mainanalysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import sys\n",
    "import csv\n",
    "import re\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "from pylab import plot, show, savefig, xlim, figure, hold, ylim, legend, boxplot, setp, axes\n",
    "\n",
    "\n",
    "#load datasets\n",
    "cubic_elep_norm_df= pd.read_csv('processedfiles/cubicnormalele.csv')\n",
    "cubic_elep_loss01_df=pd.read_csv('processedfiles/cubicloss01ele.csv')\n",
    "cubic_elep_loss05_df=pd.read_csv('processedfiles/cubicloss05ele.csv')\n",
    "cubic_elep_loss1_df=pd.read_csv('processedfiles/cubicloss1ele.csv')\n",
    "cubic_elep_dup1_df=pd.read_csv('processedfiles/cubicduplicate1ele.csv')\n",
    "cubic_elep_dup5_df=pd.read_csv('processedfiles/cubicduplicate5ele.csv')\n",
    "cubic_elep_re25_df=pd.read_csv('processedfiles/cubicreorder25ele.csv')\n",
    "cubic_elep_re50_df=pd.read_csv('processedfiles/cubicreorder50ele.csv')\n",
    "\n",
    "\n",
    "reno_elep_norm_df= pd.read_csv('processedfiles/renonormalele.csv')\n",
    "reno_elep_loss01_df=pd.read_csv('processedfiles/renoloss01ele.csv')\n",
    "reno_elep_loss05_df=pd.read_csv('processedfiles/renoloss05ele.csv')\n",
    "reno_elep_loss1_df=pd.read_csv('processedfiles/renoloss1ele.csv')\n",
    "reno_elep_dup1_df=pd.read_csv('processedfiles/renoduplicate1ele.csv')\n",
    "reno_elep_dup5_df=pd.read_csv('processedfiles/renoduplicate5ele.csv')\n",
    "reno_elep_re25_df=pd.read_csv('processedfiles/renoreorder25ele.csv')\n",
    "reno_elep_re50_df=pd.read_csv('processedfiles/renoreorder50ele.csv')\n",
    "\n",
    "\n",
    "ham_elep_norm_df= pd.read_csv('processedfiles/hamnormalele.csv')\n",
    "ham_elep_loss01_df=pd.read_csv('processedfiles/hamloss01ele.csv')\n",
    "ham_elep_loss05_df=pd.read_csv('processedfiles/hamloss05ele.csv')\n",
    "ham_elep_loss1_df=pd.read_csv('processedfiles/hamloss1ele.csv')\n",
    "ham_elep_dup1_df=pd.read_csv('processedfiles/hamduplicate1ele.csv')\n",
    "ham_elep_dup5_df=pd.read_csv('processedfiles/hamduplicate5ele.csv')\n",
    "ham_elep_re25_df=pd.read_csv('processedfiles/hamreorder25ele.csv')\n",
    "ham_elep_re50_df=pd.read_csv('processedfiles/hamreorder50ele.csv')\n",
    "\n",
    "\n",
    "bbr_elep_norm_df= pd.read_csv('processedfiles/bbrnormalele.csv')\n",
    "bbr_elep_loss01_df=pd.read_csv('processedfiles/bbrloss01ele.csv')\n",
    "bbr_elep_loss05_df=pd.read_csv('processedfiles/bbrloss05ele.csv')\n",
    "bbr_elep_loss1_df=pd.read_csv('processedfiles/bbrloss1ele.csv')\n",
    "bbr_elep_dup1_df=pd.read_csv('processedfiles/bbrduplicate1ele.csv')\n",
    "bbr_elep_dup5_df=pd.read_csv('processedfiles/bbrduplicate5ele.csv')\n",
    "bbr_elep_re25_df=pd.read_csv('processedfiles/bbrreorder25ele.csv')\n",
    "bbr_elep_re50_df=pd.read_csv('processedfiles/bbrreorder50ele.csv')\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1128caad0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1128ca590>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11297b510>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11272f0d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "#plot RTT \n",
    "fig1, ax1 = plt.subplots()\n",
    "fig2, ax2 = plt.subplots()\n",
    "fig3, ax3 = plt.subplots()\n",
    "fig4, ax4 = plt.subplots()\n",
    "\n",
    "data1 = [cubic_elep_norm_df['s_rtt_avg:52'], cubic_elep_loss01_df['s_rtt_avg:52']\n",
    ", cubic_elep_loss05_df['s_rtt_avg:52'], cubic_elep_loss1_df['s_rtt_avg:52']\n",
    ", cubic_elep_dup1_df['s_rtt_avg:52'], cubic_elep_dup5_df['s_rtt_avg:52']\n",
    ", cubic_elep_re25_df['s_rtt_avg:52'], cubic_elep_re50_df['s_rtt_avg:52']]\n",
    "\n",
    "data2 = [reno_elep_norm_df['s_rtt_avg:52'], reno_elep_loss01_df['s_rtt_avg:52']\n",
    ", reno_elep_loss05_df['s_rtt_avg:52'], reno_elep_loss1_df['s_rtt_avg:52']\n",
    ", reno_elep_dup1_df['s_rtt_avg:52'], reno_elep_dup5_df['s_rtt_avg:52']\n",
    ", reno_elep_re25_df['s_rtt_avg:52'], reno_elep_re50_df['s_rtt_avg:52']]\n",
    "\n",
    "data3 = [ham_elep_norm_df['s_rtt_avg:52'], ham_elep_loss01_df['s_rtt_avg:52']\n",
    ",ham_elep_loss05_df['s_rtt_avg:52'], ham_elep_loss1_df['s_rtt_avg:52']\n",
    ", ham_elep_dup1_df['s_rtt_avg:52'],ham_elep_dup5_df['s_rtt_avg:52']\n",
    ", ham_elep_re25_df['s_rtt_avg:52'], ham_elep_re50_df['s_rtt_avg:52']]\n",
    "\n",
    "data4 = [bbr_elep_norm_df['s_rtt_avg:52'], bbr_elep_loss01_df['s_rtt_avg:52']\n",
    ",bbr_elep_loss05_df['s_rtt_avg:52'], bbr_elep_loss1_df['s_rtt_avg:52']\n",
    ",bbr_elep_dup1_df['s_rtt_avg:52'], bbr_elep_dup5_df['s_rtt_avg:52']\n",
    ",bbr_elep_re25_df['s_rtt_avg:52'], bbr_elep_re50_df['s_rtt_avg:52']]\n",
    "\n",
    "ax1.set_ylabel('Average RTT (ms)')\n",
    "ax1.boxplot(data1, showfliers=False)\n",
    "ax1.set_title('Cubic Elephant Flows')\n",
    "ax2.set_ylabel('Average RTT (ms)')\n",
    "ax2.boxplot(data2, showfliers=False)\n",
    "ax3.set_ylabel('Average RTT (ms)')\n",
    "ax3.boxplot(data3, showfliers=False)\n",
    "ax4.set_ylabel('Average RTT (ms)')\n",
    "ax4.boxplot(data4, showfliers=False)\n",
    "#plt.subplots(nrows=2, ncols=2, constrained_layout=False)\n",
    "plt.xticks([1, 2, 3,4,5,6,7,8], ['Nrm', 'L0.1%','L0.5%',\n",
    "'L1%','D1%','D5%','R25%', 'R50%'])\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "file_extension": ".py",
  "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": "2.7.15"
  },
  "mimetype": "text/x-python",
  "name": "python",
  "npconvert_exporter": "python",
  "pygments_lexer": "ipython2",
  "version": 2
 },
 "nbformat": 4,
 "nbformat_minor": 2
}