.e10sCohort == \"test\" else False,\n",
    "           \"os\": json.loads(row.system).get(\"os\", {}).get(\"name\", None)}\n",
    "    return ping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "subset = sampled.rdd.map(row_2_ping)\n",
    "big_subset = big_sampled.rdd.map(row_2_ping)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Memory"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "IPython.core.pylabtools.figsize(20, 18)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABI4AAASqCAYAAAA/ViC5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nA3vtvDsa/EVpAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f4a9b341910>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The probability that the distributions for payload/histograms/MEMORY_TOTAL (parent) are differing by chance is 0.00.\n"
     ]
    }
   ],
   "source": [
    "compare_histograms(subset, \"payload/histograms/MEMORY_TOTAL\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Windows-only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABI4AAAQ6CAYAAAAiMDb6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n9BfqNT2pAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f4a78ae6d50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The probability that the distributions for payload/histograms/MEMORY_TOTAL (parent) are differing by chance is 0.00.\n"
     ]
    }
   ],
   "source": [
    "compare_histograms(subset.filter(lambda p: p[\"os\"] == \"Windows_NT\"), \"payload/histograms/MEMORY_TOTAL\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Mac-only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABI4AAAQ6CAYAAAAiMDb6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\njjIyMjIyMjIyMjIyMjIyMjIy6uL/A8xKo7CSuLJXAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f4aa0c99b50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The probability that the distributions for payload/histograms/MEMORY_TOTAL (parent) are differing by chance is 0.08.\n"
     ]
    }
   ],
   "source": [
    "compare_histograms(big_subset.filter(lambda p: p[\"os\"] == \"Darwin\"), \"payload/histograms/MEMORY_TOTAL\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Linux-only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJoAAAQ6CAYAAAAMdRdoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f4a78342a90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The probability that the distributions for payload/histograms/MEMORY_TOTAL (parent) are differing by chance is 0.08.\n"
     ]
    }
   ],
   "source": [
    "compare_histograms(big_subset.filter(lambda p: p[\"os\"] == \"Linux\"), \"payload/histograms/MEMORY_TOTAL\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}

## e10s_experiment.py

# coding: utf-8

# ### e10s-beta46-noapz: MEMORY_TOTAL analysis

# In[1]:

import ujson as json
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
import plotly.plotly as py
import IPython

from __future__ import division
from moztelemetry.spark import get_pings, get_one_ping_per_client, get_pings_properties
from montecarlino import grouped_permutation_test

get_ipython().magic(u'pylab inline')
IPython.core.pylabtools.figsize(16, 7)


# In[2]:

sc.defaultParallelism


# In[19]:

def chi2_distance(xs, ys, eps = 1e-10, normalize = True):
    histA = xs.sum(axis=0)
    histB = ys.sum(axis=0)

    if normalize:
        histA = histA/histA.sum()
        histB = histB/histB.sum()

    d = 0.5 * np.sum([((a - b) ** 2) / (a + b + eps)
        for (a, b) in zip(histA, histB)])

    return d

def median_diff(xs, ys):
    return np.median(xs) - np.median(ys)

def compare_histogram(histogram, e10s, none10s):
    # Normalize individual histograms
    e10s = e10s.map(lambda x: x/x.sum())
    none10s = none10s.map(lambda x: x/x.sum())

    e10s = e10s.map(lambda x: x[x.index > 75 * 1024]).map(lambda x: x[x.index < 2048 * 1024])
    none10s = none10s.map(lambda x: x[x.index > 75 * 1024]).map(lambda x: x[x.index < 2048 * 1024])

    pvalue = grouped_permutation_test(chi2_distance, [e10s, none10s], num_samples=100)

    eTotal = e10s.sum()
    nTotal = none10s.sum()

    eTotal = 100*eTotal/eTotal.sum()
    nTotal = 100*nTotal/nTotal.sum()

    fig = plt.figure()
    fig.subplots_adjust(hspace=0.3)

    ax = fig.add_subplot(1, 1, 1)
    ax2 = ax.twinx()
    width = 0.4
    ylim = max(eTotal.max(), nTotal.max())

    eTotal.plot(kind="bar", alpha=0.5, color="green", label="e10s", ax=ax, width=width, position=0, ylim=(0, ylim + 1))
    nTotal.plot(kind="bar", alpha=0.5, color="blue", label="non e10s", ax=ax2, width=width, position=1, grid=False, ylim=ax.get_ylim())

    ax.legend(ax.get_legend_handles_labels()[0] + ax2.get_legend_handles_labels()[0],
              ["e10s ({} samples".format(len(e10s)), "non e10s ({} samples)".format(len(none10s))])

    # If there are more than 100 labels, hide every other one so we can still read them
    if len(ax.get_xticklabels()) > 100:
        for label in ax.get_xticklabels()[::2]:
            label.set_visible(False)

    plt.title(histogram)
    plt.xlabel(histogram)
    plt.ylabel("Frequency %")
    plt.show()

    print "The probability that the distributions for {} are differing by chance is {:.2f}.".format(histogram, pvalue)

def normalize_uptime_hour(frame):
    frame = frame[frame["payload/simpleMeasurements/uptime"] > 0]
    frame = 60 * frame.apply(lambda x: x/frame["payload/simpleMeasurements/uptime"]) # Metric per hour
    frame.drop('payload/simpleMeasurements/uptime', axis=1, inplace=True)
    return frame

def compare_count_histograms(pings, *histograms_names):
    properties = histograms_names + ("payload/simpleMeasurements/uptime", "e10s")

    frame = pd.DataFrame(get_pings_properties(pings, properties).collect())

    e10s = frame[frame["e10s"] == True]
    e10s = normalize_uptime_hour(e10s)

    none10s = frame[frame["e10s"] == False]
    none10s = normalize_uptime_hour(none10s)

    for histogram in e10s.columns:
        if histogram == "e10s" or histogram.endswith("_parent") or histogram.endswith("_children"):
            continue

        compare_scalars(histogram + " per hour", e10s[histogram].dropna(), none10s[histogram].dropna())


def compare_histograms(pings, *histogram_names):
    frame = pd.DataFrame(get_pings_properties(pings, histogram_names + ("e10s",) , with_processes=True).collect())
    e10s = frame[frame["e10s"] == True]
    none10s = frame[frame["e10s"] == False]

    for histogram in none10s.columns:
        if histogram == "e10s" or histogram.endswith("_parent") or histogram.endswith("_children"):
            continue

        has_children = np.sum(e10s[histogram + "_children"].notnull()) > 0
        has_parent = np.sum(e10s[histogram + "_parent"].notnull()) > 0

        if has_children and has_parent:
            compare_histogram(histogram + " (parent + children)", e10s[histogram].dropna(), none10s[histogram].dropna())

        if has_parent:
            compare_histogram(histogram + " (parent)", e10s[histogram + "_parent"].dropna(), none10s[histogram].dropna())

        if has_children:
            compare_histogram(histogram + " (children)", e10s[histogram + "_children"].dropna(), none10s[histogram].dropna())

def compare_scalars(metric, *groups):
    print "Median difference in {} is {:.2f}, ({:.2f}, {:.2f}).".format(metric,
                                                                        median_diff(*groups),
                                                                        np.median(groups[0]),
                                                                        np.median(groups[1]))
    print "The probability of this effect being purely by chance is {:.2f}.".         format(grouped_permutation_test(median_diff, groups, num_samples=10000))


# #### Get e10s and non-e10s partitions

# In[4]:

dataset = sqlContext.read.load("s3://telemetry-parquet/e10s_experiment/e10s_beta46_cohorts/v20160405", "parquet")


# What are the branches, and how many clients do we have in each branch?

# In[5]:

dataset.select("e10sCohort").distinct().take(50)


# In[6]:

dataset.filter(dataset["e10sCohort"] == "test").count()


# In[7]:

dataset.filter(dataset["e10sCohort"] == "control").count()


# Sample by clientId; `sampled` is a small sample suitable for most measures, while `big_sampled` is a bigger sample used for when the small sample isn't statistically significant enough (such as for the slow script measures):

# In[8]:

sampled = dataset.filter(dataset.sampleId <= 6).filter((dataset.e10sCohort == "test") | (dataset.e10sCohort == "control"))
big_sampled = dataset.filter(dataset.sampleId <= 50).filter((dataset.e10sCohort == "test") | (dataset.e10sCohort == "control"))


# In[9]:

sampled.count(), big_sampled.count()


# How many clients have a mismatching e10s cohort?

# In[10]:

def e10s_status_mismatch(row):
    branch_status = True if row.e10sCohort == "test" else False
    e10sEnabled = json.loads(row.settings)["e10sEnabled"]
    return (row.e10sCohort, branch_status != e10sEnabled)


# In[11]:

sampled.rdd.map(e10s_status_mismatch).reduceByKey(lambda x, y: x + y).collect()


# Transform Dataframe to RDD of pings

# In[29]:

def row_2_ping(row):
    ping = {"payload": {"simpleMeasurements": json.loads(row.simpleMeasurements) if row.simpleMeasurements else {},
                        "histograms": json.loads(row.histograms) if row.histograms else {},
                        "keyedHistograms": json.loads(row.keyedHistograms) if row.keyedHistograms else {},
                        "childPayloads": json.loads(row.childPayloads) if row.childPayloads else {},
                        "threadHangStats": json.loads(row.threadHangStats)} if row.threadHangStats else {},
           "e10s": True if row.e10sCohort == "test" else False,
           "os": json.loads(row.system).get("os", {}).get("name", None)}
    return ping


# In[35]:

subset = sampled.rdd.map(row_2_ping)
big_subset = big_sampled.rdd.map(row_2_ping)


# ## Memory

# In[26]:

IPython.core.pylabtools.figsize(20, 18)


# In[22]:

compare_histograms(subset, "payload/histograms/MEMORY_TOTAL")


# ### Windows-only

# In[31]:

compare_histograms(subset.filter(lambda p: p["os"] == "Windows_NT"), "payload/histograms/MEMORY_TOTAL")


# ### Mac-only

# In[36]:

compare_histograms(big_subset.filter(lambda p: p["os"] == "Darwin"), "payload/histograms/MEMORY_TOTAL")


# ### Linux-only

# In[37]:

compare_histograms(big_subset.filter(lambda p: p["os"] == "Linux"), "payload/histograms/MEMORY_TOTAL")

	# coding: utf-8

	# ### e10s-beta46-noapz: MEMORY_TOTAL analysis

	# In[1]:

	import ujson as json
	import matplotlib.pyplot as plt
	import pandas as pd
	import numpy as np
	import plotly.plotly as py
	import IPython

	from __future__ import division
	from moztelemetry.spark import get_pings, get_one_ping_per_client, get_pings_properties
	from montecarlino import grouped_permutation_test

	get_ipython().magic(u'pylab inline')
	IPython.core.pylabtools.figsize(16, 7)


	# In[2]:

	sc.defaultParallelism


	# In[19]:

	def chi2_distance(xs, ys, eps = 1e-10, normalize = True):
	histA = xs.sum(axis=0)
	histB = ys.sum(axis=0)

	if normalize:
	histA = histA/histA.sum()
	histB = histB/histB.sum()

	d = 0.5 * np.sum([((a - b) ** 2) / (a + b + eps)
	for (a, b) in zip(histA, histB)])

	return d

	def median_diff(xs, ys):
	return np.median(xs) - np.median(ys)

	def compare_histogram(histogram, e10s, none10s):
	# Normalize individual histograms
	e10s = e10s.map(lambda x: x/x.sum())
	none10s = none10s.map(lambda x: x/x.sum())

	e10s = e10s.map(lambda x: x[x.index > 75 * 1024]).map(lambda x: x[x.index < 2048 * 1024])
	none10s = none10s.map(lambda x: x[x.index > 75 * 1024]).map(lambda x: x[x.index < 2048 * 1024])

	pvalue = grouped_permutation_test(chi2_distance, [e10s, none10s], num_samples=100)

	eTotal = e10s.sum()
	nTotal = none10s.sum()

	eTotal = 100*eTotal/eTotal.sum()
	nTotal = 100*nTotal/nTotal.sum()

	fig = plt.figure()
	fig.subplots_adjust(hspace=0.3)

	ax = fig.add_subplot(1, 1, 1)
	ax2 = ax.twinx()
	width = 0.4
	ylim = max(eTotal.max(), nTotal.max())

	eTotal.plot(kind="bar", alpha=0.5, color="green", label="e10s", ax=ax, width=width, position=0, ylim=(0, ylim + 1))
	nTotal.plot(kind="bar", alpha=0.5, color="blue", label="non e10s", ax=ax2, width=width, position=1, grid=False, ylim=ax.get_ylim())

	ax.legend(ax.get_legend_handles_labels()[0] + ax2.get_legend_handles_labels()[0],
	["e10s ({} samples".format(len(e10s)), "non e10s ({} samples)".format(len(none10s))])

	# If there are more than 100 labels, hide every other one so we can still read them
	if len(ax.get_xticklabels()) > 100:
	for label in ax.get_xticklabels()[::2]:
	label.set_visible(False)

	plt.title(histogram)
	plt.xlabel(histogram)
	plt.ylabel("Frequency %")
	plt.show()

	print "The probability that the distributions for {} are differing by chance is {:.2f}.".format(histogram, pvalue)

	def normalize_uptime_hour(frame):
	frame = frame[frame["payload/simpleMeasurements/uptime"] > 0]
	frame = 60 * frame.apply(lambda x: x/frame["payload/simpleMeasurements/uptime"]) # Metric per hour
	frame.drop('payload/simpleMeasurements/uptime', axis=1, inplace=True)
	return frame

	def compare_count_histograms(pings, *histograms_names):
	properties = histograms_names + ("payload/simpleMeasurements/uptime", "e10s")

	frame = pd.DataFrame(get_pings_properties(pings, properties).collect())

	e10s = frame[frame["e10s"] == True]
	e10s = normalize_uptime_hour(e10s)

	none10s = frame[frame["e10s"] == False]
	none10s = normalize_uptime_hour(none10s)

	for histogram in e10s.columns:
	if histogram == "e10s" or histogram.endswith("_parent") or histogram.endswith("_children"):
	continue

	compare_scalars(histogram + " per hour", e10s[histogram].dropna(), none10s[histogram].dropna())


	def compare_histograms(pings, *histogram_names):
	frame = pd.DataFrame(get_pings_properties(pings, histogram_names + ("e10s",) , with_processes=True).collect())
	e10s = frame[frame["e10s"] == True]
	none10s = frame[frame["e10s"] == False]

	for histogram in none10s.columns:
	if histogram == "e10s" or histogram.endswith("_parent") or histogram.endswith("_children"):
	continue

	has_children = np.sum(e10s[histogram + "_children"].notnull()) > 0
	has_parent = np.sum(e10s[histogram + "_parent"].notnull()) > 0

	if has_children and has_parent:
	compare_histogram(histogram + " (parent + children)", e10s[histogram].dropna(), none10s[histogram].dropna())

	if has_parent:
	compare_histogram(histogram + " (parent)", e10s[histogram + "_parent"].dropna(), none10s[histogram].dropna())

	if has_children:
	compare_histogram(histogram + " (children)", e10s[histogram + "_children"].dropna(), none10s[histogram].dropna())

	def compare_scalars(metric, *groups):
	print "Median difference in {} is {:.2f}, ({:.2f}, {:.2f}).".format(metric,
	median_diff(*groups),
	np.median(groups[0]),
	np.median(groups[1]))
	print "The probability of this effect being purely by chance is {:.2f}.". format(grouped_permutation_test(median_diff, groups, num_samples=10000))


	# #### Get e10s and non-e10s partitions

	# In[4]:

	dataset = sqlContext.read.load("s3://telemetry-parquet/e10s_experiment/e10s_beta46_cohorts/v20160405", "parquet")


	# What are the branches, and how many clients do we have in each branch?

	# In[5]:

	dataset.select("e10sCohort").distinct().take(50)


	# In[6]:

	dataset.filter(dataset["e10sCohort"] == "test").count()


	# In[7]:

	dataset.filter(dataset["e10sCohort"] == "control").count()


	# Sample by clientId; `sampled` is a small sample suitable for most measures, while `big_sampled` is a bigger sample used for when the small sample isn't statistically significant enough (such as for the slow script measures):

	# In[8]:

	sampled = dataset.filter(dataset.sampleId <= 6).filter((dataset.e10sCohort == "test") \| (dataset.e10sCohort == "control"))
	big_sampled = dataset.filter(dataset.sampleId <= 50).filter((dataset.e10sCohort == "test") \| (dataset.e10sCohort == "control"))


	# In[9]:

	sampled.count(), big_sampled.count()


	# How many clients have a mismatching e10s cohort?

	# In[10]:

	def e10s_status_mismatch(row):
	branch_status = True if row.e10sCohort == "test" else False
	e10sEnabled = json.loads(row.settings)["e10sEnabled"]
	return (row.e10sCohort, branch_status != e10sEnabled)


	# In[11]:

	sampled.rdd.map(e10s_status_mismatch).reduceByKey(lambda x, y: x + y).collect()


	# Transform Dataframe to RDD of pings

	# In[29]:

	def row_2_ping(row):
	ping = {"payload": {"simpleMeasurements": json.loads(row.simpleMeasurements) if row.simpleMeasurements else {},
	"histograms": json.loads(row.histograms) if row.histograms else {},
	"keyedHistograms": json.loads(row.keyedHistograms) if row.keyedHistograms else {},
	"childPayloads": json.loads(row.childPayloads) if row.childPayloads else {},
	"threadHangStats": json.loads(row.threadHangStats)} if row.threadHangStats else {},
	"e10s": True if row.e10sCohort == "test" else False,
	"os": json.loads(row.system).get("os", {}).get("name", None)}
	return ping


	# In[35]:

	subset = sampled.rdd.map(row_2_ping)
	big_subset = big_sampled.rdd.map(row_2_ping)


	# ## Memory

	# In[26]:

	IPython.core.pylabtools.figsize(20, 18)


	# In[22]:

	compare_histograms(subset, "payload/histograms/MEMORY_TOTAL")


	# ### Windows-only

	# In[31]:

	compare_histograms(subset.filter(lambda p: p["os"] == "Windows_NT"), "payload/histograms/MEMORY_TOTAL")


	# ### Mac-only

	# In[36]:

	compare_histograms(big_subset.filter(lambda p: p["os"] == "Darwin"), "payload/histograms/MEMORY_TOTAL")


	# ### Linux-only

	# In[37]:

	compare_histograms(big_subset.filter(lambda p: p["os"] == "Linux"), "payload/histograms/MEMORY_TOTAL")