PAM_2023_figs.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configure\n",
    "To recreate the plots from the paper (using the [published data](https://doi.org/10.5281/zenodo.7555410), set the data_folder variable below to the path of the unpacked data.zip. If you want to recreat plots from your own data running the experiments, you will have to update the folder paths at the beginning of each section. Also, set ip_target to match that of IP_TARGET in experiments.conf (\"10.70.2.2\" if using the data from the paper).\n",
    "\n",
    "For the [TCP RTT Accuracy](#tcp_rtt_accuracy) and [TCP timestamps vs SEQ/ACK no.](#tcp_timestamps_vs_seqack) results you also need to set the pping_path variable to the path of the of the PPing executable from [the build with additional output](https://doi.org/10.5281/zenodo.7589243) on the local machine. If it's been added to PATH it should be sufficient to set it to \"pping\". To be able to compare the RTTs with those calculated by tshark, the [tshark tool](https://www.wireshark.org/docs/man-pages/tshark.html) must be installed on the local machine."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_folder = \"../data\"\n",
    "ip_target = \"10.70.2.2\" # Same as IP_TARGET in experiments.conf, the IP of the receiver (M3).\n",
    "pping_path = \"~/src/pping/pping\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import os\n",
    "import re\n",
    "import scipy.stats as stats\n",
    "import sys\n",
    "\n",
    "# Python scripts used for parsing and plotting result data\n",
    "import util\n",
    "import common_plotting as complot\n",
    "import pping_ping_accuracy_viz as ppa_viz\n",
    "import pping_tcp_accuracy_viz as pta_viz\n",
    "import process_data as prodat\n",
    "import kpping_err_parsing as kp_err\n",
    "import tcp_rtt_analysis as tra\n",
    "import pcap_rtt_analysis as pra"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def interaction_plot(data, x, hue, y, dodge=\"auto\", capsize=\"auto\", errwidth=2, scale=0.5, **kwargs):    \n",
    "    n = data[hue].nunique()\n",
    "    if dodge == \"auto\":\n",
    "        dodge = 0.2/n\n",
    "    if capsize == \"auto\":\n",
    "        capsize = 0.1/n\n",
    "    \n",
    "    return sns.pointplot(data=data,  x=x, y=y, hue=hue, dodge=dodge, capsize=capsize, \n",
    "                         scale=scale, errwidth=errwidth, **kwargs)\n",
    "\n",
    "def numerize_nflows(df, inplace=True):\n",
    "    if not inplace:\n",
    "        df = df.copy()\n",
    "    \n",
    "    n_flows = [int(nflows.split(\"_\")[0]) for nflows in df[\"n_flows\"]]\n",
    "    df[\"n_flows\"] = n_flows\n",
    "    return df\n",
    "\n",
    "def numerize_run(df, inplace=True):\n",
    "    if not inplace:\n",
    "        df = df.copy()\n",
    "    \n",
    "    run_ids = [int(run_id.split(\"_\")[1]) for run_id in df[\"run\"]]\n",
    "    df[\"run\"] = run_ids\n",
    "    return df\n",
    "\n",
    "def load_data(root_folder, convert_to_gbps=True, groupby=[\"pping_setup\", \"n_flows\", \"run\"], group_size=None):\n",
    "    \"\"\"\n",
    "    Loads the data and formats it a bit.\n",
    "    - The n_flows and run columns are numerized (converted from strings to ints)\n",
    "    - Throughputs are convereted from bps to Gbps\n",
    "    - A 'network_pps' column is added containing the total packets processed (tx + rx)\n",
    "    - The relative order of \"groups\" is normalized (due to how data merging code dynamically loops through subfolders, group order may not be consistent between data sets)\n",
    "    - Keeps the size of groups consistent (tests sometimes run a bit longer than specified, causing varying numbers of datapoints per experiment)\n",
    "    \n",
    "    \"\"\"\n",
    "    df = pd.read_csv(os.path.join(root_folder, \"data.csv.xz\"))\n",
    "    df = numerize_nflows(df)\n",
    "    df = numerize_run(df)\n",
    "    \n",
    "    if convert_to_gbps:\n",
    "        for col in (\"network_txbps\", \"network_rxbps\", \"tcp_throughput\"):\n",
    "            df[col] = df[col] / 1e9\n",
    "    \n",
    "    df[\"network_pps\"] = df[\"network_txpps\"] + df[\"network_rxpps\"]\n",
    "    df[\"network_Mpps\"] = df[\"network_pps\"] / 1e6\n",
    "    \n",
    "    df_keep = list()\n",
    "    gb = df.groupby(groupby)\n",
    "    if group_size is None:\n",
    "        group_size = gb.size().min()\n",
    "    \n",
    "    for group, group_df in gb:\n",
    "        df_keep.append(group_df.iloc[:group_size])\n",
    "    \n",
    "    return pd.concat(df_keep, ignore_index=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id=\"tcp_rtt_accuracy\"></a>\n",
    "## TCP RTT accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 5.51 s, sys: 221 ms, total: 5.73 s\n",
      "Wall time: 44.4 s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Warning: 17063 entries from tshark missing in common set\n",
      "Warning: 1 entries from PPing missing in common set\n",
      "Warning: 1 entries from ePPing missing in common set\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "tcp_rtt_folder = os.path.join(data_folder, \"rtt_accuracy/tcp/2022-10-14T110904\")\n",
    "rtt_data = pta_viz.load_all_rtt_data(os.path.join(tcp_rtt_folder), pping_path=pping_path, \n",
    "                                     srcip=ip_target, tools=[\"tshark\", \"PPing\", \"ePPing\"])\n",
    "rtt_data = tra.add_delay_offset(rtt_data)\n",
    "rtt_data_matched = tra.filter_comparable_rtts(rtt_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The PostScript backend does not support transparency; partially transparent artists will be rendered opaque.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 400x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(figsize=(4, 2.5))\n",
    "complot.plot_pergroup_timeseries(rtt_data, \"rtt_ms\", print_stats=False, alpha=1.0)\n",
    "axes.set_ylabel(\"RTT (ms)\")\n",
    "axes.set_xlabel(\"Time (s)\")\n",
    "axes.set_ylim(0, 105)\n",
    "\n",
    "fig.savefig(\"PAM_TCP_RTT_timeseries.eps\", dpi=300, bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The PostScript backend does not support transparency; partially transparent artists will be rendered opaque.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 400x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#https://matplotlib.org/stable/gallery/subplots_axes_and_figures/zoom_inset_axes.html\n",
    "\n",
    "fig, axes = plt.subplots(figsize=(4, 2.5))\n",
    "complot.plot_pergroup_cdf(rtt_data, \"rtt_above_us\", axes=axes, print_stats=True, \n",
    "                          stat_kwargs={\"funcs\": (\"count\", \"min\", \"median\", \"max\"), \"fmt\":\"{:.0f}\"}, \n",
    "                          alpha=1.0)\n",
    "axes.set_xlim(0, 400)\n",
    "axes.set_xlabel(\"RTT above base delay (us)\")\n",
    "axes.set_ylabel(\"ECDF\")\n",
    "#axes.set_ylim(0, 1.05)\n",
    "axes.set_yticks([0, 0.25, 0.5, 0.75])\n",
    "\n",
    "axins = axes.inset_axes([0.6, 0.1, 0.37, 0.37])\n",
    "complot.plot_pergroup_cdf(rtt_data, \"rtt_above_us\", axes=axins, print_stats=False, alpha=1.0)\n",
    "axins.get_legend().remove()\n",
    "axins.set_xlim(155, 180)\n",
    "axins.set_ylim(0.37, 0.43)\n",
    "axins.grid(False)\n",
    "#axins.set_xticks([160, 170])\n",
    "axins.set_xticks([])\n",
    "axins.set_yticks([])\n",
    "axins.set_xlabel(\"\")\n",
    "axins.set_ylabel(\"\")\n",
    "\n",
    "axes.indicate_inset_zoom(axins, edgecolor=\"black\")\n",
    "\n",
    "fig.savefig(\"PAM_TCP_RTTabove_cdf_zoomin.eps\", dpi=300, bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY8AAAEhCAYAAACHjCx5AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAABBxElEQVR4nO3deVxTV/o/8E9kCYuYCggRBVdQkFVQBK1CEUGLuLQuRdEqdRlalVFr69jWODqg2KJTGHcHVLQurdu0iGCLK26oWLfiWpEKohgWBUKE8/uDH/drZDHRCwnkeb9evOTee3LueZIrT85dzhEwxhgIIYQQFbRSdwMIIYQ0P5Q8CCGEqIySByGEEJVR8iCEEKIySh6EEEJURsmDEEKIyih5EEIIURklD0IIISrTVXcDmrPs7Gw8efJE3c1ocjKZDEKhUN3NaHIUt3bR1rjNzc1hY2Pz2nKUPN5QdnY27O3tUVpaqu6mNDkdHR1UVlaquxlNjuLWLtoat5GREW7cuPHaBELJ4w09efIEpaWlSExMhL29vbqb02SSkpLw9ddfU9xaguLWrrhv3LiBiRMn4smTJ5Q8Gpu9vT169+6t7mY0mRs3bgCguLUFxa1dcauCLpg3YxkZGZgwYYK6m6EWCQkJ+PDDDwEABw8exOeff67mFhHydv78809s2LChwTKdO3fG1atXm6hFDaPk0Yx5eHhg+/bt6m6G2gUHB2PlypXqbgYhb0WZ5KFJKHloIIFAgKioKPTt2xddu3bFkSNHsHDhQri5uaFXr164du0aAODo0aPw8PAAUH3gmZub45tvvoG7uzu6d++OpKQkdYbBUTYeANi2bRs8PT3Ru3dvDBo0iPuWVVFRgRkzZsDOzg6+vr44e/Ys95qXeyF5eXnw9fWFu7s7evXqhdmzZ6Nm1gGJRIKQkBAMHz4cDg4OeO+99/D06dMmfCf4eS+uXLmCd999F71794aDgwOioqK413z88ccIDw/H4MGDYWdnh9GjR6OioqJJY3wdZd+Dhj7LadOmYdasWQCAp0+folu3bjh58qTaYlJVWVkZxo0bBwcHB7i4uGDIkCGYOXMmrl+/DldXVwQHBwMATpw4AScnJ/Tt2xefffYZNGoGDUbeyIULFxgAduHCBd7rBsDi4uIYY4zt3r2bGRkZsZ9//pkxxtiKFSvYRx99xBhjLC0tjbm7uzPGGLt37x4DwPbv388YY+zQoUPMzs6O97YlJiaqHLey8Zw8eZINGzaMlZeXM8YYO378OHN2dmaMMfb9998zf39/VlFRwZ4/f87c3d3ZBx98wBhjLD4+nvu9rKyMlZSUMMYYe/HiBXv//ffZnj17GGOMLV68mHXt2pUVFBQwxhgbN24ci4yMbLS4G+u9KC4u5taXlpYyV1dXdv78ecYYY5MnT2ZeXl6stLSUvXjxgnl7e7MdO3a8cXv5ivtlyr4HDX2WZWVlzNXVle3evZsFBQWxqKgo3trHWOPE/bK9e/cyf39/brmgoEDh/zNjjJWXlzMrKyuWlpbGGGNs165dDAC7cuVKo7SJMdX+rlHPQ0ONGzcOANC7d2+0atUK77//PgDA3d0dd+/erfM1xsbGGDFiBADAy8sLd+7caZrGKkGZeA4cOIDLly/D09MTrq6umDVrFh4/foyKigqkpaVh8uTJ0NPTg5GRESZOnFjnfqqqqvDFF1/AxcUFbm5uyMjIQGZmJrd96NChMDU1BaC+9+ht34uysjJ88skncHJyQr9+/XD//n2FGEePHg1DQ0Po6Oigb9++GnUc1FDmPWjoszQwMMCePXvwySefQC6X44svvlBLHG/KxcUFf/zxB8LDw7Fr1y7o6enVKpOVlQUjIyP4+PgAAMaOHQuRSNTELa0fJQ8NZWBgAKD6XvOXH1TS0dHBixcvGnxNTTlNukddmXgYY5g6dSoyMzO5n4cPH0JfX1/p7npMTAwKCgpw9uxZ/P777wgJCUF5eXmtdry676b0tu/FP/7xD1haWuLSpUu4fPkyfHx8NC7G11HmPXjdZ/nHH3/A2NgY+fn5Gndq7nW6du2K69evIzAwEKdOnYKjoyOkUqlCGWWPeXWh5EE0xvDhw7F161Y8ePAAQPU3z4yMDACAn58ftm3bhhcvXqCsrAw7duyosw6pVAqxWAwDAwM8evQIe/bsabL286mh90IqlaJjx47Q1dVFVlYWUlNT1dnURtPQZ5mdnY3w8HAcOXIEffv2RUREhPoa+gZycnIgEAgQHByMb7/9FowxmJmZoaioiCvTs2dPlJWV4fjx4wCAH3/8UWG7utFzHkRjDBw4EJGRkRgxYgQqKyshl8vx/vvvw8PDA9OnT8fvv/8OBwcHdOzYEe+++y7u379fq47Zs2djzJgxcHV1RYcOHTB48GA1RPL2GnovvvrqK4SGhmL79u3o3Lkz3nvvPXU3t1HU91m+ePEC48ePx9KlS+Hg4IDVq1fDy8sLu3bt4k6HaborV67gyy+/BGMMVVVVCA0Nhbe3N3r06AFHR0d07doVBw8exA8//IDw8HAYGhrCx8dHqWFDmkyjXXlp4Rrzgrkma+wLiZqK4qa4tQFdMCeEENKoKHkQQghRGSUPQgghKqPkQQghRGVK322lrRMf1adm1M2kpCTud21w6tQpABS3tqC4tSvue/fuKV1WwNjrn0TR5omPGtKqVStUVVWpuxlNjuLWLhS3dtHR0cGJEyfg5eXVYDmleh7aOvFRQ7R1shiKm+LWBtoad81kUMpMv6vSQ4I0Mcr/0dbJYihuilsbaGvcquDtgnnnzp3Rs2dPuLi4wNbWFiNGjEB6ejoAYN26dVi1ahVXNiwsDL169cKoUaPqXG6JZs+ejc6dO0MgENSazCU/Px+BgYGwtbWFo6OjwtDS//3vf+Hk5ARdXV3ExcUpta+pU6dCIBDg2bNnAIDCwkK4urpyP3Z2dtDV1eWGI58yZQqcnZ3h6uqKPn364Ndff+Up6jfX0Pv1st9++w2enp5wcHCAo6MjFi1axI0J9OzZMwQEBMDc3Bzm5ua1XiuVSjFhwgTY2trC3t4eX375ZaPFoyw+4r5y5QoGDhyInj17wsnJCdOnT4dMJuNee/bsWe448PPzQ25ubqPHpYxbt27B29sbdnZ26Nu3L65fv16rzNGjR2FkZKRwPJeVlQGoPl/v7u4OV1dXODk5YcyYMQrjRQkEAu44d3V1xYkTJ5ostoa8bdyv+7wTExO5uN3c3HDo0CF+Gs7XU4edOnVSGCp4//79TCQSsTNnziiUy8vLYyYmJqyysrLOZWVVVlaq/Bo+qfoE6rFjx9iDBw9qvU+MMTZlyhS2ePFixhhj586dYzY2NkwulzPGGMvMzGTXr19noaGhLDY29rX7OXjwIJs6dSoDwA1n/aqVK1eyoKAgblkqlXK/X7p0iZmZmbGqqqo6X9tUT9429H697OLFi+zOnTuMsephuvv378+2b9/OGKse0vrIkSNcTK8aOXIkW7lyJbf88OHDevfTnOK+efMmu3z5MmOseijzsWPHsn/961+MMcaqqqpYt27duGG+V65cycaPH1/vfprySWtfX18WHx/PGGNsz549rF+/frXKvDps+cvKy8tZaWkptzxnzhz297//nVtu6P/Eq5pT3A193gUFBczExIQ7tk+cOMHatWtXb1s04gnzESNGIDw8HN9++y0kEgnmz5+PwsJC+Pr6orS0FL1798by5ctrLQPAt99+i759+6J3794YNmwYNzicRCJBaGgoRo8eDVdXV+Tm5uLw4cMYMGAA3N3d4enpyQ0idvToUbi6uiI8PBwuLi7o1asXN7AcAPzyyy/o06cPXFxc4Orqyk0udP78ebz33nvw8PBA79698dNPP/HyfgwcOBAdO3asc9vu3bvx6aefAgD69OkDS0tLrvfh4uICe3t7tGr1+o+qoKAAS5YsQUxMTIPl4uPjERYWxi2/88473O+FhYUQCASv3Vdja+j9epmbmxu6du0KoHqkVldXV25Ib6FQCD8/P4X4aty+fRsXL17E3LlzuXXt27fnp/FvgY+4bW1t4ezsDKD64mefPn24bRkZGRAKhdww3zNmzMD+/fshl8sbIRrl5efn4+LFi9xQ+x988AHu3buHP//8U+k6hEIhDA0NAQCVlZV49uyZUv9v1ImPuBv6vKuqqsAYUzgLoczxpYxGHRixT58+2L9/P3r16gWg+o9UUlISPDw8uHH5x48fr7C8Y8cO3Lx5E6dPn4aOjg62bduGzz77DAcOHAAApKWl4eLFi7CwsMDdu3exZMkSJCcno02bNrh9+zYGDRrEvfHXrl3Dpk2bsGbNGqxbtw6LFi3C4cOHcfPmTYSFheH48eOws7ODXC5HaWkpCgsLMWPGDPzyyy9o3749njx5And3d/Tv3x9isbhR3qOCggJUVVWhXbt23LrOnTsjOztb5bo+/fRTSCSSBsf8P336NAoKChAUFKSw/ssvv8SePXsglUqxd+9ejUggqsrLy8OPP/6o1AyK169fh7W1NWbOnImMjAyYm5tjxYoVcHNza4KW8quhuJ8/f45NmzZhxYoVAKrvnOzUqRO33cTEBCYmJsjNzVXroHsPHjyAlZUVdHWr/yQJBALY2NggOzsbnTt3ViiblZWF3r17Q0dHB1OmTEF4eDi3raKiAn379sX9+/fh4uKCgwcPKrzWx8cHcrkcfn5+WLp0KYyNjRs9tobwFXeNVz9vc3NzrFu3Dr1794apqSnKyspw5MgRXtreqMmDvcF49Pv370dGRgbc3d0BVH+D0NHR4bYHBQXBwsICAJCcnIzbt29j4MCBCnXU9FR69OjBTdPq5eWFb7/9FgCQmpqKYcOGwc7ODgCgp6cHkUiEpKQk3L17F0OHDlWIISsrq9GSB4Baf6jf5H3bs2cP9PX1ayWFV/33v//FpEmTuIO1xvLly7F8+XIcOXIEn3/+OU6dOgV9fX2V26EuxcXFGD58OBYsWKDUBU65XI7Tp09j6dKl2LBhAw4fPozhw4fjzz//rPXeaLKG4pbL5Rg3bhyGDBnCTRIG8HO8NQZl2tW7d2/k5ORAJBIhJycHw4YNg7m5OcaOHQsA0NfXR2ZmJioqKjBr1iysW7cOCxYsAADcv38fNjY2eP78OWbOnInPP/8ca9asafzAXoOPuIG6P+/i4mKsWbMGGRkZ6NGjB/73v//hww8/xPXr19/6OG/UPt358+fh6Oio0msYY/jqq6+4CXCuXLmiMEta69atFcoGBgYqTJjz119/KXTnaygzKQ5jDM7Ozgr1ZWdnY9CgQSrFAABbt27lLmzFx8fXW87MzAwA8PjxY25dzUGuirS0NPz222/o3Lkz942lV69euHLlClfm+fPn2LVrF6ZOnVpvPYMHD0ZJSYnC65qCsu9XXUpKShAYGIjg4GCF01AN6dSpEzp06ABfX18AQEBAACoqKpCTk6Ny299GY8Utl8sxduxYtG/fHv/+97+59TY2NgqnREpKSlBSUqL2U3bW1tbIyclRmAzrwYMHtf4ftGnThutZd+zYER999FGdF7719fUxZcoUbNu2jVtXU5exsTHCw8M14oI5X3HX93mnpKRAJBKhR48eAKrniZFKpdwX7LfRaMnjwIEDWLt2rdL/mWsEBwdjzZo13J1Acrkcly5dqrPskCFDkJycrHBXyrlz5167j4CAABw6dAg3b97k9lFUVARvb2/cunULv/32G1e25luMqiZNmsQloClTpjRYdsyYMfjPf/4DoDrh5uXlYcCAASrtb82aNcjJycGff/6pcNrOycmJK7Nnzx44OzujZ8+e3LoXL17g1q1b3PK5c+eQn5/PJeCmosr79bJnz54hMDAQAQEB+Prrr5V+nbu7O9q0aYPff/8dALjrYR06dFCt4W+pMeKume/C1NQUGzZsUPhm6+7ujvLychw9ehQAsH79eowcObLOaVCbkoWFBdzc3JCYmAgA+OmnnxS+CNXIzc3lHtwrKSnBzz//zJ1qzM7OxvPnzwFUn+vfvXs3dy1AKpVyDzlXVVVh165dGnGKko+4G/q8u3btiosXLyI/Px9A9Wnrqqoqfo7z115SV/IKfKdOnViPHj2Ys7Mz69atGxs+fDg7deoUY4yxxYsXs3nz5jHGGLt3757CnS+vLjPG2KpVq5iTkxNzdnZmDg4O7LvvvqtVT43Dhw8zT09P5uzszHr27MlCQkIYY7XvTrhy5Qrr1KkTt/zLL7+w3r17MycnJ+bq6srOnj3LGGPs/PnzzMfHhzk7OzN7e3sWEBDAysrKasWr6t0Y4eHhrEOHDkxHR4dZWlqybt26cdvy8vKYv78/6969O3NwcGBHjx7ltm3bto116NCBGRkZsXfeeYd16NCBXbx4kTHG2Nq1a9nXX39d5/5Qx50lAwYMYP/9738V1pWXlzNvb2/Wq1cv5uzszLy8vNivv/5abxxNdRdKQ+9XWFgYO3DgAGOMsWXLljFdXV3m4uLC/Sxbtowr6+bmxsRiMWvVqhXr0KEDmzhxIrft/PnzrE+fPszJyYn16dOHHT9+vN72NKe4a9rq7OzMbQsPD+fqSU9PZ87OzszW1pb5+PiwnJycetvTlHcd/fHHH6xfv37M1taWubu7s6tXrzLGFOOOjY1lDg4O3N+GxYsXc3cG/vLLL8zJyYk5OTmxXr16sdDQUPbkyRPGWHXML/9NmThxIisoKKi3Lc0p7td93qtXr2b29vbM2dmZubu7syNHjtTbFlXutuIteWgbbZ0shuKmuLWBtsatEbfqEkIIabkoeRBCCFEZJQ9CCCEqU+lGX20a1/517ty5AwC4cOEC9/SmNqg5Bihu7UBxa1fcWVlZAKDUUPQ0nwchhBAFZ86cgaenZ4NllEoeAM0k+KqSkhL4+Pjg559/homJibqb06TKy8tRUFCA9u3ba/zYQXyqqKiArq4ucnNztSp2bY5bX18fVVVVWhP78+fPMWzYMBQWFjY4zBGgwmkrGxsbtY59o2mKi4sBAP369eOeEtcWcrkcSUlJ6N+/v9ofLmtq2hq7tsYNaFfsNX/XlBnbrmWnUUIIIY2CkgchhBCVUfIghBCiMkoehBBCVEbJgxBCiMooeRBCCFFZ85kyjRDy1jp/+Qv3+5/L31djS0hzRz0PQgghKqPkQQghRGUalTwkEgkEAoHCj1gs5rYzxiCRSGBlZQVDQ0P4+Pjg2rVrCnXIZDLMmjUL5ubmMDY2RnBwcK15qaVSKUJDQyESiSASiRAaGorCwsKmCJEQjdH5y1+4H0JUpVHJAwB69eqF3Nxc7ufKlSvctujoaMTExCAuLg7nz5+HWCyGv78/SkpKuDIRERHYt28fdu7ciZMnT+LZs2cICgpCZWUlVyYkJASZmZlITk5GcnIyMjMzERoa2qRxEkJIc6ZxF8x1dXUVehs1GGNYvXo1Fi1ahNGjRwMAtmzZAktLS+zYsQMzZsxAUVERNm/ejG3btmHw4MEAgMTERFhbW+PIkSMICAjAjRs3kJycrDBq5MaNG+Hl5YWsrCz06NGj6YIlhJBmSuN6Hrdu3YKVlRW6dOmC8ePH4+7duwCAe/fuIS8vD0OGDOHKCoVCDBo0COnp6QCqx96Xy+UKZaysrODo6MiVOX36NEQikcJww/369YNIJOLKEEIIaZhG9Tw8PT2xdetW2NnZ4dGjR1i2bBm8vb1x7do15OXlAQAsLS0VXmNpaYn79+8DAPLy8qCvr4+2bdvWKlPz+ry8PFhYWNTat4WFBVemLjKZDDKZjFuuGX1SLpdDLpe/QbTNV0282hY30Dxjd5Qc5n4X6tRdpsein7nfr0oCam1vjnHzRZtiVyVGjUoeQ4cO5X53cnKCl5cXunXrhi1btqBfv34Aag8VzBh77fDBr5apq/zr6omKisKSJUtqrU9LS4ORkVGD+2+pUlNT1d0EtWlOsUf3Va18UlJSvduaU9x804bYVZnwT6OSx6uMjY3h5OSEW7duYeTIkQCqew7t27fnyuTn53O9EbFYjIqKCkilUoXeR35+Pry9vbkyjx49qrWvx48f1+rVvGzhwoWYO3cut1xcXAxra2v4+vpq5Xweqamp8Pf3b/HzG7yqOcb+cs9DGfX1PJpb3HzRpthrzqgoQ6OTh0wmw40bN/Duu++iS5cuEIvFSE1NhZubG4Dqmb6OHTuGFStWAADc3d2hp6eH1NRUjB07FgCQm5uLq1evIjo6GgDg5eWFoqIinDt3Dn37Vn8lO3v2LIqKirgEUxehUAihUFhrvZ6eXos/oOpDsTeP2GWVr5/Y52UNxdWc4uabNsSuSnwalTzmz5+P4cOHw8bGBvn5+Vi2bBmKi4sxefJkCAQCREREIDIyEra2trC1tUVkZCSMjIwQEhICABCJRAgLC8O8efNgZmYGU1NTzJ8/H05OTtzdV/b29ggMDMS0adOwfv16AMD06dMRFBREd1oRQoiSNCp55OTk4KOPPsKTJ0/Qrl079OvXD2fOnEGnTp0AAAsWLEBZWRnCw8MhlUrh6emJlJQUhTnEV61aBV1dXYwdOxZlZWXw8/NDQkICdHT+70rh9u3bMXv2bO6urODgYMTFxTVtsIQQ0oxpVPLYuXNng9sFAgEkEgkkEkm9ZQwMDBAbG4vY2Nh6y5iamiIxMfFNm0kIIVpPo5IHIeTt0XAjpClo3EOChBBCNB8lD0IIISqj5EEIIURldM2DEKKAZhskyqCeByGEEJVR8iCEEKIyOm1FSAtAt+eSpkY9D0IIISqjngchpF41PRqhDlN5aHfSslHPgxBCiMooeRBCCFEZJQ9CCCEqo+RBCCFEZZQ8CCGEqIySByGEEJVR8iCEEKIySh6EEEJURsmDEEKIyih5EEIIURklD0IIISqjsa0IaaZoJF2iTtTzIIQQojJKHoQQQlRGp60IaUboVBXRFBrd84iKioJAIEBERAS3jjEGiUQCKysrGBoawsfHB9euXVN4nUwmw6xZs2Bubg5jY2MEBwcjJydHoYxUKkVoaChEIhFEIhFCQ0NRWFjYBFERQkjzp7HJ4/z589iwYQOcnZ0V1kdHRyMmJgZxcXE4f/48xGIx/P39UVJSwpWJiIjAvn37sHPnTpw8eRLPnj1DUFAQKisruTIhISHIzMxEcnIykpOTkZmZidDQ0CaLj5DmyFFymHo/BICGJo9nz55hwoQJ2LhxI9q2bcutZ4xh9erVWLRoEUaPHg1HR0ds2bIFpaWl2LFjBwCgqKgImzdvxnfffYfBgwfDzc0NiYmJuHLlCo4cOQIAuHHjBpKTk7Fp0yZ4eXnBy8sLGzduxM8//4ysrCy1xEwIIc2JRl7z+PTTT/H+++9j8ODBWLZsGbf+3r17yMvLw5AhQ7h1QqEQgwYNQnp6OmbMmIELFy5ALpcrlLGysoKjoyPS09MREBCA06dPQyQSwdPTkyvTr18/iEQipKeno0ePHrXaJJPJIJPJuOXi4mIAgFwuh1wu5zV+TVcTr7bFDag/dqEOU89+WzGFf7Xps1f3Z96UVIlR45LHzp07cfHiRZw/f77Wtry8PACApaWlwnpLS0vcv3+fK6Ovr6/QY6kpU/P6vLw8WFhY1KrfwsKCK/OqqKgoLFmypNb6tLQ0GBkZKRFZy5OamqruJqiNumJX9zziSz2qAABJSUnqbYgaaMPxXlpaqnRZjUoeDx48wJw5c5CSkgIDA4N6ywkEAoVlxlitda96tUxd5RuqZ+HChZg7dy63XFxcDGtra/j6+sLMzKzBfbc0crkcqamp8Pf3h56enrqb06TUHbuj5HCT7xOo7nEs9ajC1xmtIKsS4KokQC3tUAd1f+ZNqeaMijI0KnlcuHAB+fn5cHd359ZVVlbi+PHjiIuL465H5OXloX379lyZ/Px8rjciFotRUVEBqVSq0PvIz8+Ht7c3V+bRo0e19v/48eNavZoaQqEQQqGw1no9Pb0Wf0DVh2Jv+thllQ1/SWr0/VcJIKsUaOXnrg3HuyrxadQFcz8/P1y5cgWZmZncj4eHByZMmIDMzEx07doVYrFYoftYUVGBY8eOcYnB3d0denp6CmVyc3Nx9epVroyXlxeKiopw7tw5rszZs2dRVFTElSGE1K/zl78o/BDto1E9DxMTEzg6OiqsMzY2hpmZGbc+IiICkZGRsLW1ha2tLSIjI2FkZISQkBAAgEgkQlhYGObNmwczMzOYmppi/vz5cHJywuDBgwEA9vb2CAwMxLRp07B+/XoAwPTp0xEUFFTnxXJCCCGKNCp5KGPBggUoKytDeHg4pFIpPD09kZKSAhMTE67MqlWroKuri7Fjx6KsrAx+fn5ISEiAjo4OV2b79u2YPXs2d1dWcHAw4uLimjweQghpjjQ+eRw9elRhWSAQQCKRQCKR1PsaAwMDxMbGIjY2tt4ypqamSExM5KmVhBCiXTQ+eRCi7eiaAtFEGnXBnBBCSPNAPQ9CyFt7uXf05/L31dgS0lSo50EIIURlvCWPe/fu8VUVIYQQDcdb8ujevTt8fX2RmJiI8vJyvqolhBCigXhLHpcvX4abmxvmzZsHsViMGTNmKDzBTQghpOXgLXk4OjoiJiYGf/31F+Lj45GXl4cBAwagV69eiImJwePHj/naFSGEEDXj/YK5rq4uRo0ahd27d2PFihW4c+cO5s+fj44dO2LSpEnIzc3le5eEEEKaGO/JIyMjA+Hh4Wjfvj1iYmIwf/583LlzB7/99hv++usvjBgxgu9dEkIIaWK8PecRExOD+Ph4ZGVlYdiwYdi6dSuGDRuGVq2q81OXLl2wfv169OzZk69dEkIIURPeksfatWsxdepUTJkyBWKxuM4yNjY22Lx5M1+7JKTFoiFJiKbjLXncunXrtWX09fUxefJkvnZJCCFETXi75hEfH489e/bUWr9nzx5s2bKFr90QQgjRALwlj+XLl8Pc3LzWegsLC0RGRvK1G0IIIRqAt9NW9+/fR5cuXWqt79SpE7Kzs/naDSEtFl3nIM0Jbz0PCwsL/P7777XWX758GWZmZnzthhBCiAbgrecxfvx4zJ49GyYmJhg4cCAA4NixY5gzZw7Gjx/P124IIRqOhmfXDrwlj2XLluH+/fvw8/ODrm51tVVVVZg0aRJd8yCEkBaGt+Shr6+PXbt2YenSpbh8+TIMDQ3h5OSETp068bULQgghGoL3mQTt7OxgZ2fHd7WEEEI0CG/Jo7KyEgkJCfj111+Rn5+Pqqoqhe2//fYbX7sihBCiZrwljzlz5iAhIQHvv/8+HB0dIRAI+KqaEEKIhuEteezcuRO7d+/GsGHD+KqSEEKIhuLtOQ99fX107979repYu3YtnJ2d0aZNG7Rp0wZeXl44dOgQt50xBolEAisrKxgaGsLHxwfXrl1TqEMmk2HWrFkwNzeHsbExgoODkZOTo1BGKpUiNDQUIpEIIpEIoaGhKCwsfKu2E0Jq6/zlL9wPaVl4Sx7z5s3Dv//9bzDG3riOjh07Yvny5cjIyEBGRgbee+89jBgxgksQ0dHRiImJQVxcHM6fPw+xWAx/f3+UlJRwdURERGDfvn3YuXMnTp48iWfPniEoKAiVlZVcmZCQEGRmZiI5ORnJycnIzMxEaGjomwdPCCFahrfTVidPnkRaWhoOHTqEXr16QU9PT2H73r17X1vH8OHDFZb/9a9/Ye3atThz5gwcHBywevVqLFq0CKNHjwYAbNmyBZaWltixYwdmzJiBoqIibN68Gdu2bcPgwYMBAImJibC2tsaRI0cQEBCAGzduIDk5GWfOnIGnpycAYOPGjfDy8kJWVhZ69OjBx9tBCCEtGm/J45133sGoUaP4qg6VlZXYs2cPnj9/Di8vL9y7dw95eXkYMmQIV0YoFGLQoEFIT0/HjBkzcOHCBcjlcoUyVlZWcHR0RHp6OgICAnD69GmIRCIucQBAv379IBKJkJ6eXm/ykMlkkMlk3HJxcTEAQC6XQy6X8xZ3c1ATr7bFDTRu7EKdN++1NzZhK6bw75torseLNh3vqsTIW/KIj4/npZ4rV67Ay8sL5eXlaN26Nfbt2wcHBwekp6cDACwtLRXKW1pa4v79+wCAvLw86Ovro23btrXK5OXlcWUsLCxq7dfCwoIrU5eoqCgsWbKk1vq0tDQYGRmpFmQLkZqaqu4mqE1jxB7dl/cqebfUo+r1heqRlJTEY0uanjYc76WlpUqX5fUhwRcvXuDo0aO4c+cOQkJCYGJigocPH6JNmzZo3bq1UnX06NEDmZmZKCwsxE8//YTJkyfj2LFj3PZXbwFmjL32tuBXy9RV/nX1LFy4EHPnzuWWi4uLYW1tDV9fX60b+FEulyM1NRX+/v61Tk+2dHzG7ig5zFOrGp+wFcNSjyp8ndEKsqo3uw3/qiSA51Y1DW063mvOqCiD1yHZAwMDkZ2dDZlMBn9/f5iYmCA6Ohrl5eVYt26dUvW8fNeWh4cHzp8/j3//+9/44osvAFT3HNq3b8+Vz8/P53ojYrEYFRUVkEqlCr2P/Px8eHt7c2UePXpUa7+PHz+u1at5mVAohFAorLVeT0+vxR9Q9aHY3y52WWXzexZKViV443Y392NFG453VeLj7W6rOXPmwMPDA1KpFIaGhtz6UaNG4ddff33jehljkMlk6NKlC8RisULXsaKiAseOHeMSg7u7O/T09BTK5Obm4urVq1wZLy8vFBUV4dy5c1yZs2fPoqioiCtDCCGkYbzebXXq1Cno6+srrO/UqRP++usvper4xz/+gaFDh8La2holJSXYuXMnjh49iuTkZAgEAkRERCAyMhK2trawtbVFZGQkjIyMEBISAgAQiUQICwvDvHnzYGZmBlNTU8yfPx9OTk7c3Vf29vYIDAzEtGnTsH79egDA9OnTERQURHdaEdKIaKj2loW35FFVVaXwLEWNnJwcmJiYKFXHo0ePEBoaitzcXIhEIjg7OyM5ORn+/v4AgAULFqCsrAzh4eGQSqXw9PRESkqKQv2rVq2Crq4uxo4di7KyMvj5+SEhIQE6Ojpcme3bt2P27NncXVnBwcGIi4t7m/AJIUSr8JY8/P39sXr1amzYsAFA9UXpZ8+eYfHixUoPWbJ58+YGtwsEAkgkEkgkknrLGBgYIDY2FrGxsfWWMTU1RWJiolJtIoQQUhtvyWPVqlXw9fWFg4MDysvLERISglu3bsHc3Bw//PADX7shhBCiAXhLHlZWVsjMzMQPP/yAixcvoqqqCmFhYZgwYYLCBXRCCCHNH6/PeRgaGmLq1KmYOnUqn9USQgjRMLwlj61btza4fdKkSXztipBmi0aXJS0Fr5NBvUwul6O0tBT6+vowMjKi5EEIIS0Ibw8JSqVShZ9nz54hKysLAwYMoAvmhBDSwvCWPOpia2uL5cuX1+qVEEIIad4aNXkAgI6ODh4+fNjYuyGEENKEeLvmcfDgQYVlxhhyc3MRFxeH/v3787UbQgghGoC35DFy5EiFZYFAgHbt2uG9997Dd999x9duCCGEaABex7YihBCiHXh9SJAQUhs921EbjbDb/PGWPF6eZe91YmJi+NotIYQQNeAteVy6dAkXL17EixcvuHkxbt68CR0dHfTu3Zsr97opYwkhhGg+3pLH8OHDYWJigi1btnBTwEqlUkyZMgXvvvsu5s2bx9euCCGEqBlvz3l89913iIqKUpg7vG3btli2bBndbUUIIS0Mb8mjuLgYjx49qrU+Pz8fJSUlfO2GEEKIBuAteYwaNQpTpkzBjz/+iJycHOTk5ODHH39EWFgYRo8ezdduCCGEaADernmsW7cO8+fPx8SJEyGXy6sr19VFWFgYVq5cydduCCEtDN222zzxljyMjIywZs0arFy5Enfu3AFjDN27d4exsTFfuyCEEKIheB8YMTc3F7m5ubCzs4OxsTEYY3zvghBCiJrxljwKCgrg5+cHOzs7DBs2DLm5uQCATz75hG7TJYSQFoa35PH3v/8denp6yM7OhpGREbd+3LhxSE5O5ms3hBBCNABv1zxSUlJw+PBhdOzYUWG9ra0t7t+/z9duCGkWaDwr0tLx1vN4/vy5Qo+jxpMnTyAUCpWqIyoqCn369IGJiQksLCwwcuRIZGVlKZRhjEEikcDKygqGhobw8fHBtWvXFMrIZDLMmjUL5ubmMDY2RnBwMHJychTKSKVShIaGQiQSQSQSITQ0FIWFhaoFTQghWoq35DFw4EBs3bqVWxYIBKiqqsLKlSvh6+urVB3Hjh3Dp59+ijNnziA1NRUvXrzAkCFD8Pz5c65MdHQ0YmJiEBcXh/Pnz0MsFsPf31/hQcSIiAjs27cPO3fuxMmTJ/Hs2TMEBQWhsrKSKxMSEoLMzEwkJycjOTkZmZmZCA0N5eGdIISQlo+301YrV66Ej48PMjIyUFFRgQULFuDatWt4+vQpTp06pVQdr14biY+Ph4WFBS5cuICBAweCMYbVq1dj0aJF3IOHW7ZsgaWlJXbs2IEZM2agqKgImzdvxrZt2zB48GAAQGJiIqytrXHkyBEEBATgxo0bSE5OxpkzZ+Dp6QkA2LhxI7y8vJCVlcUN7EgIIaRuvCUPBwcH/P7771i7di10dHTw/PlzjB49Gp9++inat2//RnUWFRUBAExNTQEA9+7dQ15eHoYMGcKVEQqFGDRoENLT0zFjxgxcuHABcrlcoYyVlRUcHR2Rnp6OgIAAnD59GiKRiEscANCvXz+IRCKkp6fXmTxkMhlkMhm3XFxcDACQy+XcQ5HaoiZebYsbUD52oU7LukVd2Iop/NtYeiz6WWH5qiSgUfenDG063lWJkZfkUfPHev369ViyZAkfVYIxhrlz52LAgAFwdHQEAOTl5QEALC0tFcpaWlpyF+Xz8vKgr6+vMEBjTZma1+fl5cHCwqLWPi0sLLgyr4qKiqoztrS0tDqv9WiD1NRUdTdBbV4Xe3TfJmpIE1vq0bQzhiYlJTXp/hqiDcd7aWmp0mV5SR56enq4evUqr3N1fPbZZ/j9999x8uTJWtte3Q9j7LX7frVMXeUbqmfhwoUKE14VFxfD2toavr6+MDMza3DfLY1cLkdqair8/f2hp6en7uY0KWVjd5QcbsJWNT5hK4alHlX4OqMVZFVNNyePpvQ8tOV4rzmjogzeTltNmjQJmzdvxvLly9+6rlmzZuHgwYM4fvy4wq2/YrEYQHXP4eVTYfn5+VxvRCwWo6KiAlKpVKH3kZ+fD29vb65MXSMAP378uFavpoZQKKzzrjE9Pb0Wf0DVh2KvP3ZZZcuc9ExWJWjS2DTp+NKG412V+HhLHhUVFdi0aRNSU1Ph4eFRa0wrZaaeZYxh1qxZ2LdvH44ePYouXboobO/SpQvEYjFSU1Ph5ubG7ffYsWNYsWIFAMDd3R16enpITU3F2LFjAVQPmXL16lVER0cDALy8vFBUVIRz586hb9/q8wtnz55FUVERl2AIIYTU762Tx927d9G5c2dcvXqVm2725s2bCmWUPZ316aefYseOHThw4ABMTEy46w8ikQiGhoYQCASIiIhAZGQkbG1tYWtri8jISBgZGSEkJIQrGxYWhnnz5sHMzAympqaYP38+nJycuLuv7O3tERgYiGnTpmH9+vUAgOnTpyMoKIjutCJvjB4MJNrkrZOHra0tcnNzkZaWBqB6OJLvv/++3tM/DVm7di0AwMfHR2F9fHw8Pv74YwDAggULUFZWhvDwcEilUnh6eiIlJQUmJiZc+VWrVkFXVxdjx45FWVkZ/Pz8kJCQAB0dHa7M9u3bMXv2bO6urODgYMTFxancZkII0UZvnTxeHTX30KFDCg/1vU1ddREIBJBIJJBIJPWWMTAwQGxsLGJjY+stY2pqisTExDdpJiGkidBcH5qL9yHZaQh2Qghp+d46eQgEglrXNPi8ZZcQQojm4eW01ccff8zdxlpeXo6ZM2fWuttq7969b7srQgghGuKtk8fkyZMVlidOnPi2VRJCCNFwb5084uPj+WgHIYSQZoS3hwQJ0Ub0bAfRVrzfbUUIIaTlo+RBCCFEZZQ8CCGEqIySByGEEJVR8iCEEKIyutuKENIs0DhXmoV6HoQQQlRGyYMQQojKKHkQQghRGV3zIOQNOEoOt9h5yglRBvU8CCGEqIySByGEEJVR8iCEEKIySh6EEEJURhfMCSHNDj0wqH7U8yCEEKIySh6EEEJURsmDEEKIyjQueRw/fhzDhw+HlZUVBAIB9u/fr7CdMQaJRAIrKysYGhrCx8cH165dUygjk8kwa9YsmJubw9jYGMHBwcjJyVEoI5VKERoaCpFIBJFIhNDQUBQWFjZydIQQ0jJoXPJ4/vw5XFxcEBcXV+f26OhoxMTEIC4uDufPn4dYLIa/vz9KSkq4MhEREdi3bx927tyJkydP4tmzZwgKCkJlZSVXJiQkBJmZmUhOTkZycjIyMzMRGhra6PERQkhLoHF3Ww0dOhRDhw6tcxtjDKtXr8aiRYswevRoAMCWLVtgaWmJHTt2YMaMGSgqKsLmzZuxbds2DB48GACQmJgIa2trHDlyBAEBAbhx4waSk5Nx5swZeHp6AgA2btwILy8vZGVloUePHk0TLGk2au7uEeowRPdVc2MI0QAa1/NoyL1795CXl4chQ4Zw64RCIQYNGoT09HQAwIULFyCXyxXKWFlZwdHRkStz+vRpiEQiLnEAQL9+/SASibgyhBBC6qdxPY+G5OXlAQAsLS0V1ltaWuL+/ftcGX19fbRt27ZWmZrX5+XlwcLColb9FhYWXJlXyWQyyGQybrm4uBgAIJfLIZfL3zCi5qkmXm2KW6jDqv9tpfivttDkuBv7ONSm412VGJtV8qghECiOZsoYq7XuVa+Wqat8Q/VERUVhyZIltdanpaXByMhImWa3OKmpqepuQpN59VTVUo8q9TREzTQx7qSkpCbZjzYc76WlpUqXbVbJQywWA6juObRv355bn5+fz/VGxGIxKioqIJVKFXof+fn58Pb25so8evSoVv2PHz+u1aupsXDhQsydO5dbLi4uhrW1NXx9fWFmZvb2wTUjcrkcqamp8Pf3h56enrqb0yQcJYcBVH/zXupRha8zWkFWpT1DsjeXuK9KAnivU5uO95ozKspoVsmjS5cuEIvFSE1NhZubGwCgoqICx44dw4oVKwAA7u7u0NPTQ2pqKsaOHQsAyM3NxdWrVxEdHQ0A8PLyQlFREc6dO4e+fau/Up49exZFRUVcgnmVUCiEUCistV5PT6/FH1D10abYX527Q1Yl0Mr5PDQ97sY8HrXheFclPo1LHs+ePcPt27e55Xv37iEzMxOmpqawsbFBREQEIiMjYWtrC1tbW0RGRsLIyAghISEAAJFIhLCwMMybNw9mZmYwNTXF/Pnz4eTkxN19ZW9vj8DAQEybNg3r168HAEyfPh1BQUF0pxUhhChB45JHRkYGfH19ueWaU0WTJ09GQkICFixYgLKyMoSHh0MqlcLT0xMpKSkwMTHhXrNq1Sro6upi7NixKCsrg5+fHxISEqCjo8OV2b59O2bPns3dlRUcHFzvsyWEEEIUaVzy8PHxAWP139EhEAggkUggkUjqLWNgYIDY2FjExsbWW8bU1BSJiYlv01RCCNFaGpc8CNEELw/5TQiprVk9JEgIIUQzUPIghBCiMkoehBBCVEbXPAghLQZNT9t0qOdBCCFEZZQ8CCGEqIxOWxHy/9HtuYQoj3oehBBCVEbJgxBCiMooeRBCCFEZXfMghLRIdNtu46KeByGEEJVR8iCEEKIySh6EEEJURtc8iFajZzsIeTPU8yCEEKIySh6EEEJURqetCCEtHt22yz/qeRBCCFEZ9TyI1qGL5IS8Pep5EEIIURklD0IIISqj01ZEK9CpKlLj1WOBLqC/Ga3veaxZswZdunSBgYEB3N3dceLECXU3ifCk85e/cD+EEH5pdc9j165diIiIwJo1a9C/f3+sX78eQ4cOxfXr12FjY6Pu5pE3QImCkKah1ckjJiYGYWFh+OSTTwAAq1evxuHDh7F27VpERUWpuXVEWZQwCGl6Wps8KioqcOHCBXz55ZcK64cMGYL09HQ1tYpQIiBNjR4gfDNamzyePHmCyspKWFpaKqy3tLREXl5erfIymQwymYxbLioqAgA8ffq0cRuqgeRyOUpLS+G6aC9kVQKFbWcX+nG/e0b9qnLdmn5A6lYxlJZWQVfeCpWvxN6SaUvc3efvrrVO2IrhK7cqFBQUQE9PTw2tajolJSUAAMbYa8tq+v/VRicQKP5HYIzVWgcAUVFRWLJkSa31dnZ2jda25sj8O3W3oPGFqLsBaqKtcQPaF3tJSQlEIlGDZbQ2eZibm0NHR6dWLyM/P79WbwQAFi5ciLlz53LLhYWF6NSpE7Kzs1/7Jrc0xcXFsLa2xoMHD9CmTRt1N6dJaWvs2ho3oF2xM8ZQUlICKyur15bV2uShr68Pd3d3pKamYtSoUdz61NRUjBgxolZ5oVAIoVBYa71IJGrxB1R92rRpQ7FrGW2NG9Ce2JX9Mqy1yQMA5s6di9DQUHh4eMDLywsbNmxAdnY2Zs6cqe6mEUKIRtPq5DFu3DgUFBTgn//8J3Jzc+Ho6IikpCR06tRJ3U0jhBCNptXJAwDCw8MRHh6u8uuEQiEWL15c56mslo5i177YtTVuQLtjb4iAKXNPFiGEEPISrR/bihBCiOooeRBCCFEZJQ9CCCEqo+RBCCFEZZQ8eCaTyeDq6gqBQIDMzEx1N6fRBQcHw8bGBgYGBmjfvj1CQ0Px8OFDdTer0f35558ICwtDly5dYGhoiG7dumHx4sWoqKhQd9OaxL/+9S94e3vDyMgI77zzjrqb02hovp/6UfLg2YIFC5R6tL+l8PX1xe7du5GVlYWffvoJd+7cwYcffqjuZjW6P/74A1VVVVi/fj2uXbuGVatWYd26dfjHP/6h7qY1iYqKCowZMwZ/+9vf1N2URlMz38+iRYtw6dIlvPvuuxg6dCiys7PV3TTNwAhvkpKSWM+ePdm1a9cYAHbp0iV1N6nJHThwgAkEAlZRUaHupjS56Oho1qVLF3U3o0nFx8czkUik7mY0ir59+7KZM2cqrOvZsyf78ssv1dQizUI9D548evQI06ZNw7Zt22BkZKTu5qjF06dPsX37dnh7e7f4oavrUlRUBFNTU3U3g/CgZr6fIUOGKKyn+X7+DyUPHjDG8PHHH2PmzJnw8PBQd3Oa3BdffAFjY2OYmZkhOzsbBw4cUHeTmtydO3cQGxtL46K1EKrO96ONKHk0QCKRQCAQNPiTkZGB2NhYFBcXY+HChepuMi+UjbvG559/jkuXLiElJQU6OjqYNGmSUpPJaCJVYweAhw8fIjAwEGPGjOGmNG6O3iT2lk7Z+X60EQ1P0oAnT57gyZMnDZbp3Lkzxo8fj//9738KB1VlZSV0dHQwYcIEbNmypbGbyitl4zYwMKi1PicnB9bW1khPT4eXl1djNbHRqBr7w4cP4evrC09PTyQkJKBVq+b7fexNPveEhARERESgsLCwkVvXtCoqKmBkZIQ9e/YoTNkwZ84cZGZm4tixY2psnWbQ+oERG2Jubg5zc/PXlvv++++xbNkybvnhw4cICAjArl274Onp2ZhNbBTKxl2Xmu8iL0/Z25yoEvtff/0FX19fuLu7Iz4+vlknDuDtPveWRtX5frQRJQ8e2NjYKCy3bt0aANCtWzd07NhRHU1qEufOncO5c+cwYMAAtG3bFnfv3sU333yDbt26NctehyoePnwIHx8f2NjY4Ntvv8Xjx4+5bWKxWI0taxrZ2dl4+vQpsrOzUVlZyT3T1L17d+74b+5ovp+GUfIgb8zQ0BB79+7F4sWL8fz5c7Rv3x6BgYHYuXNnix++OiUlBbdv38bt27drfUHQhjPB33zzjcLpWDc3NwBAWloafHx81NQqftF8Pw2jax6EEEJU1rxP0hJCCFELSh6EEEJURsmDEEKIyih5EEIIURklD0IIISqj5EEIIURllDwIIYSojJIHIYQQlVHyIEQFCQkJCtOuSiQSuLq6KpSRSCSwtLSEQCDA/v37613X0mRlZUEsFqOkpISX+uLi4hAcHMxLXaQRqG0aKkJUMHnyZAaAAWA6OjrM2tqazZw5kz19+pSlpaVx2+r76dSpU4PbBw0apFQ7Xp05r6SkhD158oRbvn79OgPA9u3bx3Jzc1l5eXmd61qiDz74gP3zn//krb7y8nImFovZiRMneKuT8IfGtiLNRmBgIOLj4/HixQtcv34dU6dORWFhIbZs2YLc3Fyu3Jw5c1BcXIz4+HhunVwu52Y3TE9PxwcffICsrCy0adMGQPUoqm+idevWCgMB3rlzBwAwYsQIboj+uta9iZdj0DQ5OTk4ePAgVq9ezVudQqEQISEhiI2NxYABA3irl/CDTluRZkMoFEIsFqNjx44YMmQIxo0bh5SUFOjr60MsFnM/hoaGXNmaH2tra+73mqliLSwsaq17VUJCAmxsbGBkZIRRo0ahoKBAYfvLp60kEgmGDx8OAGjVqhUEAkGd62rEx8fD3t4eBgYG6NmzJ9asWcNt+/PPPyEQCLB79274+PjAwMAAiYmJSr9u79698PX1hZGREVxcXHD69GmFdp86dQqDBg2CkZER2rZti4CAAEilUgDVAztGR0eja9euMDQ0hIuLC3788ccGP5vdu3fDxcVFYZDIuk7prV69Gp07d+aWjx49ir59+8LY2BjvvPMO+vfvj/v373Pbg4ODsX//fpSVlTW4f9L0qOdBmqW7d+8iOTm5Ub+Jnz17FlOnTkVkZCRGjx6N5ORkLF68uN7y8+fPR+fOnTFlyhSuJ9S6deta6wBg48aNWLx4MeLi4uDm5oZLly5h2rRpMDY2xuTJk7lyX3zxBb777jvEx8dDKBQq/bpFixbh22+/ha2tLRYtWoSPPvoIt2/fhq6uLjIzM+Hn54epU6fi+++/h66uLtLS0lBZWQkA+Oqrr7B3716sXbsWtra2OH78OCZOnIh27dph0KBBdcZ+/PhxladgfvHiBUaOHIlp06bhhx9+QEVFBc6dO6eQYD08PCCXy3Hu3Ll6903URN3nzQhRxuTJk5mOjg4zNjZmBgYG3LWKmJiYOsuOGDGi3rpqrpFIpdIG9/nRRx+xwMBAhXXjxo1TuOaxePFi5uLiwi3v27ePvfrfqq511tbWbMeOHQrrli5dyry8vBhjjN27d48BYKtXr36j123atInbfu3aNQaA3bhxg4urf//+dcb87NkzZmBgwNLT0xXWh4WFsY8++qjO1zDGmIuLS63rHa++N4wxtmrVKtapUyfGGGMFBQUMADt69Gi99TLGWNu2bVlCQkKDZUjTo54HaTZ8fX2xdu1alJaWYtOmTbh58yZmzZrFS929evXiTpe8++67OHToEG7cuKEwixwAeHl5ITk5+a329fjxYzx48ABhYWGYNm0at/7FixcQiUQKZV/+Nq/K65ydnbnf27dvDwDIz89Hz549kZmZiTFjxtTZtuvXr6O8vBz+/v4K6ysqKrg5O+pSVlZW57TEDTE1NcXHH3+MgIAA+Pv7Y/DgwRg7dizX3hqGhoYoLS1VqW7S+Ch5kGbD2NgY3bt3B1A99a+vry+WLFmCpUuXvnXdSUlJkMvlAKr/WAGNN6lTVVUVgOpTV69OU6yjo6OwbGxs/Eave/l0Xs1poJrX18TXUNt++eUXdOjQQWFbQxN8mZubc9dMarRq1arWe1jzHteIj4/H7NmzkZycjF27duGrr75Camoq+vXrx5V5+vQp2rVrV+++iXpQ8iDN1uLFizF06FD87W9/g5WV1VvVVdfscA4ODjhz5ozCuleX34SlpSU6dOiAu3fvYsKECY3+ulc5Ozvj119/xZIlS2ptc3BwgFAoRHZ2tkrXGNzc3HD9+nWFde3atUNeXh4YY1wCq5mu9tXXurm5YeHChfDy8sKOHTu45HHnzh2Ul5c32Osh6kHJgzRbPj4+6NWrFyIjIxEXF8d7/bNnz4a3tzeio6MxcuRIpKSkvPUpqxoSiQSzZ89GmzZtMHToUMhkMmRkZEAqlWLu3Lm8v+5lCxcuhJOTE8LDwzFz5kzo6+sjLS0NY8aMgbm5OebPn4+///3vqKqqwoABA1BcXIz09HS0bt1a4aL8ywICAvDJJ5+gsrKS6wX5+Pjg8ePHiI6Oxocffojk5GQcOnSIuz363r172LBhA4KDg2FlZYWsrCzcvHkTkyZN4uo9ceIEunbtim7duin71pImQrfqkmZt7ty52LhxIx48eMB73f369cOmTZsQGxsLV1dXpKSk4KuvvuKl7k8++QSbNm1CQkICnJycMGjQICQkJKBLly6N8rqX2dnZISUlBZcvX0bfvn3h5eWFAwcOQFe3+rvk0qVL8c033yAqKgr29vYICAjA//73vwb3MWzYMOjp6eHIkSPcOnt7e6xZswb/+c9/4OLignPnzmH+/PncdiMjI/zxxx/44IMPYGdnh+nTp+Ozzz7DjBkzuDI//PCDwvUdojloDnNCCC/WrFmDAwcO4PDhw7zUd/XqVfj5+eHmzZu1bggg6kenrQghvJg+fTqkUilKSkpgYmLy1vU9fPgQW7dupcShoajnQQghRGV0zYMQQojKKHkQQghRGSUPQgghKqPkQQghRGWUPAghhKiMkgchhBCVUfIghBCiMkoehBBCVEbJgxBCiMr+H++dPJnd0hu6AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 400x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "d1 = rtt_data_matched[\"ePPing\"]\n",
    "d2 = rtt_data_matched[\"PPing\"]\n",
    "bins = np.arange(-4, 1, 0.05)\n",
    "\n",
    "fig, axes = plt.subplots(figsize=(4, 2.5))\n",
    "ppa_viz.plot_rtt_diff(d1, d2, unit=\"us\", axes=axes, stat_kwargs={\"fmt\": \"{:.3f}\"}, \n",
    "                      density=False, histtype=\"stepfilled\", bins=bins)\n",
    "axes.set_xlim(bins[0], bins[-1])\n",
    "axes.get_legend().remove()\n",
    "fig.savefig(\"PAM_TCP_RTTdiff_ePPing_PPing.eps\", dpi=300, bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 400x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "d1 = rtt_data_matched[\"ePPing\"]\n",
    "d2 = rtt_data_matched[\"tshark\"]\n",
    "\n",
    "bins = np.arange(-1, 7, 0.05)\n",
    "\n",
    "fig, axes = plt.subplots(figsize=(4, 2.5))\n",
    "ppa_viz.plot_rtt_diff(d1, d2, unit=\"us\", axes=axes, stat_kwargs={\"fmt\": \"{:.3f}\"}, \n",
    "                      density=False, histtype=\"stepfilled\", bins=bins)\n",
    "axes.set_xlim(bins[0], bins[-1])\n",
    "axes.get_legend().remove()\n",
    "fig.savefig(\"PAM_TCP_RTTdiff_ePPing_tshark.eps\", dpi=300, bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Multicore-performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "multicpu_folder = os.path.join(data_folder, \"performance_test/multi_core/2022-10-11T093803\")\n",
    "data_multicpu = load_data(multicpu_folder)\n",
    "\n",
    "kpping_err_multicpu = kp_err.load_all_kpping_err(multicpu_folder, omit=20)\n",
    "\n",
    "kpping_err_df_multicpu = prodat.flatten_per_flow_dict(kpping_err_multicpu)\n",
    "kpping_err_df_multicpu = numerize_nflows(kpping_err_df_multicpu)\n",
    "kpping_err_df_multicpu = numerize_run(kpping_err_df_multicpu)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_multicpu[\"cpu_total_norm\"] = data_multicpu[\"cpu_total\"] / 6 # CPU cores on middlebox\n",
    "df = data_multicpu.query(\"n_flows == 10\")\n",
    "df_mean = df.groupby([\"pping_setup\", \"run\"], as_index=False).mean() # Variation between runs is higher than variation within runs - so treat as 10 idenpendent runs instead of 1000 independent values\n",
    "order = [\"baseline\", \"PPing\", \"ePPing\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 250x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(figsize=(2.5, 2.5))\n",
    "sns.barplot(data=df_mean, x=\"pping_setup\", y=\"network_txbps\", order=order, axes=axes)\n",
    "\n",
    "axes.spines[\"right\"].set_visible(False)\n",
    "axes.spines[\"top\"].set_visible(False)\n",
    "\n",
    "axes.set_xlabel(\"\")\n",
    "axes.set_ylabel(\"Throughput (Gbps)\")\n",
    "axes.grid(axis=\"y\")\n",
    "\n",
    "fig.savefig(\"PAM_10flows_allcores_throughput_bar.eps\", bbox_inches=\"tight\", dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pping_setup\n",
       "PPing       56.954062\n",
       "baseline    56.948054\n",
       "ePPing      56.152551\n",
       "Name: network_txbps, dtype: float64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_mean.groupby(\"pping_setup\")[\"network_txbps\"].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 250x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(figsize=(2.5, 2.5))\n",
    "sns.barplot(data=df_mean, x=\"pping_setup\", y=\"cpu_total_norm\", order=order)\n",
    "\n",
    "axes.spines[\"right\"].set_visible(False)\n",
    "axes.spines[\"top\"].set_visible(False)\n",
    "\n",
    "axes.set_xlabel(\"\")\n",
    "axes.set_ylabel(\"CPU (%)\")\n",
    "axes.grid(axis=\"y\")\n",
    "\n",
    "fig.savefig(\"PAM_10flows_allcores_cpu_bar.eps\", bbox_inches=\"tight\", dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pping_setup\n",
       "PPing       77.13711\n",
       "baseline    47.38321\n",
       "ePPing      57.39245\n",
       "Name: cpu_total_norm, dtype: float64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_mean.groupby(\"pping_setup\")[\"cpu_total_norm\"].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "frac_data = kpping_err_df_multicpu.query(\"n_flows == 10 and timestamp < 99\").copy()\n",
    "frac_data[\"fraction\"] = (frac_data[\"processed_packets\"].values / \n",
    "                         df.query(\"n_flows == 10 and pping_setup == 'PPing' and timestamp < 99\")[\"network_pps\"].values)\n",
    "\n",
    "frac_data = pd.concat([frac_data, pd.DataFrame({\"pping_setup\": \"ePPing\", \"run\": [1, 2], \n",
    "                                                \"fraction\": [1.0, 1.0]})]) #ePPing never miss packets\n",
    "frac_data[\"fraction\"] *= 100\n",
    "frac_data_mean = frac_data.groupby([\"pping_setup\", \"run\"], as_index=False).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 180x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(figsize=(1.8, 2.5))\n",
    "sns.barplot(data=frac_data_mean, x=\"pping_setup\", y=\"fraction\", order=order[1:], palette=[\"C1\", \"C2\"])\n",
    "\n",
    "axes.spines[\"right\"].set_visible(False)\n",
    "axes.spines[\"top\"].set_visible(False)\n",
    "\n",
    "axes.set_xlabel(\"\")\n",
    "axes.set_ylabel(\"Processed packets (%)\")\n",
    "axes.grid(axis=\"y\")\n",
    "\n",
    "fig.savefig(\"PAM_10flows_allcores_fracproc_bar.eps\", bbox_inches=\"tight\", dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pping_setup\n",
       "PPing       6.20687\n",
       "ePPing    100.00000\n",
       "Name: fraction, dtype: float64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "frac_data_mean.groupby(\"pping_setup\")[\"fraction\"].mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Baseline performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "singlecore_folder = os.path.join(data_folder, \"performance_test/single_core/2022-10-10T185529\")\n",
    "data = load_data(singlecore_folder)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "kpping_err = kp_err.load_all_kpping_err(singlecore_folder, omit=20)\n",
    "kpping_err_df = prodat.flatten_per_flow_dict(kpping_err)\n",
    "kpping_err_df = numerize_nflows(kpping_err_df)\n",
    "kpping_err_df = numerize_run(kpping_err_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1 = data[[\"n_flows\", \"pping_setup\", \"run\", \"timestamp\", \"network_pps\", \"network_Mpps\"]].copy()\n",
    "df1[\"pping_setup\"] = df1[\"pping_setup\"].map({\"PPing\": \"PPing-fw\", \n",
    "                                             \"ePPing\": \"ePPing\", \"baseline\": \n",
    "                                             \"baseline\"})\n",
    "\n",
    "df2 = kpping_err_df.copy()\n",
    "df2[\"pping_setup\"] = \"PPing-proc\"\n",
    "df2 = df2.rename(columns={\"processed_packets\": \"network_pps\"})\n",
    "df2[\"network_Mpps\"] = df2[\"network_pps\"] / 1e6\n",
    "\n",
    "df = pd.concat([df1, df2], ignore_index=True)\n",
    "df_mean = df.groupby([\"pping_setup\", \"n_flows\", \"run\"], as_index=False).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The PostScript backend does not support transparency; partially transparent artists will be rendered opaque.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 400x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "labels = [\"baseline\", \"PPing-fw\", \"PPing-proc\", \"ePPing\"]\n",
    "ls = [\"-\", \"--\", \":\", \"-\"]\n",
    "colors = [\"C0\", \"C1\", \"C1\", \"C2\"]\n",
    "handles = [axes.plot([], [], c=colors[i], ls=ls[i])[0] for i in range(len(labels))]\n",
    "\n",
    "fig, axes = plt.subplots(figsize=(4, 2.5))\n",
    "interaction_plot(df_mean, \"n_flows\", \"pping_setup\", \"network_Mpps\", ax=axes, \n",
    "                 hue_order=labels, linestyles=ls, palette=colors, dodge=0)\n",
    "\n",
    "axes.legend(handles=handles, labels=labels, bbox_to_anchor=[0.995, 0.02], loc=\"lower right\", ncol=1)\n",
    "axes.set_xlabel(\"No. flows\")\n",
    "axes.set_ylabel(\"Packet rate (Mpps)\")\n",
    "axes.grid(axis=\"y\")\n",
    "axes.set_axisbelow(True) # For some reason the first line (baseline) is drawn behind the gridlines otherwise, regardless of zorder\n",
    "axes.set_ylim(0, 2.05)\n",
    "fig.savefig(\"PAM_performance_overview_packetrate.eps\", bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Nr reports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.0, 131716.963)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 400x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(figsize=(4, 2.5))\n",
    "d = data.query(\"pping_setup != 'baseline'\").groupby([\"pping_setup\", \"n_flows\", \"run\"], as_index=False).mean()\n",
    "interaction_plot(d, \"n_flows\", \"pping_setup\", \"pping_filtered_rtt_events\", ax=axes, \n",
    "                 hue_order=[\"PPing\", \"ePPing\"], palette=[\"C1\", \"C2\"], dodge=0)\n",
    "axes.legend() # Gets rid of legend title\n",
    "axes.set_xlabel(\"No. flows\")\n",
    "axes.set_ylabel(\"RTT reports/s\")\n",
    "axes.grid(axis=\"y\")\n",
    "axes.set_axisbelow(True)\n",
    "axes.set_ylim(0)\n",
    "#fig.savefig(\"performance_overview_reports.pdf\", bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The PostScript backend does not support transparency; partially transparent artists will be rendered opaque.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 400x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(figsize=(4, 2.5))\n",
    "interaction_plot(d, \"n_flows\", \"pping_setup\", \"pping_filtered_rtt_events\", ax=axes, \n",
    "                 hue_order=[\"PPing\", \"ePPing\"], palette=[\"C1\", \"C2\"], dodge=0)\n",
    "axes.legend() # Gets rid of legend title\n",
    "axes.set_xlabel(\"No. flows\")\n",
    "axes.set_ylabel(\"RTT reports/s\")\n",
    "axes.grid(axis=\"y\")#, which=\"both\")\n",
    "axes.set_axisbelow(True)\n",
    "axes.set_yscale(\"log\")\n",
    "fig.savefig(\"PAM_report_frequency_log.eps\", bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pping_setup  n_flows\n",
       "PPing        1             232.257\n",
       "             10           1586.556\n",
       "             100          2458.024\n",
       "             1000         6849.346\n",
       "ePPing       1            1000.003\n",
       "             10           9785.336\n",
       "             100         60428.875\n",
       "             1000       125049.765\n",
       "Name: pping_filtered_rtt_events, dtype: float64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d.groupby([\"pping_setup\", \"n_flows\"])[\"pping_filtered_rtt_events\"].mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Number of processed packets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "pps_table = data.query(\"pping_setup == 'baseline'\").groupby(\"n_flows\")[\n",
    "    [\"network_txpps\", \"network_rxpps\", \"network_pps\"]].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\begin{tabular}{lrrr}\n",
      "\\toprule\n",
      "{} & network\\_txpps & network\\_rxpps & network\\_pps \\\\\n",
      "n\\_flows &               &               &             \\\\\n",
      "\\midrule\n",
      "1       &          1.86 &          0.04 &        1.90 \\\\\n",
      "10      &          1.86 &          0.09 &        1.95 \\\\\n",
      "100     &          1.72 &          0.21 &        1.92 \\\\\n",
      "1000    &          1.64 &          0.29 &        1.93 \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\n"
     ]
    }
   ],
   "source": [
    "ff = lambda x: \"{:.2f}\".format(x)\n",
    "print((pps_table / 1e6).to_latex(formatters={col: ff for col in pps_table.columns}))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Impact of sampling and aggregation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "agg_config_map = {\"individual\": os.path.join(data_folder, \"sampling_rate_impact/individual/2022-10-17T123956\"),\n",
    "                  \"aggregated\": os.path.join(data_folder, \"sampling_rate_impact/aggregated/2022-10-18T201936\")}\n",
    "\n",
    "dfs = list()\n",
    "for agg, path in agg_config_map.items(): \n",
    "    for r in (0, 10, 100, 1000):\n",
    "        tmp_df = load_data(os.path.join(path, \"r_{}\".format(r)))\n",
    "        tmp_df[\"sample limit (ms)\"] = r\n",
    "        tmp_df[\"report_mode\"] = agg\n",
    "        dfs.append(tmp_df)\n",
    "\n",
    "rtest_data = pd.concat(dfs, ignore_index=True)\n",
    "rtest_data_mean = rtest_data.groupby([\"report_mode\", \"sample limit (ms)\", \"n_flows\", \"run\"], as_index=False).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The PostScript backend does not support transparency; partially transparent artists will be rendered opaque.\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEFCAYAAAAPCDf9AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAABSKUlEQVR4nO3dd1xT1/8/8NfNIBCmLIHKEAciUEXctqLWhXVVrW0dxWodv7a2dte2X8UOtba1dlntp3XW0aG4aq22imhVFJA6UFxQRNl7Z53fHxcCYSaYkEDez8cjD5KTm9x3rvF9Ts499xyOMcZACCHELAiMHQAhhJDWQ0mfEELMCCV9QggxI5T0CSHEjFDSJ4QQM0JJnxBCzAglfUIIMSOU9AkhxIxQ0ieEEDNCSZ8QQsyIUZN+dHQ0JkyYAA8PD3Ach3379jX7msrKSrz33nvw9vaGRCJBly5dsGnTJsMHSwgh7YDImDsvLS1Fr1698Nxzz2Hq1KlavWb69OnIzMzEjz/+iK5duyIrKwsKhcLAkRJCSPtg1KQfFhaGsLAwrbc/cuQITp48iTt37sDR0REA4OPjY6DoCCGk/TFq0tfVgQMH0LdvX6xZswbbt2+HtbU1Jk6ciA8//BBWVlYNvqayshKVlZXqxyqVCnl5eXBycgLHca0VOiGEGAxjDMXFxfDw8IBA0HSvfZtK+nfu3MHp06dhaWmJyMhI5OTk4IUXXkBeXl6j/fqrVq3CihUrWjlSQghpfXfv3kWnTp2a3IYzlfn0OY5DZGQkJk+e3Og2o0ePxqlTp5CRkQF7e3sAwN69ezFt2jSUlpY22Nqv29IvLCyEl5cXkpOTYWtrq/fPQQghra24uBidO3dGQUGBOjc2pk219N3d3fHQQw9pfCh/f38wxpCWloZu3brVe41EIoFEIqlX7ujoCDs7O4PGSwghrUEsFgOAVl3WbWqc/pAhQ3D//n2UlJSoy27cuAGBQNDsTxpCCCFGTvolJSVISEhAQkICACA5ORkJCQlITU0FACxduhTPPvusevsZM2bAyckJzz33HBITExEdHY0333wTc+fObfRELiGEkBpGTfqxsbEIDg5GcHAwAOC1115DcHAwli1bBgBIT09XVwAAYGNjg2PHjqGgoAB9+/bFzJkzMWHCBHz11VdGiZ8QQtoakzmR21qKiopgb2+PwsJC6tMnhLQLuuS1NtWnTwgh5MFQ0ieEEDNCSZ8QQsxImxqnT0hdTC5Hwb59kKfehdjLEw6TJ4OrGrNMCKmPkr6JuDNlKiqvXwdUKkAggKRHD/ju3WPssEwGU6kApRJMqQRTKAGlAiqZDLdHjwErL1dvl7lqNfxizlHiJ6QRlPRNAJPLUXo9ESJVVYFKhdLriVCVlgICAZhSBSgVVQlPAahU6sTHlMp6yVBzeyWg4v8ypUK9LZTNlalqntN4j5p9MKUCUCg1Y6guUynVz9WUqQCFoqZMqeJfW11W6zVQKMBUNc9Dy0FmrKwMSQMHQeToCIG1ddVNCqGNDX9fal1TXl1mLVWXCWuVc5aWNCkfaXco6ZuA3D2/QKjSLBOpgKSQvsYJyBgEAnBCISAS8X/FYgiEQs0yoQCcUKQuU2RnQ5GTg9ppmQFARQXkd+8+eExCIQTSqgrBplalUF15qCuN2hVJrW1qVy5SKR83IUZGSd8EXPtnJxwEqGnpg09epZZAgTWgEgBKAaDiqv5WPVYKOKg4/jETcGBCDuAEYMLqG58o+ZuQvwn4+5xIqC7jhEIIRKKqchEEQhE4oQgCUc1foUgMTiTm7wvFEIjEEIqr/oosIKz+KxZDJJZAJLKAUGTB3xfX3BeLJeoysUgCsYUlRCJJs9PBNiTnl93IWlZ/BlXXiP+D4+QpUJWVQVVaCmVJCVSlpVCVllX9LYWqpASqstKax6WlUJZoPq7eTpGZCWQ8+OUsnJWVxi8LoXXDlYbGrw7rhisXzsKixb9CqCvRvFHSNwEZ1hVwUtUvPxzKIStIBIVKDoVKATkHKMBBwXHq+3KOg4JD1WMGBceggApyDmCG6ppQAZBV3fRExIkgEoggFoghEtS/39DjfFEO3hMDlvKa96kQAx/YR6N3UhmsRFaQiqWwkljBytqKfyxyhJWIv28l5v+KBU33/zOVCqy8HEp1RVD1t0yzclBqVBhlfMVSpxJRFhZCmZPz4AdMLIZQKq3/i0KjcqiqXGqVcxIJKm/c4BM+AKhUqLx5E0wup/MgZoKuyDUBe468im6vH4FYWVMmFwK31oZhypi1fIFKBciKgfJ8oLwAqCho4j7/WFlRAEV5IeSyIiiqKofqiqKm0gD/HKorj6r7Ag5ysRQKC2soLKSQi60gF1tBIbaEQiSBQiSBXGQBhUAMhUgMuUAEhUDI3wDIVQoomAJylRzyqkqr+tbUY43nWPPLYAqVDEOvMLjlM2R04BAdyEEp1K2yEwvE6opAKpbWVAoiK0hF0nqVRO0yqUgKK3Gdsqr3sBRaQiio36Wjksnq/ZqoVzmUlGj+MqmzrbKMr1hYWZlOn7UxIldXWHTxhcjRCSJnJwidnCFyqrqvLnOCwMJCL/sj+qVLXqOkbwLk8jIcGxkCryyAYwDjgFRXYNRfcRCLpQ++A6UCqCyqVTHkqysGdSWhriwKNCsReanu+xOIAEsHwMoBsOrQ8H3Lqsd174trJs5jjNVUBEwBuVKurgwUKgUO3zmM9QnrUbdTf1LXSfB38ke5olx9K5OXaT5WVD2W1zyuUFSAQb//HSRCiWbFUVV51C1rtLJpoKKpvnEcB6ZUqrux6lcatX51VFUYZefPo+LmzXrnQbStJgV2dhA5OUHo5AhRVcUgdHaqVVk4QeTMlwukevjuEq3okteoe8cUCCzxTuhqBFjvgtgiG3KZC66WPoNRAkv9vL9QBEgd+ZuuFLIGKoZ87e6XtaAbQyhRVwCcpQPEVh0gbrCScMDz9oH4H2OQ1+rGEjOG5QPebVFlyRhDhbKi8YqisYqjqvLQeFznNfmV+bofiyZw4GApsmy84pBYQWpd+9eHDaxELujoUAKPmzfrvZ/zu+/A7tGhUObmQpGTC0VuDn8/N4+/n5MLRW4ulLm5kCUnA8nJzcdoZcX/WnBygtDZGSJHR76CcHKGyLmqvOp5gZ0djZRqJZT0TcCeuDQUyUQ4K5utUb43Pg1P9fMyUlRVRBaAjSt/05W8vJFfFM10UZXlACUZzb69GEBXj45IsrCACvzl5V3lMohjtwCDXtA5XI7j1ElS31RMhQpFBV8x1Kog6lYUdSuW5iqbElkJ8irytI5DaM2wqYHzIAut92JAdga6OnRFF68u8HUYBEeLhluMqrIyKPLyoMzJgSK3pjJQ1KoYqsvlaWmQp6U1GxcnFvMVQJ2KobpS4CuIqjIHBxoJ9QCoe8cEfHLkOr6Lul2vvFcneyyfGIDenRwgEJhRK4gxQFba/C+KlNNAzk2goS4ZtyDAdxjgOxzwGgRYtN+uBoVKoXXFcfzucVxKj9fqPIirlSt8HXzR1aErfB180cW+C7o4dIG9pOnl+GpjMhkUeXl1KoYcKHOry3L4srw8KPPyak4wN0UggLBDB81zDtXdTNUVQ1V3k8jREZwZnIegPv0mmGLS330+Fe/svdzo8+72lggLdMfjD7sh2LODeVUATYnbChx8uX651Akoy615LLQAvAbWVALuvYAGTrCagz039iDibES98v/X6/+hs31n3Cq4hTsFd3C78DZSi1KhZMp62zpbOaOLQxd1JVB938HS4YFiY0ollAUFUORUVQbVXUsNVBaK3FxALm/+TQEI7O0hcnSs6Waqc4K6drmgicWYTHmoKyX9Jphi0pcrVei57Ajkypp/CrGQw9tje+Do1Uxc+C9PfUGqm50lxga64fGH3RHiZeYVgFIOrPQAlLXGjgotgHfvA4VpwJ0o/pZ8kv91UM2qA9B5aE0l4Ni5lQM3HrlKjkd2PYIyRc2oH6lIitPPnK43dFWmlOG/ov9wu+A2bhfe5v8W8JVBQyOrnCyd0MWhC3zta34ddHXoig6WHfT+ORhjUBUVQZGbV1VB8BWDMq9WN1N191NentajnDiptNZ5iJoT1AIHB2R9sgZQ1qoExWL0iI8ziaGulPSbYIpJHwDGf3UKielFUDFAwAE93e1w6OVHAQCZRRU4ciUDhy+n43xKTQXgaitBWKAbxgW5o6+PI4TmWAFsGApkXgGYEuCEQMdAYFG05jYqJZBxCbh9gq8EUs8Bysqa5x28gS7D+Uqgc2jLTni3IXKVHAdvH8Td4rvwtPXEhC4Tmr1WQeP1SjlfGRTexp2CO/yvg8I7SClKgUJVvzJwtHSEr72v+ldBV4eu8LX3haOlY6udvFWVlVVVDDlQ5uXV/HJo4DyEqqhI6/d1/+hDOEybZsDItUNJvwmmmvS1lVVcgT+vZuLwpXTEJOdCVfWv52IrwdgAvgLo39lMKwBtycqAu+dqKoGMS7We5PjuH99hfEXgORAQ62kUVTsnV8lxt+gubhfeVncT3Sq41Whl4CBxqN9N5NAFTpZORh3Jo5LJ1BVD9TmHgv37UH7+Qr1tnRYsgOtrrxohSk2U9JvQ1pN+bdnFlTiayP8COHu7pgJwtrHAmKoKYEBnR4iEtGxCk0pz+C6g6kqgsNa8PSJL/kRwdSXQMYifyoJoTaFS4G7xXXX3UHV3UXJhMuSq+v3y9hJ7jYqgurvI2crZaJVB/q+/IuP/ltUrp5Z+G9Cekn5tuSWVOJqYicOX03Hmdi6UVTWAk7UFRge4YVyQGwb5OlEF0BzGgLw7wJ0TfCWQfAqoLKx5XurEdwFVVwIORh5S24YpVAqkFadpnC+4XcBXBjJV/Tk+bC1s1V1DtU8gu0pdDV4ZMLkcSQMGapwb4KRSk5nGm5J+E9pr0q8tr1SGY4kZ+P1yBs7cyoGiqgLoIBVjTIAbwoLcMbiLE8RUATRPqQDSE6oqgSjgbgxQu3Xq6MufDPYdBnR+lD9JTB6IUqXEvZJ76nMF1ZXBncI7qKx9LqaKrdiWH1Jap6uoo7SjXisDJpejcP9+yFLvwsLLE/aTJplEwgco6TfJHJJ+bQVlMvUvgNM3ayoAB6kYo3t2RFiQO4Z0cYaFiCoArchKgf/O1IwMyrxS8xwnADyCayoBz/6ASGKkQNsfpUqJ+yX3Nc4ZVHcTlSvK621vLbZusJvIzdqt3V39S0m/CeaW9GsrLJPj2DW+Ajh1M1s9RNTOUqTuAnqkqwtVALoozgSSo2u6g4rv1zwnlgLeg2sqgY4BQDtLNqZAxVR8ZVBnaOmdwjsNVgZSkVRdCdQ+gexu7Q4B1za/+5T0m2DOSb+2wnI5/q6qAKJv5ECm5K+EtLUUYVTPjhgX6I5HuztDIjLPi5hahDH+CuE7UXwlkHyKnxm1mrVL1bUBw/iKwP4hIwVqHlRMhYzSDI2RRNXdRbWvU6hmJbKqd76gi0MXeNh4QMAJIFfJceDWAfVQ14ldJ+o01NWQKOk3gZJ+fUUVchy/loXfL6fj5I1syBRVFYBEhJE9OyIs0A1Du7vAUkwVgE6UCuBeXE0lkHYBqD100albzfUBPo8AltpPb0BajjGGjNKMeieQbxfeRmkDs8paCi3hY+eDmwU3Na5QbuyiNmOgpN8ESvpNK6lUqH8BRCVlo7KqArCRiPCYvyvCAt0xzI8qgBapLAZS/qmpBLKv1zzHCYGHQmoqgU79AKHxk4k5YYwhsyxToxKovl8iL2nwNSsGr8CUblNaOdL6DJr0Kysrcf78eaSkpKCsrAwuLi4IDg5G585t41J2SvraK61U4Pj1LBy+nI4TSVmokPMVgLWFECP8O2JcoBuG+bnCyoIqgBYpSq85IXwnSnNmUQsbwHtITSXg0oPOBxgJYwwrY1Zid9Lues89H/Q8XunzihGi0mSQ+fTPnDmDr7/+Gvv27YNMJoODgwOsrKyQl5eHyspK+Pr6YsGCBVi0aBFsbW0f+EMQ47OWiDChlwcm9PJAmUyBE9ezcfhyOo5fz8LBf+/j4L/3YSUWYkQPV4wLcsfwHi6QWtBs3Vqzcwd6P8PfGONb/tUXiKWcBm7+yd8AwMat1vmAYfxrSavgOA49HHs0+JynrWcrR/PgtGrpT5o0CRcuXMCMGTMwceJE9O3bF9Jaq+LcuXMHp06dwq5du/Dvv/9i27ZtGDVqVLM7j46Oxqeffoq4uDikp6cjMjISkydPbnT7qKgoDB8+vF75tWvX0KNHw/8odVFL/8GVy5SISuLPARy/noUyGd/PaSkWYLgfXwGM6OEKawlVAC2mkAH3YvkK4PYJ/txA7RkvXXrUjAryGQJIqKFlSLpMVGcMeu/e+fbbbzF//nxYaDEv9dWrV3H//n2tkv4ff/yBf/75B3369MHUqVO1TvpJSUkaH8zFxQVCLRdVoKSvXxVyJaKS+F8Af1/LRGlVBSAR8RVAWJAbHvPvCBuqAB5MRSHf+q+uBHJrrX4lEPHnAKorgYdC+NXSiF496ER1htQmT+RyHKd10s/Pz4eDg0OL9kNJ33Aq5EpE3+ArgL+uZaGkkh+pYiESYFh3F4wLcsdj/q6wtTSN/yhtWvXU0dXdQbWXppTY8aOBqisB526a5wM2hgKZV2vNTBoALDzZyh+A6JNB18i9e/cuOI5Dp06dAADnz5/Hzp070bNnTyxYsKBlEesoODgYFRUV6NmzJ95///0Gu3yqVVZWorKy5tLtoqppU+VyOeRaLsJAtCMEMLy7E4Z3d0KlXInTt3Nx5Eom/rqejaOJmTiamAkLkQCPdnXC2ICOeKyHC1UALSXtCAQ+xd+YCshKhCD5JLjkk+BSz4JLOgwkHQYAMFsPsM6hUHUOBfMZClFxBrjqqSSYCqw4Awr6v9Cm6ZLLdG7pP/roo1iwYAFmz56NjIwM+Pn5ISAgADdu3MDLL7+MZcvqz0SnVSBatPSTkpIQHR2NkJAQVFZWYvv27diwYQOioqIwdOjQBl8TERGBFStW1CvfuXOnxnkJYjgKFZBUyCEhl8PlPA7lSr7VKeQYejgw9HZkCHRkkFKPhF4IVHJ0KL0F1+IrcCm+CoeyZHC1lpRUgQMHBg78QpMKgQR/PPwdGEf/AG1VWVkZZsyYYZjunQ4dOuDcuXPw8/PDV199hZ9//hn//PMPjh49ikWLFuHOnTstClqbpN+QCRMmgOM4HDhwoMHnG2rpe3p6Iicnh7p3jECmUOHsnVz8cTUTf13LQmE53wUkFnIY3IX/BTCyhyscpPQLQG/KC8D9dxpcchQE1w+Bq90VBD7xK0esABv0onHiIw+sqKgIzs7OhunekcvlkEj4SaT++usvTJw4EQDQo0cPpKentyDcBzNw4ED89NNPjT4vkUjU8dYmFoshNpEZ8syJWAyMDPDAyAAPyJUqnLmdi8OX0vFnYgZO3sjByRs5+D9BIgZ3dcbjQW4Y3dMNHazb/8LWBiV2AYKe4G/SDsCZr/jVxKpwAETHlwNp54CQOUDXkXQiuI3RJZfp/C8bEBCADRs24PHHH8exY8fw4YcfAgDu378PJycnXd/ugV28eBHu7jRmuS0SCwUI7e6C0O4u+EgZiHN3cnH4cjr+vJqJ6BvZiL6RjXcjr2BwFyeMC3LHmAA3OFIF8GA6dNZI+GoWNsCNP/ibrQcQPAvoM5vWC2iHdO7eiYqKwhNPPIGioiKEh4dj06ZNAIB3330X169fx969e7V+r5KSEty6dQsAf3J27dq1GD58OBwdHeHl5YWlS5fi3r172LZtGwBg3bp18PHxQUBAAGQyGX766SesXr0ae/bswZQp2l0KTaN3TJ9CqUJMch5+v5yOP69kILeUX1BDKOAw0NdRXQE420iaXFuYNKCxxeTfSgFu/wXEbQVuHwff6cMBXUbwrX+/MJoWwoQZfMimUqlEUVEROnSoWTAiJSUFUqkUrq6uWr9PYxdbhYeHY8uWLZgzZw5SUlIQFRUFAFizZg2+//573Lt3D1ZWVggICMDSpUsxbtw4rfdJSb9tUaoYYpL5XwBHrmQip4Q/PyPggH4+johNyYOy1jdYLOSQ+MFYWiCmKc0tJp//H3DxJ/5WPVW0tSvQewbQ51nAqYtx4iaNapVx+llZWUhKSgLHcejevbtOyd6YKOm3XUoVw4WUPPxxOR1/XMlAVnH9VZQA4JOpQXiqH3VLPDClArh1jG/93/yTHxoKAD6P8q1//wm0SIyJMGjSLyoqwosvvojdu3dDqeT7BoVCIZ566il8++23sLc37elhKem3DyoVw2u/JGBfwv16z70wrAveGqvdtBxES0X3gYs7gPhtQGEqX2blCPR6BggJB1z8jBufmdMlr+n8G/j5559HTEwMDh06hIKCAhQWFuLQoUOIjY3F/PnzWxw0IboQCDgM9G144MCFlDwUVdDFRnpl5wGEvgm88i8waw/gPxGoLALOfQt82x/4cQyQsBOQ1V+chJgWnVv61tbW+PPPP/HII49olJ86dQpjx45FaWn9RQhMCbX02w+5UoXeK46q5/upzd3eEqunPozQ7i5GiMxMlGTxiT5+K5BXdX2OxB54eDrf+ncLMm58ZsSg3TteXl74/fffERSk+Q966dIljBs3DmlpabpH3Ioo6bcvcqUKe+PT8F9uGbydpBjm54qPfr+Gg//y3T5P9fXEe+P9YUfTPRgOY/xkcHFbgGsHakYGefThk3/gVJoF1MAMmvS///57/Prrr9i2bZt6fHxGRgbCw8MxZcoULFy4sOWRtwJK+ubhyJV0vL/vCnJKZNTqb01lecC/u/nWf/XKYBY2fOIPCecrAloMRu8MmvSDg4Nx69YtVFZWwsuLHyGRmpoKiUSCbt26aWwbHx+vY+iGR0nffOSXyrD8wFUcoFZ/62MMuHueb/1fjQQU5Xx5xyA++Qc9CVg5GDPCdsWgSb+hycsas3z5cl3eulVQ0jc/R65kVLX6K+Fub4mVU4Iw3K9tDDFuF8oLgMu/8q3/jMt8mcgKCJjMD/30HECt/wfUJufTby2U9M1TfqkMEQevYn/VEM/pfTvhvcd7wt6KWv2thjHg/kW+9X9lDyCrWmzc2Y9v/fd6BpA6GjXEtqpVkn5sbCyuXbsGjuPg7++PkJCQFgXb2ijpm7c/r2bgvUi+1e9mZ4lVU6nVbxSVJXzij9/KLwUJ8NNB+E/gW/8+j1LrXwcGTfppaWl45pln8M8//6hXryooKMDgwYOxa9cueHqa9kLBlPRJ3Vb/kyGd8P54avUbTcYVPvn/+zNQWciXOfryUz70ngnYUKXcHIMm/dGjR6OoqAhbt26Fnx9/FV5SUhLmzp0La2trHD16tOWRtwJK+qRavVb/lCAM70EJxmhkZUDifr4CSD3LlwlE/GRvfeYAXYYDAu3WwjY3Bk36VlZWOHPmDIKDgzXK4+PjMWTIEJSXl+secSuipE9qKyiTIeLAVfV0DtNCOuH/qNVvfNlJ/JQPCTuB8jy+zN6Ln+45eBZ/hTBRM+g0DF5eXg2ux6hQKPDQQw/p+naEGJWD1ALrng7G97ND4GIrwW9xaRj9xUmcuJ5l7NDMm4sfMOZj4PXrwNQfgc5D+Tl/TnwMfBEA7HwKuH6YnxSO6ETnlv7+/fuxcuVKfPvttwgJCQHHcYiNjcXixYvx9ttv67zcYWujlj5pTEGZDCsOJiLy4j0A1Oo3Obm3gYvb+YnfSqsqZVt3vuUfPBvo4G3c+IzIoN07HTp0QFlZGRQKBUQifuGt6vvW1tYa2+bl5ekYuuFR0ifNOZaYiXcjLyO7uBId7SRYNSUII3p0NHZYpJpSDiT9wff93/obNQu+DAf6hAN+4wCRea2wZtCkv3XrVq23DQ8P1+WtWwUlfaKNuq3+qX06Ydn4nrCnBdtNS0Eqv9hL/PZaC764VC34Em42C77QxVlNoKRPdEGt/jZCqQBu/cW3/m/8ya8KBvDj/fuE8+P/xZbGjdGAKOk3gZI+0VVBmQwfHErE3ni+1T+lz0NYPj6AWv2mqigdSPiJH/1TUL3gSwfg4af5K39d/Y0bnwEYJOkLhdqNj61eTctUUdInLfVXVas/q7gSrrZ8q/8xf2r1myyVCkiO4qd9uP47oKoa6eM5gG/9BzwBWEiNGaHeGCTpCwQCeHt7Izw8vN4Y/domTZqkW7StjJI+eRCFZXKsOHSVWv1tTUk28O9Ofr3fvNt8mcSOX/ClTzjg/rBx43tABkn6Fy5cwKZNm7B792507twZc+fOxcyZM9GhQwe9BN1aKOkTffj7WiaW7q1p9a98Iggje1Kr3+RVL/gSvxVIPAAoK/lyj2A++QdNa5MLvhi0T7+iogK//fYbNm/ejHPnzmHChAmYN28eRo0a9UBBtxZK+kRfCsvk+OBQIvbE86vFTQl+CMsnUKu/zSjLAy79zLf+s6/xZWJrIHAKP+nbQyFtZtK3VjuRm5ycjHnz5uHkyZPIzs6Go6PpT4tKSZ/o2/HrfKs/s4ha/W1S9YIv8VuBK3trLfgSyLf+H57OL/iyYSiQeYUfGcQJ+ecXRRs19GoGT/ppaWnYsmULtmzZgvLycsyePRsfffSR+mItU0ZJnxhCYZkcH/6eiN/i+Fb/E8EPYfmEnnCQmtdFQm1eRSG/4EvcViDjEl8msgT8JwJX99acDAb4qaDfvQ8Ijf/LziBJXyaTITIyEj/++CNOnTqFsLAwzJ07F+PGjYNAoPMUPkZDSZ8YUu1Wv0tVq38UtfrbpuoFXy7/VrPgS10Tv+angDYygyR9Jycn2NraIjw8HLNnz4ara8NT0Jp6IqWkTwytsFyODw9Rq7/dqCwBfp0D3DqmWS4QAYNfBkYaf1lYgw3ZVL+ogZMbjDFwHEfj9AmpcuJ6Ft7Ze4la/e1B3Fbg4Mv1y9tzS//kyZNa7Tw0NFSr7QAgOjoan376KeLi4pCeno7IyEitZ+n8559/EBoaisDAQCQkJGi9T0r6pDUVlsvx0aFE/FrV6p/c2wMREwOo1d/WKOXASg9AKaspa6N9+lqfedUlmWurtLQUvXr1wnPPPYepU6dq/brCwkI8++yzeOyxx5CZman3uAjRF3srMT59shfGBblj6d7L2JdwH6dv5WLlE4EYHeBm7PCItoRiwMVfc/SOi79JJHxdadXSLy0trTdtsj63B/guI21b+k8//TS6desGoVCIffv2UUuftAmF5XJ8/HsifonlW/2TensgYkIAOlhTq588GL239Lt27YrFixdjzpw58PBoeJkyxhj++usvrF27FkOHDsXSpUt1j1wLmzdvxu3bt/HTTz/ho48+anb7yspKVFZWqh8XFRUBAORyeYMrgFVTKpVQKBQws/noSBWO4yASibSec0obUhHw8aSeGO3vgvf2J2J/wn38cysHH0zoiVE9aW1e0nJN5bK6tEr6UVFReP/997FixQr07t0bffv2hYeHBywtLZGfn4/ExEScPXsWYrEYS5cuxYIFC1ocfFNu3ryJd955B6dOndL6moBVq1ZhxYoV9cqPHj0KqbThyZZsbW1ha2vbpoaiEv1TqVQoLi5GcXGx3t/7VT8gMkWAmGwZXtiVgBBnFab6qGDd9noLiAkoKyvTeludLs5KS0vDr7/+iujoaKSkpKC8vBzOzs4IDg7GmDFjHmjMfnPdO0qlEgMHDsS8efOwaNEiAEBERESz3TsNtfQ9PT2Rk5PT4M+gzMxMFBUVwcXFBVKptMGRSqT9Y4yhrKwM2dnZsLOzQ8eOhhl1E30zB+/tu4qMoko421hQq5+0SFFREZydndvWfPrNJf2CggJ06NBB4+e2SqUCYwxCoRBHjx7FiBEjmt1PU31fSqUSN27cgKurK5ycnB7o85D2ITc3F1lZWejevbteu3pqK6qQ4+ND1/Bz7F0AwMReHlgxkfr6ifYMMnrH2Ozs7HD58mWNsvXr1+P48eP47bff0Llz5wfeR3W/WGPdPsT8VH8X5HK5wZK+naUYn0x7GGFBbli69zIO/HsfZ27n4KPJQRgbSCN8iH4ZNemXlJTg1q1b6sfJyclISEiAo6MjvLy8sHTpUty7dw/btm2DQCBAYGCgxutdXV1haWlZr/xBUZcOqdaa34Vhfq7489WhWPn7Ney+cBeLforDhKpWvyO1+omeGPVMZWxsLIKDg9WLsrz22msIDg7GsmXLAADp6elITU01ZoiEtCo7SzFWT30YW+f2h7u9JQ7+ex+jvziJI1fSjR0aaSeMmvSHDRsGxli925YtWwAAW7ZsQVRUVKOvj4iI0GmMfns1Z84cra9kboyPjw/WrVunfsxxHPbt22e0uJKSkuDm5maQkTPVpk2bhrVr1xrs/R9EaHcX/PnqUDzdzxM5JTIs+ikei3ddRF6prPkXE9IEGpNIGpSeno6wsLAHeo8vv/xSXYEDfCW/ZMkSrV773nvv4cUXX4StreFWMVq2bBk+/vhj9bUbpqa61b9tbn94UKuf6EmLkv6pU6cwa9YsDBo0CPfu8WuFbt++HadPn9ZrcMR43NzcIJFIHug97O3t4eDgoPPr0tLScODAATz33HMPtP/mPPzww/Dx8cGOHTsMup8HNbSq1f9M/5pW/0s746nVT1pE56S/Z88ejBkzBlZWVrh48aJ6DHxxcTFWrlyp9wBNyW+//YagoCBYWVnByckJI0eORGlpKQB+DeFRo0bB2dkZ9vb2CA0NRXx8vMbrOY7Dxo0bMX78eEilUvj7++Ps2bO4desWhg0bBmtrawwaNAi3b99WvyYiIgK9e/fGxo0b4enpCalUiieffBIFBQWNxskYw5o1a+Dr6wsrKyv06tULv/32m06ftXb3TkpKCjiOwy+//IJHH30UVlZW6NevH27cuIELFy6gb9++sLGxwdixY5Gdna1+j9rdO3PmzMHJkyfx5ZdfguM4cByHlJSUBvf9yy+/oFevXujUqZO6bMuWLXBwcMChQ4fg5+cHqVSKadOmobS0FFu3boWPjw86dOiAxYsXa8z0un79enTr1g2Wlpbo2LEjpk2bprGviRMnYteuXTodG2OwtRRj1ZSaVv+hS+kYtfYk/rhMrX6iI6aj3r17s61btzLGGLOxsWG3b99mjDF28eJF1rFjR13frtUVFhYyAKywsLDec+Xl5SwxMZGVl5fXe+7+/ftMJBKxtWvXsuTkZHbp0iX27bffsuLiYsYYY3///Tfbvn07S0xMZImJiWzevHmsY8eOrKioSP0eANhDDz3Efv75Z5aUlMQmT57MfHx82IgRI9iRI0dYYmIiGzhwIBs7dqz6NcuXL2fW1tZsxIgR7OLFi+zkyZOsa9eubMaMGeptwsPD2aRJk9SP3333XdajRw925MgRdvv2bbZ582YmkUhYVFRUo8fF29ubffHFFxqxRkZGMsYYS05OZgDU71kdZ58+fdiwYcPY6dOnWXx8POvatStbtGhRg3EVFBSwQYMGsfnz57P09HSWnp7OFApFg7FMmjRJ430YY2zz5s1MLBazUaNGsfj4eHby5Enm5OTERo8ezaZPn86uXr3KDh48yCwsLNju3bsZY4xduHCBCYVCtnPnTpaSksLi4+PZl19+qfG+hw8fZhKJhFVUVDQYS1PfCWMpKpexd/ZcYt5vH2Lebx9iL+yIYznFDcdPzENTea0unZO+lZUVS05OZoxpJv3bt28ziUSi69u1upYm/bi4OAaApaSkaLUfhULBbG1t2cGDB9VlANj777+vfnz27FkGgP3444/qsl27djFLS0v14+XLlzOhUMju3r2rLvvjjz+YQCBg6enpjDHN5FpSUsIsLS3ZmTNnNOKZN28ee+aZZxqNV5uk/8MPP2jECYD9/fff6rJVq1YxPz8/9eO6lVFoaCh75ZVXGo2hWq9evdgHH3ygUbZ582YGgN26dUtdtnDhQiaVStUVL2OMjRkzhi1cuJAxxtiePXuYnZ2dRsVb17///tvkv6spJv1q0Tey2OBVfzPvtw+xPh8cZb9fum/skIiR6JL0de7ecXd31xhbX+306dPw9fVt2c+NNqBXr1547LHHEBQUhCeffBL/+9//kJ+fr34+KysLixYtQvfu3WFvbw97e3uUlJTUG3L68MMPq+9XX9ofFBSkUVZRUaFxctHLy0ujq2PQoEFQqVRISkqqF2diYiIqKiowatQo2NjYqG/btm3T6DZqCW1iz8rKeqB9AEB5eTksLS3rlUulUnTp0kVjfz4+PrCxsWkwhlGjRsHb2xu+vr6YPXs2duzYUW+OEisrKwC6zV1iKh7t5oIjSx7FM/29kFsqwws74vHijnjkllQ2/2JitnRO+gsXLsQrr7yCmJgYcByH+/fvY8eOHXjjjTfwwgsvGCJGkyAUCnHs2DH88ccf6NmzJ77++mv4+fkhOTkZAN9nHRcXh3Xr1uHMmTNISEiAk5MTZDLNk21icc2MWtUX/jRUplKpGo2lepuGLhyqft3vv/+OhIQE9S0xMVHnfv26tIm9qbi15ezsrFGhNrT/6v01VFYdg62tLeLj47Fr1y64u7tj2bJl6NWrl8b5kLy8PACAi4vLA8dtDHxffxB+mjcADzlY4ffL6Rj1RTR+v0R9/aRhOif9t956C5MnT8bw4cNRUlKCoUOH4vnnn8fChQvx0ksvGSJGk8FxHIYMGYIVK1bg4sWLsLCwQGRkJAB+RNPLL7+McePGISAgABKJBDk5OXrZb2pqKu7fv69+fPbsWQgEAnTv3r3etj179oREIkFqaiq6du2qcfP09NRLPC1lYWGh1XKawcHBSExM1Ms+RSIRRo4ciTVr1uDSpUtISUnB8ePH1c9fuXIFnTp1grOzs172ZyyPdHPGn68OxYwBXsgrleHFnXyrP4da/aSOFk3D8PHHH+O9995DYmIiVCoVevbsqfETuz2KiYnB33//jdGjR8PV1RUxMTHIzs6Gv78/AH7Nge3bt6Nv374oKirCm2++qe46eFCWlpYIDw/HZ599hqKiIrz88suYPn063Nzqz8tia2uLN954A6+++ipUKhUeeeQRFBUV4cyZM7CxsUF4eLheYmoJHx8fxMTEICUlBTY2NnB0dGxwVtYxY8bg+eefh1KpfKD5bg4dOoQ7d+5g6NCh6NChAw4fPgyVSgU/Pz/1NqdOncLo0aNbvA9TYiMRYeUTQRgX6I6391zC75fTcfZOLj6cFIjHH3Y3dnjEROjc0p87dy6Ki4shlUrRt29f9O/fHzY2NigtLcXcuXMNEaNJsLOzQ3R0NMaNG4fu3bvj/fffx+eff66+gGnTpk3Iz89HcHAwZs+ejZdffhmurvqZIrdr166YMmUKxo0bh9GjRyMwMBDr169vdPsPP/wQy5Ytw6pVq+Dv748xY8bg4MGDepmU7kG88cYbEAqF6NmzJ1xcXBqdYmPcuHEQi8X466+/Hmh/Dg4O2Lt3L0aMGAF/f39s2LABu3btQkBAAACgoqICkZGRmD9//gPtx9RUt/pn1mr1v7Ajjlr9BEALplYWCoVIT0+vl9BycnLg5uYGhUKh1wD1rakpSCsqKpCcnIzOnTs3eCLRGLRZM6A9Wr9+Pfbv348///zTYPv49ttvsX//fhw9erTRbUzxO6GLf27l4K3fLuFeQTkcrS3wwaQAjH+44dXvSNtlkKmVi4qK1HPjFBcXa/wHUCqVOHz4sN5atoQsWLAA+fn5KC4uNthUDGKxGF9//bVB3ttUDOnKt/pXHb6GHTGpeGnnRSzdewmllUqoGCDggJ7udjj08qPGDpW0Eq2TvoODg/pKyoZOIHIc1+CyhIS0hEgkwnvvvWfQfRhqWU9TYyMR4eMngjAuyB1v/vov7hdWqJ9TMSApsxhypQpiIU3FZQ60TvonTpwAYwwjRozAnj174OjoqH7OwsIC3t7ejS6aTlouIiICERERxg6DtANDujpjQWgXRBy4qlEuVzLsjU/DU/28jBQZaU1aJ/3Q0FAA/EInnp6etGg4IW1QZlFFg+X/5ba9i9NIy+g8ZNPb2xsAfwVjampqvYuPal+1SQgxLd6ODS8FWlxh2gMwiP7onPSzs7Px3HPP4Y8//mjweW0uviGEGMfUkE748FAiSmWa/093n0/FyJ4dEdq9bV6ZTLSncx/NkiVLkJ+fj3PnzsHKygpHjhzB1q1b0a1bNxw4cMAQMRJC9EQsFCBh+Wh8MjUILwzrgk+mBuGjyYGQqxgWbo9FzJ1cY4dIDEznlv7x48exf/9+9OvXDwKBAN7e3hg1ahTs7OywatUqPP7444aIkxCiJ2KhoN5J20qFCh8eSsS8rbHY8fwA9PJ0ME5wxOB0bumXlpaqx+M7OjqqF80ICgqqt2gIIaRtmPdIZ7w+qjtKKhV4dtN5XEs3zSUkyYPTOen7+fmpp/StXtHp3r172LBhA9zdaX4PQtqql0Z0xcJQXxSWyzH7xxjcyS4xdkjEAFrUp5+ezk/bunz5chw5cgReXl746quv2v1yieZu/fr16ukIQkJCcOrUKWOHRPSI4zi8M7YHZg/0Rk6JDDN/iMHdPBrK2d7o3Kc/c+ZM9f3g4GCkpKTg+vXr8PLyavPT05LG/fzzz1iyZAnWr1+PIUOGYOPGjQgLC0NiYiK8vOiinvaC4zismBiAMpkSe+LTMOvHGPyycBA62rW9eYdIw3Rq6cvlcvj6+mrMdS6VStGnTx9K+O3c2rVrMW/ePDz//PPw9/fHunXr4Onpie+++87YoRE9Ewg4fDI1CI8HueO/3DLM+iGGVuNqR3RK+mKxGJWVlQ2u2ERaj1ypwu7zqfjkyHXsPp8KufLBV6tqikwmQ1xcXL1550ePHo0zZ84YdN/EOERCAb54qjdG9HDFzawSPLvpPArL5cYOi+iBzn36ixcvxieffGLyUyi3V3KlCrN+iME7ey/jf9F38M7ey5j1Q4xBE39OTg6USqV6XdxqHTt2REZGhsH2S4zLQiTA+pl9MMjXCVfvF+G5zedRWkn/79s6nfv0q1eQOnr0KIKCgmBtba3x/N69e/UWHKlvT1waYpL5dV0VKn4phJjkvFaZMKvuLzzGGP3qa+csxUL8EN4Xs3+MQXxqAeZvi8WmOf1gKW75imbEuHRO+g4ODpg6daohYiFa+C+vDCIBp074ACAScAadMMvZ2RlCobBeqz4rK6te65+0P9YSETY/1x/PfH8OZ27n4oUd8dgwKwQWIpp0sS3SOelv3rxZbzuPjo7Gp59+iri4OKSnpyMyMhKTJ09udPvTp0/j7bffxvXr11FWVgZvb28sXLgQr776qt5iMnXejlKNhA/wLX5vp4Yn0tIHCwsLhISE4NixY3jiiSfU5ceOHcOkSZMMtl9iOuytxNg+rz+e+v4cjl/Pwqs/J+CrZ4IhFNAvvbbGqFV1aWkpevXqhW+++Uar7a2trfHSSy8hOjoa165dw/vvv4/3338f33//vYEjNR1TQzphQGd+LQNR1X+4AZ0dMaVPJ4Pu97XXXsMPP/yATZs24dq1a3j11VeRmpqKRYsWGXS/xHQ42Ujw07wB8HKU4vfL6Xh7zyWoVDqttkpMgM4tfX0KCwtTLyyujeDgYAQHB6sf+/j4YO/evTh16pTZrIIkFgrw0/MDsDc+Df/llsHbSYopfToZfNWjp556Crm5ufjggw+Qnp6OwMBAHD58WD3VNjEPbvaW2PH8AEzfeBa/xaXB2kKIiIkBdG6nDTFq0n9QFy9exJkzZ/DRRx81uk1lZSUqK2vGGBcV8XOKyOVyyOWaQ9DkcjkYY1CpVFCpDDsM8kEIOeDJEM2WfWvEu2jRonote1M+TvqgUqnAGINcLodQSCcvAcDNVowt4SGY8eMFbD37HyQiDm+M6kaJ34jq5rKmtMmk36lTJ2RnZ0OhUCAiIgLPP/98o9uuWrWqwbV7jx49CqlUsx9cJBLBzc0NJSUl9RaHIeZJJpOhvLwc0dHRNEy5jnldgG+uCvH9qRTcS7mN0Z2oq8dYysq0H8jBMcZ0+pfatm0bnnrqKUgkEo1ymUyG3bt349lnn9Xl7WoC4bhmT+RWS05ORklJCc6dO4d33nkH33zzDZ555pkGt22ope/p6YmcnBzY2dlpbFtRUYG7d+/Cx8cHlpZ02TnhvxMpKSnw9PSk70QD/k0rRPjmWJTKlHg3zA/PDabuPmMoKiqCs7MzCgsL6+W1unRO+kKhEOnp6erplavl5ubC1dW1xStn6ZL0a/voo4+wfft29cyfzSkqKoK9vX2DB6eiogLJycnqScUIoe9E82Lu5CJ883lUyFVYPSUIT/enuZhaW1N5rS6dz/41dkFOWloa7O3tdX27B8YY02jJE0Ja1wBfJ2yc3RdiIYelkZexP+GesUMiTdC6Tz84OBgcx4HjODz22GMQiWpeqlQqkZycjLFjx+q085KSEty6dUv9ODk5GQkJCXB0dISXlxeWLl2Ke/fuYdu2bQCAb7/9Fl5eXujRowcAftz+Z599hsWLF+u0X0KIfoV2d8HXz/TBizvj8dov/8JKLMToADdjh0UaoHXSr+52SUhIwJgxY2BjY6N+zsLCAj4+PjpfqRsbG4vhw4erH7/22msAgPDwcGzZsgXp6elITU1VP69SqbB06VIkJydDJBKhS5cuWL16NRYuXKjTfgkh+jc20A2fP9kLr/6SgJd2XsQP4X0xlBZaNzk69+lv3boVTz31VJvt36Q+faIL+k7obmdMKt6NvAxLsQDb5g5A/6qLCYnhGLRPPzw8HBUVFfjhhx+wdOlS5OXxk3/Fx8fj3j3qyyPE3M0Y4IX3H/dHhVyFuVsu4FJagbFDIrXonPQvXbqE7t2745NPPsFnn32GgoICAEBkZCSWLl2q7/gIIW3Q84/64rVaC61fz6CF1k2Fzkn/1VdfxZw5c3Dz5k2Nn7thYWGIjo7Wa3CEkLZr8YiuWDjUFwVlcsz64TwttG4idE76sbGxDZ44feihh2hBjXYsOjoaEyZMgIeHBziOw759+zSeZ4whIiICHh4esLKywrBhw3D16lXjBEtMAsdxeCesB2YN9EJOSSVm/RCDtHxaaN3YdE76lpaW6vlraktKSoKLC52pb6+amxF1zZo1WLt2Lb755htcuHABbm5uGDVqFIqLi1s5UmJKOI7DBxMDMaXPQ7hfWIGZP8Qgq6jC2GGZNZ2T/qRJk/DBBx+oJ/jhOA6pqal45513aHGVdiwsLAwfffQRpkyZUu85xhjWrVuH9957D1OmTEFgYCC2bt2KsrIy7Ny50wjRElMiEHBYM/VhjAtyw3+5ZZj5QwzySmluK2PROel/9tlnyM7OhqurK8rLyxEaGoquXbvC1tYWH3/8sSFiJHUp5UDcVuCvCP6v0rgLVicnJyMjI0Nj4XSJRILQ0FBaOJ0A4BdaX/dUMIb7ueBmVglm/xhDC60bic6zbNrZ2eH06dM4fvw44uPjoVKp0KdPH4wcORI6DvknLaGUA9smA/+dBgQiQKUALv0CPLsPEIqNElL1uZyGFk7/77//jBESMUEWIgG+mxWCOZvP49ydPMzdcgHb5/WH1KJNTvbbZunc0l+1ahUAYMSIEXjjjTfw1ltvYeTIkVAqlZgxY4beAyR1JOzkEz7AJ3yAf/zvLuPFVIUWTifN4Rda74dgLwfE/ZeP+dtiUSFv2SSNpGV0Tvrr1q2rtzyhUqnE008/jYSEBH3FRRqTn8y38GsTiIC8ZOPEA8DNjZ9jhRZOJ9qwkYiwZU5/9HS3wz+3cvHijnjIle17MR5TonPSP3z4MN5++2388ssvAPgVW5588klcvXoVJ06c0HuApI4OnWta+NVUCsCxs3HiAdC5c2e4ubnh2LFj6jKZTIaTJ09i8ODBRouLmC57Kb/QeldXG/x9PQtLfk6AktbbbRU6d6aFhIQgMjISkyZNgkQiwY8//ojbt2/jxIkT1KprDb1n8H34tfv0vR8BejW8iIy+NDcj6pIlS7By5Up069YN3bp1w8qVKyGVSqnLjzSqeqH1Jzeewe+X0mElFmLN1IchEFCXoEGxFtq/fz8TiUQsKCiIZWdnt/RtWl1hYSEDwAoLC+s9V15ezhITE1l5ebkRItOBQsZY3FbGjkXwfxUyg+/yxIkTDEC9W3h4OGOMMZVKxZYvX87c3NyYRCJhQ4cOZZcvXzZ4XIbWZr4TbVhqbikb8PFfzPvtQ2zZvstMpVIZO6Q2p6m8VpdWs2w2NDYbAM6dO4euXbvC2dlZXbZ37149VEWGQ7NsEl3Qd6J13M4uwVMbzyKnRIYXhnXBW2N7GDukNkWXWTa16t5pbEWsMWPG6B4dIYTU0cXFBtvnDcDT35/D+qjbsJaI8OLwrsYOq13SKulv3rzZ0HEQQsycv7sdts7tj5n/O4dP/0yClViIuY8Yb4BCe6Xz6J3k5GTcvHmzXvnNmzeRkpKij5gIIWaqt6cDfpzTDxKRAB8cSsTPF1KbfxHRic5Jf86cOQ1eWh8TE4M5c+boIyZCiBkb6OuEjbNDIBZyeGcvLbSubzon/YsXL2LIkCH1ygcOHEgXZxFC9GKYnyu+fqYPBByH1375F0ev0rTt+qJz0uc4rsHpcgsLC6FU0uXUhBD9GBvohs+efBgqxvDSzos4dTPb2CG1Czon/UcffRSrVq3SSPBKpRKrVq3CI488otfgCCHm7YngTvhociBkShXmb4vFhZQ8Y4fU5ul8Re6aNWswdOhQ+Pn54dFHHwUAnDp1CkVFRTh+/LjeAySEmLeZA7xRLlPio9+vYe7mC9gxfwAe7uRg7LDaLJ1b+j179sSlS5cwffp0ZGVlobi4GM8++yyuX7+OwMBAQ8RICDFzzz/qi1dHdkdx1ULrSRm0IltLtWgiaw8PD6xcuVLfsRBCSKNefqwrymQKbIy+g5k/xODXRYPQ2dna2GG1OTq39KuVlZXh+vXruHTpksaNtE/6WBi9srISixcvhrOzM6ytrTFx4kSkpaW14qcgbVndhdZn/u8cLbTeAjon/ezsbIwfPx62trYICAhAcHCwxo20T/pYGH3JkiWIjIzE7t27cfr0aZSUlGD8+PE06otoTb3QejAttN5ius7mNmPGDDZ48GB2/vx5Zm1tzY4ePcq2b9/O/Pz82KFDh3SfHq6VtYtZNo0MAIuMjFQ/VqlUzM3Nja1evVpdVlFRwezt7dmGDRsYY4wVFBQwsVjMdu/erd7m3r17TCAQsCNHjrRa7Lqi74RpkiuUbNH2WOb99iE2am0Uyy2pNHZIRqXLLJs6t/SPHz+OL774Av369YNAIIC3tzdmzZqFNWvWqJdS1FZzXQZ17d27F6NGjYKLiwvs7OwwaNAg/Pnnn7p+hDZPrpJjz409WBe3Dntu7IFcZfoLo8fFxUEul2ts4+HhgcDAQFo8nehMJBTgy6eDMczPBTcyS/DsphgUVdBC69rQOemXlpbC1dUVAODo6IjsbP6CiaCgIMTHx+v8Xk11GdQVHR2NUaNG4fDhw4iLi8Pw4cMxYcIEXLx4UbcP0YbJVXIsOLoAEWcjsOXqFkScjcCCowuMmvibWhi9+rmMjAxYWFigQ4cOjW5DiC4sRAJsmBWCgb6OuHKvCM9tvoAymaL5F5o5nUfv+Pn5ISkpCT4+Pujduzc2btwIHx8fbNiwAe7u7jq9V1hYGMLCwrTeft26dRqPV65cif379+PgwYNmcz7hwK0DiM2MBQAoGd8XHpsZi4O3D2JKt4bXPWgtLVkYXZttCGlM9ULrs36IUS+0/mN4P1iKhcYOzWTpnPSXLFmC+/fvAwCWL1+OMWPGYMeOHbCwsMCWLVv0HV+TVCoViouL4ejo2Og2lZWVqKysVD8uKioCwK/tK5drto7lcjkYY1CpVFCpTHOh5tSiVAg5oTrhA4CQEyK1KLVVY659jKp/+d2/f1+jtZ+ZmQlXV1eoVCq4urpCJpMhNzdXo7WflZWFQYMGmezxVqlUYIxBLpdDKKREYookAuCH2cGYtSkW/9zKxQs/xeGbZ3pBLGzx4MQ2p24ua4rOSX/mzJnq+8HBwUhJScH169fh5eWlsYJWa/j8889RWlqK6dOnN7rNqlWrsGLFinrlR48ehVQq1SgTiURwc3NDSUkJZDKZ3uPVB2exs0bCB/gWv7PIWV2htYby8nL1/pycnNCxY0ccOnQIXbp0AVCzMHpERASKiorQrVs3iMViHDhwAE888QQAvsvnypUrWLZsWavGrguZTIby8nJER0dDoaCuA1M2qxPwdaEQx5OyMfPro3i2mwrmstxuWZn2Q1e1TvplZWV48803sW/fPsjlcowcORJfffUVnJ2d0adPnxYF+iB27dqFiIgI7N+/X93SbMjSpUvx2muvqR8XFRXB09MTo0ePbnC5xLt378LGxsZkl8abHjAdf9//G3FZceoWf4hrCJ4MeBJigdhg+627MHpmZibu3LmjsTD66tWrERgYiG7dumHVqlWwtrbG3LlzYWtrCzs7O8ydOxfLli1Dp06d4OjoiLfeegtBQUGYOHGiybaiKyoqYGVlhaFDh5rsd4LUCB1egWd+uICLueXw9e6ElZMCzGKhdZ0aTdoOCXrjjTeYVCpl8+fPZ4sXL2bOzs5s2rRpLR5iVBfqDANsyu7du5mVlVWLhoi2hyGbMqWM7bmxh62LW8f23NjDZMq2sTB6eXk5e+mll5ijoyOzsrJi48ePZ6mpqQaP/UG0le8EqVF7ofXl+6+YxULrugzZ1Drp+/r6sl27dqkfx8TEMJFIxBQKRcuirBuIlkl/586dzNLSUusKoq72kPRJ66HvRNt0M7OY9fngKPN++xBbc+SascMxOIOM07979656Vk0A6N+/P0QikfqkbkuUlJQgISFBvfhKcnIyEhISkJrKL5G2dOlSPPvss+rtd+3ahWeffRaff/45Bg4ciIyMDGRkZKCwsLDFMRBC2p+urvxC63aWInx74ja+PXGr+ReZCa2TvlKphIWFhUaZSCR6oJNbsbGxGtM3vPbaawgODsayZcsAAOnp6eoKAAA2btwIhUKBF198Ee7u7urbK6+80uIYCCHtU08PfqF1awshPv0zCZv/STZ2SCaBY4wxbTYUCAQICwuDRCJRlx08eBAjRoyAtXXNTHd79+7Vf5R6VFRUBHt7exQWFjZ4Ijc5ORmdO3emk3YEAH0n2oNzd3IRvuk8KhUqrJn6MKb38zR2SHrXVF6rS+vRO+Hh4fXKZs2apXt0hBDSigb6OmHD7BAs2BaLt/degqWFEBN7eRg7LKPROulv3rzZkHEQQojBDPdzxdfPBOPFnRfx2s8JsBILMapnx+Zf2A6ZzyVrhBCzNjbQHZ89+TCUjOHFHfE4fTPH2CEZBSV9QojZoIXWKekTQszMzAHeeG+cP8rlSszdfAGX08xryDclfUKI2Zk/1BdLRnZDcaUCszfFmNVC65T0iVZaa43c/Px8zJ49G/b29rC3t8fs2bNRUFBg4E9HzNErj3XDgqG+KCiTY9aPMUjOKTV2SK2Ckj7RSmutkTtjxgwkJCTgyJEjOHLkCBISEjB79myDfz5ifjiOw9KwHpg5wAvZxWa00Lqh54QwNTT3zoODgdbITUxMZADYuXPn1NucPXuWAWDXr1838KdqGH0n2j+lUsVe3X2Reb99iIWuOc4yC9vev7VB18glpC59rZF79uxZ2NvbY8CAAeptBg4cCHt7e1pHlxiMQMBhzbSHMTbADSm5ZZj1YwzyS01zPQ19oKTfBjG5HPm//oqsz9ci/9dfwXRYNccQ9LVGbkZGRoNrI7i6utI6usSgREIBvnqm9kLr59vtQus6r5xFjIvJ5UidOw9lFy4AIhGgUKDowEF4bfoRnNhwi6hoQx9r5Da0vTbvQ8iDql5oPXzTecQk52Hu5gvYNq8/pBbtK01SS7+NKdi3j0/4AFA1w2nZhQso3L/faDG5ubkBQL3WeFZWlrr17+bmBplMhvz8/Ca3yczMrPf+2dnZ9X5FEGIIlmIhfpzTD709HRD7Xz4WbItDhVzZ/AvbEEr6bYw89S7fwq9NJIIs9a5xAgLQuXNnuLm54dixY+qy6jVyBw8eDAAICQmBWCzW2CY9PR1XrlxRbzNo0CAUFhbi/Pnz6m1iYmJQWFio3oYQQ7ORiLD1uf7wd7fD6Vs5eGnnRciVKmOHpTeU9NsYsZenuoWvplDAwsuw08U2teANx3FYsmQJVq5cicjISFy5cgVz5syBVCrFjBkzAAD29vaYN28eXn/9dfz999+4ePEiZs2ahaCgIIwcORIA4O/vj7Fjx2L+/Pk4d+4czp07h/nz52P8+PHw8/Mz6OcjpDZ7qRjb5/WHr4s1/rqWidd++RdKlVaz0Ju89tVZZQYcJk9G0YGDGn360n79YD9pkkH3Gxsbi+HDh6sfVy82Hx4eji1btuCtt95CeXk5XnjhBeTn52PAgAE4evQobG1t1a/54osvIBKJMH36dJSXl+Oxxx7Dli1bNBZF37FjB15++WX1KJ+JEyc2em0AIYbkbCPBzucH4smNZ3Dw3/uQCDmEeDsiNb8M3o5STA3pBLGw7bWbtV5Epb1oD4uoMLkchfv3Q5Z6FxZenrCfNMnoJ3Hbq7bynSCGczevDFO/+wdZxZrDOK0thEhYPtokEr9BFlEhpoMTi+EwbZqxwyDELHg6SjFroA/WHruhUV4qU2JvfBqe6udlpMhaxvhVFCGEmLjyRkbw/Jfb9qZtoKRPCCHN8HaUNlzu1HC5KaOkTwghzZga0gnWFkKNMmsLIab06WSkiFqO+vQJIaQZYqEACctHY298Gv7LLYO3kxRT+rTN0TuU9BtgZgOaSBPou0CqiYWCNnfStiFtr5oyIHHVsMeysrZ3coYYRvV3QUxDYkk7QS39WoRCIRwcHJCVlQUAkEqlNNGXmWKMoaysDFlZWXBwcNC4gIyQtoySfh3Vk4dVJ35i3hwcHNTfCULaA0r6dXAcB3d3d7i6ukJu5HnqiXGJxWJq4ZN2h5J+I4RCIf2HJ4S0O0Y9kRsdHY0JEybAw8MDHMdh3759TW6fnp6OGTNmwM/PDwKBAEuWLGmVOAkhpL0watIvLS1Fr169tJ5FsbKyEi4uLnjvvffQq1cvA0dHCCHtj1G7d8LCwhAWFqb19j4+Pvjyyy8BAJs2bTJUWIQQ0m61+z79yspKVFZWqh8XFhYCAPLy8uhELSGkXSguLgag3cWE7T7pr1q1CitWrKhX3rlzZyNEQwghhlNcXAx7e/smt2n3SX/p0qXqVZ4AQKVSIS8vD05OTiZ34VVRURE8PT1x9+7dZhdCIDXouOmOjlnLmOpxY4yhuLgYHh4ezW7b7pO+RCKBRCLRKHNwcDBOMFqys7MzqS9UW0HHTXd0zFrGFI9bcy38ajT3DiGEmBGjtvRLSkpw69Yt9ePk5GQkJCTA0dERXl5eWLp0Ke7du4dt27apt0lISFC/Njs7GwkJCbCwsEDPnj1bO3xCCGlzjJr0Y2NjMXz4cPXj6r738PBwbNmyBenp6UhNTdV4TXBwsPp+XFwcdu7cCW9vb6SkpLRKzIYkkUiwfPnyet1RpGl03HRHx6xl2sNx4xhNGE4IIWaD+vQJIcSMUNInhBAzQkmfEELMCCV9QggxI5T0TYCuU0ybo+aOEWMMERER8PDwgJWVFYYNG4arV68aJ1gj0ccxqqysxOLFi+Hs7Axra2tMnDgRaWlprfgpDKu1jlF+fj5mz54Ne3t72NvbY/bs2SgoKDDwp9MOJX0ToOsU0+aouWO0Zs0arF27Ft988w0uXLgANzc3jBo1Sj0RlTnQxzFasmQJIiMjsXv3bpw+fRolJSUYP348lEpla30Mg2qtYzRjxgwkJCTgyJEjOHLkCBISEjB79myDfz6tMGJSALDIyEhjh2HS6h4jlUrF3Nzc2OrVq9VlFRUVzN7enm3YsMEIERpfS45RQUEBE4vFbPfu3ept7t27xwQCATty5Eirxd5aDHWMEhMTGQB27tw59TZnz55lANj169cN/KmaRy190uYlJycjIyMDo0ePVpdJJBKEhobizJkzRozMdGhzjOLi4iCXyzW28fDwQGBgoFkcR30do7Nnz8Le3h4DBgxQbzNw4EDY29ubxHGkpE/avIyMDABAx44dNco7duyofs7caXOMMjIyYGFhgQ4dOjS6TXumr2OUkZEBV1fXeu/v6upqEseRkj5pN+pOlc0YM7nps42tJcfI3I6jPo5RQ9ubynGkpE/aPDc3NwCo14rKysqq12ozV9ocIzc3N8hkMuTn5ze6TXumr2Pk5uaGzMzMeu+fnZ1tEseRkj5p8zp37gw3NzccO3ZMXSaTyXDy5EkMHjzYiJGZDm2OUUhICMRiscY26enpuHLlilkcR30do0GDBqGwsBDnz59XbxMTE4PCwkKTOI7tfhGVtqC5KaZJ88doyZIlWLlyJbp164Zu3bph5cqVkEqlmDFjhhGjbl0Peozs7e0xb948vP7663BycoKjoyPeeOMNBAUFYeTIkcb6WHrVGsfI398fY8eOxfz587Fx40YAwIIFCzB+/Hj4+fm1/oeuy5hDhwjvxIkTDEC9W3h4uLFDMxnNHSOVSsWWL1/O3NzcmEQiYUOHDmWXL182btCtTB/HqLy8nL300kvM0dGRWVlZsfHjx7PU1FQjfBrDaK1jlJuby2bOnMlsbW2Zra0tmzlzJsvPz2+lT9k0mlqZEELMCPXpE0KIGaGkTwghZoSSPiGEmBFK+oQQYkYo6RNCiBmhpE8IIWaEkj4hhJgRSvqEGMC+ffvQtWtXCIVCLFmyBFu2bIGDg4OxwyKEkj4xD3PmzAHHcVi9erVG+b59+wwy8+HChQsxbdo03L17Fx9++KHe35+QlqKkT8yGpaUlPvnkk3ozJOpbSUkJsrKyMGbMGHh4eMDW1tag+yNEF5T0idkYOXIk3NzcsGrVqia327NnDwICAiCRSODj44PPP/9c631ERUWpk/yIESPAcRyioqIa3Pa7775Dly5dYGFhAT8/P2zfvl393Ouvv44JEyaoH69btw4cx+H3339Xl/n5+akn9IqKikL//v1hbW0NBwcHDBkyBP/995/WcRPzQUmfmA2hUIiVK1fi66+/RlpaWoPbxMXFYfr06Xj66adx+fJlRERE4P/+7/+wZcsWrfYxePBgJCUlAeArj/T09Aan042MjMQrr7yC119/HVeuXMHChQvx3HPP4cSJEwCAYcOG4dSpU1CpVACAkydPwtnZGSdPngTAz/l+48YNhIaGQqFQYPLkyQgNDcWlS5dw9uxZLFiwwCQW7CAmyNgzvhHSGsLDw9mkSZMYY4wNHDiQzZ07lzHGWGRkJKv932DGjBls1KhRGq998803Wc+ePbXeV35+PgPATpw4oS7bvHkzs7e3Vz8ePHgwmz9/vsbrnnzySTZu3DjGGL8At0AgYLGxsUylUjEnJye2atUq1q9fP8YYYzt37mQdO3ZkjPEzOgJgUVFRWsdIzBe19InZ+eSTT7B161YkJibWe+7atWsYMmSIRtmQIUNw8+ZNKJVKvcXQ2H6uXbsGgJ+3vXfv3oiKisLly5chEAiwcOFC/PvvvyguLkZUVBRCQ0MBAI6OjpgzZw7GjBmDCRMm4Msvv0R6erreYiXtCyV9YnaGDh2KMWPG4N133633HGtgHVNmoNnHm1uLddiwYYiKisLJkycRGhqKDh06ICAgAP/88w+ioqIwbNgw9babN2/G2bNnMXjwYPz888/o3r07zp07Z5C4SdtGSZ+YpdWrV+PgwYM4c+aMRnnPnj1x+vRpjbIzZ86ge/fuEAqFetu/v79/g/vx9/dXP67u1z9+/Lg6wYeGhmL37t3q/vzagoODsXTpUpw5cwaBgYHYuXOn3uIl7YiRu5cIaRW1+/SrzZ49m1laWmr06cfFxTGBQMA++OADlpSUxLZs2cKsrKzY5s2b1duMGDGCff31143uS5s+/cjISCYWi9l3333Hbty4wT7//HMmFAo1XlPdry8UCtmVK1cYY4zt27ePCYVC5uLiot7uzp077J133mFnzpxhKSkp7M8//2SOjo5s/fr1uh0kYhYo6ROz0FDST0lJYRKJhNVt+/z222+sZ8+eTCwWMy8vL/bpp59qPO/t7c2WL1/e6L60SfqMMbZ+/Xrm6+vLxGIx6969O9u2bVu99woJCWEuLi5MpVIxxviTthzHsWnTpqm3ycjIYJMnT2bu7u7MwsKCeXt7s2XLljGlUtnEESHmipZLJIQQM0J9+oQQYkYo6RNCiBmhpE8IIWaEkj4hhJgRSvqEEGJGKOkTQogZoaRPCCFmhJI+IYSYEUr6hBBiRijpE0KIGaGkTwghZoSSPiGEmJH/D8yKKahkkCkXAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 400x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 400x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i, agg in enumerate([\"individual\", \"aggregated\"]):\n",
    "    fig, axes = plt.subplots(figsize=(4, 2.5))\n",
    "    \n",
    "    d = rtest_data_mean.query(\"report_mode == @agg\")\n",
    "    interaction_plot(d, \"n_flows\", \"sample limit (ms)\", \"network_Mpps\", ax=axes, dodge=0)\n",
    "    axes.set_xlabel(\"No. flows\")\n",
    "    axes.set_ylabel(\"Packet rate (Mpps)\")\n",
    "    axes.grid(axis=\"y\")\n",
    "    axes.set_axisbelow(True) # For some reason the first line (baseline) is drawn behind the gridlines otherwise, regardless of zorder\n",
    "    axes.set_ylim(1.1, 1.6)\n",
    "    if (i > 0):\n",
    "        axes.legend().remove()\n",
    "    fig.savefig(\"PAM_r_test_{}.eps\".format(agg), bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "sample limit (ms)  n_flows\n",
       "0                  1          1.547766e+06\n",
       "                   10         1.481198e+06\n",
       "                   100        1.347177e+06\n",
       "                   1000       1.135781e+06\n",
       "10                 1          1.562592e+06\n",
       "                   10         1.548978e+06\n",
       "                   100        1.495129e+06\n",
       "                   1000       1.415567e+06\n",
       "100                1          1.562454e+06\n",
       "                   10         1.558305e+06\n",
       "                   100        1.549477e+06\n",
       "                   1000       1.516134e+06\n",
       "1000               1          1.564298e+06\n",
       "                   10         1.567258e+06\n",
       "                   100        1.556887e+06\n",
       "                   1000       1.538977e+06\n",
       "Name: network_pps, dtype: float64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rtest_data.query(\"report_mode == 'individual'\").groupby([\"sample limit (ms)\", \"n_flows\"])[\"network_pps\"].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "sample limit (ms)  n_flows\n",
       "0                  1          1.561145e+06\n",
       "                   10         1.553300e+06\n",
       "                   100        1.524449e+06\n",
       "                   1000       1.452821e+06\n",
       "10                 1          1.561747e+06\n",
       "                   10         1.562042e+06\n",
       "                   100        1.549921e+06\n",
       "                   1000       1.513076e+06\n",
       "100                1          1.559455e+06\n",
       "                   10         1.558755e+06\n",
       "                   100        1.554222e+06\n",
       "                   1000       1.536773e+06\n",
       "1000               1          1.559478e+06\n",
       "                   10         1.562633e+06\n",
       "                   100        1.559820e+06\n",
       "                   1000       1.542927e+06\n",
       "Name: network_pps, dtype: float64"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rtest_data.query(\"report_mode == 'aggregated'\").groupby([\"sample limit (ms)\", \"n_flows\"])[\"network_pps\"].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id=\"tcp_timestamps_vs_seqack\"></a>\n",
    "### TCP timestamp vs seq/ack no. differences\n",
    "For this analysis we actually use PPing (and not ePPing) due to PPing being able to run on packet traces. That way we can ensure all differences are due to the different algorithms (matching TSval/TSecr vs SEQ/ACK, and ePPing uses same algorithm as PPing), without it being affected by the minor difference in clocks between tcpdump and ePPing (already analyzed in the RTT accuracy section). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 274 ms, sys: 59 ms, total: 333 ms\n",
      "Wall time: 31.8 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "tcp_ex_folder = os.path.join(data_folder, \"rtt_error_analysis/2022-10-28T101527\")\n",
    "tool_rtts = pta_viz.load_all_rtt_data(tcp_ex_folder, pping_path=pping_path, srcip=ip_target, \n",
    "                                      tools=[\"tshark\", \"PPing\", \"ePPing\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 5min 29s, sys: 137 ms, total: 5min 29s\n",
      "Wall time: 5min 29s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "pcap_rtts = pra.calculate_rtts_from_pcap(pta_viz.get_pcap_file(tcp_ex_folder))\n",
    "\n",
    "pcap_rtts = pta_viz._filter_and_format_data(pcap_rtts)\n",
    "ref = pcap_rtts[\"time\"].min() - np.timedelta64(int(tool_rtts[\"tshark\"][\"timestamp\"].min() * 1e9), \"ns\")\n",
    "pcap_rtts[\"timestamp\"] = util.normalize_timestamps(pcap_rtts[\"time\"], ref)\n",
    "pcap_rtts[\"rtt\"] = pcap_rtts[\"max_rtt\"] # RTT with delayed ACK component"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Warning: 9323 entries from tshark missing in common set\n",
      "Warning: 3 entries from PPing missing in common set\n",
      "Warning: 1 entries from ePPing missing in common set\n",
      "Warning: 9323 entries from pcap_rtt missing in common set\n"
     ]
    }
   ],
   "source": [
    "tool_rtts[\"pcap_rtt\"] = pcap_rtts\n",
    "matching_rtts = tra.filter_comparable_rtts(tool_rtts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tshark == minimum rtt for each ACK\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"tshark == minimum rtt for each ACK\")\n",
    "all(matching_rtts[\"tshark\"][\"rtt\"] == matching_rtts[\"pcap_rtt\"][\"min_rtt\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "any(matching_rtts[\"pcap_rtt\"][\"retrans\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 10min 6s, sys: 220 ms, total: 10min 6s\n",
      "Wall time: 10min 7s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "unsync_err = pra.find_unsync_tsval_update(pta_viz.get_pcap_file(tcp_ex_folder), verbose=False)\n",
    "unsync_err_tsecr = [err[\"TSecr\"] for err in unsync_err[\"errors\"]]\n",
    "\n",
    "retrans_err = pra.find_too_fast_retrans(pta_viz.get_pcap_file(tcp_ex_folder), verbose=False)\n",
    "retrans_err_tsecr = [err[\"TSecr\"] for err in retrans_err[\"errors\"]] +\\\n",
    "                        [err[\"TSecr\"] for err in retrans_err[\"weak_errors\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 400x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "bins = np.arange(0, 200, 5)\n",
    "fig, axes = plt.subplots(figsize=(4, 2.5))\n",
    "ppa_viz.plot_rtt_diff(matching_rtts[\"pcap_rtt\"], matching_rtts[\"tshark\"], unit=\"us\", axes=axes, \n",
    "                      stat_kwargs={\"funcs\": (\"count\", \"min\", \"median\", \"max\"), \"fmt\":\"{:.0f}\"}, \n",
    "                      density=False, histtype=\"stepfilled\", stats_label=\"\", bins=bins)\n",
    "axes.get_legend().remove()\n",
    "axes.set_xlim(bins[0], bins[-1])\n",
    "axes.set_xlabel(\"Delayed ACK latency (us)\")\n",
    "fig.savefig(\"PAM_tcp_rtt_diff_delack_example.eps\", bbox_inches=\"tight\", dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 400x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "unsync_mask = matching_rtts[\"PPing\"][\"tsecr\"].isin(unsync_err_tsecr)\n",
    "\n",
    "bins = np.arange(0, 200, 5)\n",
    "fig, axes = plt.subplots(figsize=(4, 2.5))\n",
    "ppa_viz.plot_rtt_diff(matching_rtts[\"PPing\"].loc[unsync_mask], matching_rtts[\"pcap_rtt\"].loc[unsync_mask], unit=\"us\", axes=axes, \n",
    "                      stat_kwargs={\"funcs\": (\"count\", \"min\", \"median\", \"max\"), \"fmt\":\"{:.0f}\"}, \n",
    "                      density=False, histtype=\"stepfilled\", stats_label=\"\", bins=bins)\n",
    "axes.get_legend().remove()\n",
    "axes.set_xlim(bins[0], bins[-1])\n",
    "axes.set_xlabel(\"RTT overestimation (us)\")\n",
    "fig.savefig(\"PAM_tcp_rtt_diff_edgecase2_example.eps\", bbox_inches=\"tight\", dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}