4a_hardware_utilization_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Section 4a: Hardware Utilization Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pyspark.sql.functions as F\n",
    "from util.read_and_print_df import *\n",
    "from util.handle_nan import *\n",
    "from util.extract_json_attributes import *\n",
    "from util.plotting import *\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "from matplotlib.ticker import FormatStrFormatter\n",
    "import matplotlib.patches as mpatches\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Assigning 541 GB of memory per spark driver and executor, and use 126 cores.\n"
     ]
    }
   ],
   "source": [
    "spark = get_spark_session()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+------+---------+-------------------------+--------------------+------------------+--------------+---------------------+---------+--------------------+----------------------+-------------+--------------------------------+---------+-------------------------+----------------+------------+---------+----------------------+-------------+\n",
      "|node  |cpu_model|cpu_core_count_per_socket|cpu_threads_per_core|cpu_tdp_per_socket|cpu_t_case_max|cpu_per_core_temp_max|cpu_count|cpu_core_count_total|cpu_memory_bytes_total|cpu_tdp_total|node_filesystem_size_bytes_total|gpu_model|gpu_memory_bytes_per_card|gpu_tdp_per_card|gpu_temp_max|gpu_count|gpu_memory_bytes_total|gpu_tdp_total|\n",
      "+------+---------+-------------------------+--------------------+------------------+--------------+---------------------+---------+--------------------+----------------------+-------------+--------------------------------+---------+-------------------------+----------------+------------+---------+----------------------+-------------+\n",
      "|r10n1 |gold_6130|16                       |1                   |125.0             |87.0          |101.0                |1        |16                  |1.03079215104E11      |125.0        |1.912248332288E12               |n/a      |0.0                      |0.0             |0.0         |0        |0.0                   |0.0          |\n",
      "|r10n10|gold_6130|16                       |1                   |125.0             |87.0          |101.0                |1        |16                  |1.03079215104E11      |125.0        |1.912248332288E12               |n/a      |0.0                      |0.0             |0.0         |0        |0.0                   |0.0          |\n",
      "|r10n11|gold_6130|16                       |1                   |125.0             |87.0          |101.0                |1        |16                  |1.03079215104E11      |125.0        |1.912248332288E12               |n/a      |0.0                      |0.0             |0.0         |0        |0.0                   |0.0          |\n",
      "|r10n12|gold_6130|16                       |1                   |125.0             |87.0          |101.0                |1        |16                  |1.03079215104E11      |125.0        |1.912248332288E12               |n/a      |0.0                      |0.0             |0.0         |0        |0.0                   |0.0          |\n",
      "|r10n13|gold_6130|16                       |1                   |125.0             |87.0          |101.0                |1        |16                  |1.03079215104E11      |125.0        |1.912248332288E12               |n/a      |0.0                      |0.0             |0.0         |0        |0.0                   |0.0          |\n",
      "+------+---------+-------------------------+--------------------+------------------+--------------+---------------------+---------+--------------------+----------------------+-------------+--------------------------------+---------+-------------------------+----------------+------------+---------+----------------------+-------------+\n",
      "only showing top 5 rows\n",
      "\n"
     ]
    }
   ],
   "source": [
    "df_node_hardware_info = spark.read.parquet(path_node_hardware_info)\n",
    "df_node_hardware_info.show(5, False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+-----+---------+-------------------+-----------------+-----------+----------------+---+------------------------+-----------------------+---------------------------+----------+----------+------------------------+----------------------------+-------------------------+-----------------------+-----------------+------------------------+----------------------------+---------------+-----------------------+------------------------+----------------------+-----------------------------+--------------------------+------------------+-------------------------+-------------------------+------------------+----------------+---------------------------+----------------------------+---------------------------+-------------------------------+-------------------------+-------------------------------------+--------------------------------+---------------------------------+------------------------------+--------------------------+---------------------------+--------------------------+------------------------------+--------------------------------+--------------------------------+--------------------------------+--------------------------------+--------------------------------+----------------------------------+-----------------------------------+----------------------------------+--------------------+-------------------------------+--------------------------------+-------------------------------+------------------------------+-----------------------------------+-----------------------------------+-----------------------------------+-----------------------------------+-------------------------------------+-------------------------------------+--------------------------------------+-------------------------------------+------------------------------+------------------------------------+-------------------------+--------------------------+-------------------------+---------------------------------------+-------------------+------------------------------------+--------------------------------------+----------------------------------------+--------------------+----------------------------------+-----------------------------------+-----------------+-------------------------------------------+---------------------------------------+------------------------------------+------------------------------------------+---------------------------------------------+------------------------------------------+--------------------------------------------+----------------------------------------------+----------------------------------------+-----------------------------------------+-----------------------+---------------------------+--------+---------+-------------------------+--------------------+------------------+--------------+---------------------+---------+--------------------+----------------------+-------------+--------------------------------+----------------+-------------------------+----------------+------------+---------+----------------------+-------------+\n",
      "|node |id       |timestamp          |node_time_seconds|node_load15|node_power_usage|up |node_netstat_Tcp_OutSegs|node_netstat_Tcp_InErrs|node_context_switches_total|node_load5|node_load1|node_memory_Active_bytes|node_netstat_Tcp_RetransSegs|node_netstat_Udp_InErrors|node_memory_Dirty_bytes|node_ambient_temp|node_netstat_Icmp_InMsgs|node_netstat_Udp_InDatagrams|node_intr_total|node_netstat_Tcp_InSegs|node_memory_Percpu_bytes|node_boot_time_seconds|node_netstat_Udp_OutDatagrams|node_netstat_Icmp_InErrors|node_procs_blocked|node_netstat_Icmp_OutMsgs|node_memory_MemFree_bytes|node_procs_running|node_forks_total|node_hwmon_temp_celsius_min|node_hwmon_temp_celsius_mean|node_hwmon_temp_celsius_max|node_filesystem_avail_bytes_sum|node_filesystem_files_sum|node_network_transmit_bytes_total_sum|node_filesystem_device_error_sum|node_disk_written_bytes_total_sum|node_filesystem_free_bytes_sum|node_thermal_zone_temp_min|node_thermal_zone_temp_mean|node_thermal_zone_temp_max|node_disk_read_bytes_total_sum|nvidia_gpu_memory_used_bytes_sum|nvidia_gpu_temperature_celsius_0|nvidia_gpu_temperature_celsius_1|nvidia_gpu_temperature_celsius_2|nvidia_gpu_temperature_celsius_3|nvidia_gpu_temperature_celsius_min|nvidia_gpu_temperature_celsius_mean|nvidia_gpu_temperature_celsius_max|node_arp_entries_sum|nvidia_gpu_fanspeed_percent_min|nvidia_gpu_fanspeed_percent_mean|nvidia_gpu_fanspeed_percent_max|node_filesystem_files_free_sum|nvidia_gpu_power_usage_milliwatts_0|nvidia_gpu_power_usage_milliwatts_1|nvidia_gpu_power_usage_milliwatts_2|nvidia_gpu_power_usage_milliwatts_3|nvidia_gpu_power_usage_milliwatts_min|nvidia_gpu_power_usage_milliwatts_sum|nvidia_gpu_power_usage_milliwatts_mean|nvidia_gpu_power_usage_milliwatts_max|node_filesystem_size_bytes_sum|node_disk_writes_completed_total_sum|nvidia_gpu_duty_cycle_min|nvidia_gpu_duty_cycle_mean|nvidia_gpu_duty_cycle_max|node_network_transmit_packets_total_sum|node_udp_queues_sum|node_network_receive_bytes_total_sum|node_network_receive_packets_total_sum|node_network_receive_multicast_total_sum|node_disk_io_now_sum|node_rapl_package_joules_total_sum|node_network_receive_drop_total_sum|timestamp_seconds|node_network_transmit_bytes_total_sum_delta|node_disk_written_bytes_total_sum_delta|node_disk_read_bytes_total_sum_delta|node_disk_writes_completed_total_sum_delta|node_network_transmit_packets_total_sum_delta|node_network_receive_bytes_total_sum_delta|node_network_receive_packets_total_sum_delta|node_network_receive_multicast_total_sum_delta|node_rapl_package_joules_total_sum_delta|node_network_receive_drop_total_sum_delta|timestamp_seconds_delta|node_rapl_package_power_sum|gpu_node|cpu_model|cpu_core_count_per_socket|cpu_threads_per_core|cpu_tdp_per_socket|cpu_t_case_max|cpu_per_core_temp_max|cpu_count|cpu_core_count_total|cpu_memory_bytes_total|cpu_tdp_total|node_filesystem_size_bytes_total|gpu_model       |gpu_memory_bytes_per_card|gpu_tdp_per_card|gpu_temp_max|gpu_count|gpu_memory_bytes_total|gpu_tdp_total|\n",
      "+-----+---------+-------------------+-----------------+-----------+----------------+---+------------------------+-----------------------+---------------------------+----------+----------+------------------------+----------------------------+-------------------------+-----------------------+-----------------+------------------------+----------------------------+---------------+-----------------------+------------------------+----------------------+-----------------------------+--------------------------+------------------+-------------------------+-------------------------+------------------+----------------+---------------------------+----------------------------+---------------------------+-------------------------------+-------------------------+-------------------------------------+--------------------------------+---------------------------------+------------------------------+--------------------------+---------------------------+--------------------------+------------------------------+--------------------------------+--------------------------------+--------------------------------+--------------------------------+--------------------------------+----------------------------------+-----------------------------------+----------------------------------+--------------------+-------------------------------+--------------------------------+-------------------------------+------------------------------+-----------------------------------+-----------------------------------+-----------------------------------+-----------------------------------+-------------------------------------+-------------------------------------+--------------------------------------+-------------------------------------+------------------------------+------------------------------------+-------------------------+--------------------------+-------------------------+---------------------------------------+-------------------+------------------------------------+--------------------------------------+----------------------------------------+--------------------+----------------------------------+-----------------------------------+-----------------+-------------------------------------------+---------------------------------------+------------------------------------+------------------------------------------+---------------------------------------------+------------------------------------------+--------------------------------------------+----------------------------------------------+----------------------------------------+-----------------------------------------+-----------------------+---------------------------+--------+---------+-------------------------+--------------------+------------------+--------------+---------------------+---------+--------------------+----------------------+-------------+--------------------------------+----------------+-------------------------+----------------+------------+---------+----------------------+-------------+\n",
      "|r28n3|106894969|2022-10-11 04:17:00|1.66546E9        |3.98       |672.0           |1.0|1.94794E9               |0.0                    |1.05184E12                 |4.2       |4.16      |2.35916E10              |271230.0                    |0.0                      |16384.0                |25.0             |423.0                   |7272060.0                   |1.41723E10     |1.43194E9              |8.20511E7               |1.66134E9             |7401900.0                    |198.0                     |0.0               |419.0                    |1.57859E11               |4.0               |3.75065E7       |26.0                       |29.19                       |36.0                       |2.564558897152E12              |6.8497933E8              |8.46171547648E12                     |0.0                             |1.705329791488E12                |2.564612583424E12             |31.0                      |33.5                       |36.0                      |3.7710436352E10               |1.5011938304E10                 |59.0                            |32.0                            |35.0                            |34.0                            |32.0                              |40.0                               |59.0                              |229.0               |41.0                           |41.0                            |41.0                           |6.84435874E8                  |104137.0                           |16758.0                            |27738.0                            |21007.0                            |16758.0                              |169640.0                             |42410.0                               |104137.0                             |2.602202996736E12             |7.3244452E7                         |0.0                      |15.0                      |60.0                     |1.956308731E9                          |0.0                |2.6822646841991E13                  |3.843764456E9                         |0.0                                     |0.0                 |348559.42                         |0.0                                |1665461820       |89727.0                                    |37376.0                                |0.0                                 |8.0                                       |97.0                                         |19008.0                                   |111.0                                       |0.0                                           |2675.9099999999744                      |0.0                                      |30.0                   |89.19699999999915          |1       |gold_5118|12                       |1                   |105.0             |81.0          |101.0                |2        |24                  |2.06158430208E11      |210.0        |2.602202996736E12               |NVIDIA_TITAN_RTX|2.5769803776E10          |280.0           |89.0        |4        |1.03079215104E11      |1120.0       |\n",
      "|r28n3|106894970|2022-10-11 04:17:30|1.66546E9        |3.89       |672.0           |1.0|1.94794E9               |0.0                    |1.05184E12                 |3.92      |3.04      |2.35326E10              |271230.0                    |0.0                      |16384.0                |25.0             |423.0                   |7272060.0                   |1.41723E10     |1.43197E9              |8.20511E7               |1.66134E9             |7401910.0                    |198.0                     |0.0               |419.0                    |1.56576E11               |22.0              |3.75068E7       |25.0                       |28.23                       |34.0                       |2.564558897152E12              |6.8497933E8              |8.461716147725E12                    |0.0                             |1.705329832448E12                |2.564612583424E12             |30.0                      |31.5                       |33.0                      |3.7710436352E10               |1.5011938304E10                 |56.0                            |32.0                            |35.0                            |35.0                            |32.0                              |39.5                               |56.0                              |229.0               |40.0                           |40.75                           |41.0                           |6.84435874E8                  |75301.0                            |16863.0                            |27634.0                            |21074.0                            |16863.0                              |140872.0                             |35218.0                               |75301.0                              |2.602202996736E12             |7.3244457E7                         |0.0                      |0.0                       |0.0                      |1.956313201E9                          |0.0                |2.682418133668E13                   |3.843935316E9                         |0.0                                     |0.0                 |350707.96                         |0.0                                |1665461850       |671245.0                                   |40960.0                                |0.0                                 |5.0                                       |4470.0                                       |1.534494689E9                             |170860.0                                    |0.0                                           |2148.5400000000373                      |0.0                                      |30.0                   |71.61800000000125          |1       |gold_5118|12                       |1                   |105.0             |81.0          |101.0                |2        |24                  |2.06158430208E11      |210.0        |2.602202996736E12               |NVIDIA_TITAN_RTX|2.5769803776E10          |280.0           |89.0        |4        |1.03079215104E11      |1120.0       |\n",
      "|r28n3|106894971|2022-10-11 04:18:00|1.66546E9        |3.8        |672.0           |1.0|1.94794E9               |0.0                    |1.05184E12                 |3.64      |2.23      |2.358E10                |271230.0                    |0.0                      |8192.0                 |25.0             |423.0                   |7272060.0                   |1.41724E10     |1.43197E9              |8.20511E7               |1.66134E9             |7401910.0                    |198.0                     |0.0               |419.0                    |1.71775E11               |18.0              |3.75068E7       |26.0                       |28.08                       |33.0                       |2.564558897152E12              |6.8497933E8              |8.461716223075E12                    |0.0                             |1.705329928704E12                |2.564612583424E12             |30.0                      |31.5                       |33.0                      |3.7710436352E10               |1.5011938304E10                 |54.0                            |32.0                            |35.0                            |34.0                            |32.0                              |38.75                              |54.0                              |229.0               |41.0                           |41.0                            |41.0                           |6.84435874E8                  |74283.0                            |16835.0                            |27373.0                            |21099.0                            |16835.0                              |139590.0                             |34897.5                               |74283.0                              |2.602202996736E12             |7.3244472E7                         |0.0                      |1.5                       |6.0                      |1.956313281E9                          |0.0                |2.6824181349974E13                  |3.843935417E9                         |0.0                                     |0.0                 |352836.64                         |0.0                                |1665461880       |75350.0                                    |96256.0                                |0.0                                 |15.0                                      |80.0                                         |13294.0                                   |101.0                                       |0.0                                           |2128.679999999993                       |0.0                                      |30.0                   |70.95599999999976          |1       |gold_5118|12                       |1                   |105.0             |81.0          |101.0                |2        |24                  |2.06158430208E11      |210.0        |2.602202996736E12               |NVIDIA_TITAN_RTX|2.5769803776E10          |280.0           |89.0        |4        |1.03079215104E11      |1120.0       |\n",
      "|r28n3|106894972|2022-10-11 04:18:30|1.66546E9        |3.71       |672.0           |1.0|1.94794E9               |0.0                    |1.05184E12                 |3.39      |1.75      |2.35799E10              |271230.0                    |0.0                      |16384.0                |25.0             |423.0                   |7272060.0                   |1.41724E10     |1.43197E9              |8.20511E7               |1.66134E9             |7401910.0                    |198.0                     |0.0               |419.0                    |1.71782E11               |17.0              |3.75068E7       |25.0                       |27.92                       |32.0                       |2.564558897152E12              |6.8497933E8              |8.461716278853E12                    |0.0                             |1.705330016768E12                |2.564612583424E12             |30.0                      |32.0                       |34.0                      |3.7710436352E10               |1.5011938304E10                 |52.0                            |32.0                            |35.0                            |34.0                            |32.0                              |38.25                              |52.0                              |229.0               |41.0                           |41.0                            |41.0                           |6.84435874E8                  |73393.0                            |16807.0                            |27234.0                            |21013.0                            |16807.0                              |138447.0                             |34611.75                              |73393.0                              |2.602202996736E12             |7.3244481E7                         |0.0                      |0.0                       |0.0                      |1.956313328E9                          |0.0                |2.682418135776E13                   |3.843935496E9                         |0.0                                     |0.0                 |354952.28                         |0.0                                |1665461910       |55778.0                                    |88064.0                                |0.0                                 |9.0                                       |47.0                                         |7786.0                                    |79.0                                        |0.0                                           |2115.640000000014                       |0.0                                      |30.0                   |70.5213333333338           |1       |gold_5118|12                       |1                   |105.0             |81.0          |101.0                |2        |24                  |2.06158430208E11      |210.0        |2.602202996736E12               |NVIDIA_TITAN_RTX|2.5769803776E10          |280.0           |89.0        |4        |1.03079215104E11      |1120.0       |\n",
      "|r28n3|106894973|2022-10-11 04:19:00|1.66546E9        |3.86       |672.0           |1.0|1.94796E9               |0.0                    |1.05184E12                 |3.88      |4.49      |3.73465E10              |271230.0                    |0.0                      |8380420.0              |25.0             |423.0                   |7272070.0                   |1.41726E10     |1.43213E9              |8.20511E7               |1.66134E9             |7401920.0                    |198.0                     |0.0               |419.0                    |1.50142E11               |24.0              |3.75074E7       |26.0                       |30.69                       |36.0                       |2.564558811136E12              |6.8497933E8              |8.461719749204E12                    |0.0                             |1.705351074816E12                |2.564612497408E12             |35.0                      |35.5                       |36.0                      |3.7710436352E10               |1.5011938304E10                 |51.0                            |32.0                            |35.0                            |34.0                            |32.0                              |38.0                               |51.0                              |229.0               |41.0                           |41.0                            |41.0                           |6.84435865E8                  |73411.0                            |16632.0                            |27784.0                            |20959.0                            |16632.0                              |138786.0                             |34696.5                               |73411.0                              |2.602202996736E12             |7.3245375E7                         |0.0                      |2.25                      |9.0                      |1.956334291E9                          |0.0                |2.6831838806489E13                  |3.844791768E9                         |0.0                                     |0.0                 |357469.73                         |0.0                                |1665461940       |3470351.0                                  |2.1058048E7                            |0.0                                 |894.0                                     |20963.0                                      |7.657448729E9                             |856272.0                                    |0.0                                           |2517.4499999999534                      |0.0                                      |30.0                   |83.91499999999844          |1       |gold_5118|12                       |1                   |105.0             |81.0          |101.0                |2        |24                  |2.06158430208E11      |210.0        |2.602202996736E12               |NVIDIA_TITAN_RTX|2.5769803776E10          |280.0           |89.0        |4        |1.03079215104E11      |1120.0       |\n",
      "+-----+---------+-------------------+-----------------+-----------+----------------+---+------------------------+-----------------------+---------------------------+----------+----------+------------------------+----------------------------+-------------------------+-----------------------+-----------------+------------------------+----------------------------+---------------+-----------------------+------------------------+----------------------+-----------------------------+--------------------------+------------------+-------------------------+-------------------------+------------------+----------------+---------------------------+----------------------------+---------------------------+-------------------------------+-------------------------+-------------------------------------+--------------------------------+---------------------------------+------------------------------+--------------------------+---------------------------+--------------------------+------------------------------+--------------------------------+--------------------------------+--------------------------------+--------------------------------+--------------------------------+----------------------------------+-----------------------------------+----------------------------------+--------------------+-------------------------------+--------------------------------+-------------------------------+------------------------------+-----------------------------------+-----------------------------------+-----------------------------------+-----------------------------------+-------------------------------------+-------------------------------------+--------------------------------------+-------------------------------------+------------------------------+------------------------------------+-------------------------+--------------------------+-------------------------+---------------------------------------+-------------------+------------------------------------+--------------------------------------+----------------------------------------+--------------------+----------------------------------+-----------------------------------+-----------------+-------------------------------------------+---------------------------------------+------------------------------------+------------------------------------------+---------------------------------------------+------------------------------------------+--------------------------------------------+----------------------------------------------+----------------------------------------+-----------------------------------------+-----------------------+---------------------------+--------+---------+-------------------------+--------------------+------------------+--------------+---------------------+---------+--------------------+----------------------+-------------+--------------------------------+----------------+-------------------------+----------------+------------+---------+----------------------+-------------+\n",
      "only showing top 5 rows\n",
      "\n"
     ]
    }
   ],
   "source": [
    "df_prom = spark.read.parquet(path_node_dataset)\n",
    "\n",
    "df_prom_node_data_2 = df_prom.join(df_node_hardware_info, how='left', on=\"node\")\n",
    "df_prom_node_data_2.show(5, False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We calculate memory usage ratio as `node_memory_Active_bytes / node_memory_MemTotal_bytes`\n",
    "(https://www.tigera.io/learn/guides/prometheus-monitoring/prometheus-metrics/)\n",
    "`node_memory_Active_bytes` is present as metric in the dataset, but `node_memory_MemTotal_bytes` is not. We used the `cpu_memory_bytes_total` from the node hardware info dataset as substitute for `node_memory_MemTotal_bytes`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_prom_node_data_3 = df_prom_node_data_2.withColumn('node_load_1_ratio', F.col('node_load1') / F.col('cpu_core_count_total'))\n",
    "df_prom_node_data_3 = df_prom_node_data_3.withColumn('cpu_power_sum_ratio', F.col('node_rapl_package_power_sum') / F.col('cpu_tdp_total'))\n",
    "df_prom_node_data_3 = df_prom_node_data_3.withColumn('cpu_mem_ratio', F.col('node_memory_Active_bytes') / F.col('cpu_memory_bytes_total'))\n",
    "# normalize by taking ambient temp as min value\n",
    "# cpu package temp ratios\n",
    "df_prom_node_data_3 = df_prom_node_data_3.withColumn('cpu_temp_mean_ratio', (F.col('node_thermal_zone_temp_mean') - F.col('node_ambient_temp')) / (F.col('cpu_t_case_max') - F.col('node_ambient_temp')))\n",
    "df_prom_node_data_3 = df_prom_node_data_3.withColumn('cpu_temp_min_ratio', (F.col('node_thermal_zone_temp_min') - F.col('node_ambient_temp')) / (F.col('cpu_t_case_max') - F.col('node_ambient_temp')))\n",
    "df_prom_node_data_3 = df_prom_node_data_3.withColumn('cpu_temp_max_ratio', (F.col('node_thermal_zone_temp_max') - F.col('node_ambient_temp')) / (F.col('cpu_t_case_max') - F.col('node_ambient_temp')))\n",
    "# cpu core temp ratios\n",
    "df_prom_node_data_3 = df_prom_node_data_3.withColumn('cpu_core_temp_min_ratio', (F.col('node_hwmon_temp_celsius_min') - F.col('node_ambient_temp')) / (F.col('cpu_per_core_temp_max') - F.col('node_ambient_temp')))\n",
    "df_prom_node_data_3 = df_prom_node_data_3.withColumn('cpu_core_temp_max_ratio', (F.col('node_hwmon_temp_celsius_max') - F.col('node_ambient_temp')) / (F.col('cpu_per_core_temp_max') - F.col('node_ambient_temp')))\n",
    "df_prom_node_data_3 = df_prom_node_data_3.withColumn('cpu_core_temp_mean_ratio', (F.col('node_hwmon_temp_celsius_mean') - F.col('node_ambient_temp')) / (F.col('cpu_per_core_temp_max') - F.col('node_ambient_temp')))\n",
    "# gpu ratios\n",
    "df_prom_node_data_3 = df_prom_node_data_3.withColumn('gpu_power_sum_ratio', F.col('nvidia_gpu_power_usage_milliwatts_sum') / 1000 / F.col('gpu_tdp_total'))\n",
    "df_prom_node_data_3 = df_prom_node_data_3.withColumn('gpu_mem_sum_ratio', F.col('nvidia_gpu_memory_used_bytes_sum') / F.col('gpu_memory_bytes_total'))\n",
    "df_prom_node_data_3 = df_prom_node_data_3.withColumn('gpu_temp_mean_ratio', (F.col('nvidia_gpu_temperature_celsius_mean') - F.col('node_ambient_temp')) / (F.col('gpu_temp_max') - F.col('node_ambient_temp')))\n",
    "df_prom_node_data_3 = df_prom_node_data_3.withColumn('gpu_temp_min_ratio', (F.col('nvidia_gpu_temperature_celsius_min') - F.col('node_ambient_temp')) / (F.col('gpu_temp_max') - F.col('node_ambient_temp')))\n",
    "df_prom_node_data_3 = df_prom_node_data_3.withColumn('gpu_temp_max_ratio', (F.col('nvidia_gpu_temperature_celsius_max') - F.col('node_ambient_temp')) / (F.col('gpu_temp_max') - F.col('node_ambient_temp')))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# clip all ratios to min 0 (due to numerical inaccuracies in the collected data)\n",
    "for c in df_prom_node_data_3.columns:\n",
    "    if 'ratio' in c:\n",
    "        df_prom_node_data_3 = df_prom_node_data_3.withColumn(c, when(F.col(c) < 0, 0).otherwise(F.col(c)))\n",
    "\n",
    "# change all cpu-only node gpu column values to 0.0\n",
    "for c in df_prom_node_data_3.columns:\n",
    "    if 'gpu' in c and 'ratio' in c:\n",
    "        df_prom_node_data_3 = df_prom_node_data_3.withColumn(c, when(F.col('gpu_node') == 0, 0.0).otherwise(F.col(c)))\n",
    "\n",
    "# clip node_load_1_ratio to max 1.0, as a cpu with at least as many threads as cores is already at 100% load\n",
    "df_prom_node_data_3 = df_prom_node_data_3.withColumn('node_load_1_ratio', when(F.col('node_load_1_ratio') > 1, 1).otherwise(F.col('node_load_1_ratio')))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    }
   ],
   "source": [
    "# 10 min runtime\n",
    "# all newly generated columns\n",
    "cols = ['node_load_1_ratio', 'cpu_power_sum_ratio', 'cpu_mem_ratio', 'cpu_temp_max_ratio', 'cpu_core_temp_max_ratio', 'gpu_power_sum_ratio', 'gpu_mem_sum_ratio', 'gpu_temp_max_ratio']\n",
    "cols_dict = {\n",
    "    'node_load_1_ratio': 'Node Load 1 [\\%]',\n",
    "    'cpu_power_sum_ratio': 'CPU Power [\\%]',\n",
    "    'cpu_mem_ratio': 'CPU Memory [\\%]',\n",
    "    'cpu_temp_max_ratio': 'CPU Package Temp. [\\%]',\n",
    "    'cpu_core_temp_max_ratio': 'CPU Core Temp. [\\%]',\n",
    "    'gpu_power_sum_ratio': 'GPU Power [\\%]',\n",
    "    'gpu_mem_sum_ratio': 'GPU Memory [\\%]',\n",
    "    'gpu_temp_max_ratio': 'GPU Temp. [\\%]'\n",
    "}\n",
    "\n",
    "df_double_regular = df_prom_node_data_3.filter(F.col('gpu_node') == 0).select(cols).toPandas()\n",
    "df_double_ml = df_prom_node_data_3.filter(F.col('gpu_node') == 1).select(cols).toPandas()\n",
    "\n",
    "df_double_regular = df_double_regular * 100\n",
    "df_double_ml = df_double_ml * 100\n",
    "\n",
    "# clip to 99.99th percentile to remove outliers\n",
    "for c in cols:\n",
    "    df_double_regular[c] = df_double_regular[c].clip(upper=df_double_regular[c].quantile(0.9999))\n",
    "    df_double_ml[c] = df_double_ml[c].clip(upper=df_double_ml[c].quantile(0.9999))\n",
    "\n",
    "# add gpu_node col back\n",
    "df_double_regular_2 = df_double_regular.copy()\n",
    "df_double_regular_2['gpu_node'] = 0\n",
    "df_double_ml_2 = df_double_ml.copy()\n",
    "df_double_ml_2['gpu_node'] = 1\n",
    "df_double = pd.concat([df_double_regular_2, df_double_ml_2], axis=0).round(0) # round for easier plotting, so multiple outliers stack on top of each other"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 900x1800 with 16 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 10-30 min runtime (varies greatly with sns version, tested with 0.12.2 (10 min) and 0.13.2 (30 min))\n",
    "set_one_column_figure_font( height_multiplier= 3)\n",
    "\n",
    "ncols = 2\n",
    "nrows = int(len(cols))\n",
    "fig, axes = plt.subplots(nrows=nrows, ncols=ncols, gridspec_kw={'width_ratios': [4, 1]})\n",
    "fig.subplots_adjust(hspace=0.7)\n",
    "fig.subplots_adjust(wspace=0.0)\n",
    "\n",
    "for i, c in enumerate(cols):\n",
    "    ax = axes[i, 0]\n",
    "    df = df_double[[c, 'gpu_node']]\n",
    "    if 'gpu' not in c:\n",
    "        sns.histplot(df_double_regular[c].round(0), kde=False, label=f'Non-GPU', color=non_gpu_color, ax=ax, line_kws={'linewidth': 2}, bins=range(0, 111, 5), stat=\"density\")\n",
    "    sns.histplot(df_double_ml[c].round(0), kde=False, label=f'GPU', color=gpu_color, ax=ax, line_kws={'linewidth': 2}, bins=range(0, 111, 5), stat=\"density\")\n",
    "    \n",
    "    ax.set_ylabel('Density')\n",
    "    ax.set_xlabel(cols_dict[c])\n",
    "    ax.set_xticks(np.arange(0, 101, 10))\n",
    "    ax.set_xlim(0, 110)\n",
    "    ax.yaxis.set_major_formatter(FormatStrFormatter('%.2f'))\n",
    "    ax.legend().remove()\n",
    "\n",
    "    ax2 = axes[i, 1]\n",
    "    # put ax2 yaxis on the right side of the subplot\n",
    "    ax2.yaxis.tick_right()\n",
    "    ax2.yaxis.set_label_position(\"right\")\n",
    "\n",
    "    bp = sns.boxplot(data=df, x='gpu_node', y=c, hue='gpu_node', ax=ax2, palette=[non_gpu_color, gpu_color], order=[0, 1], showfliers=True,  hue_order=[0, 1], dodge=False)\n",
    "\n",
    "\n",
    "    # remove all duplicated outlier dots to reduce pdf size\n",
    "    # millions of overlapping outliers are not visible anyway\n",
    "    for l in bp.lines:\n",
    "        if len(l.get_ydata()) > 2:\n",
    "            new_ydata = np.unique(l.get_ydata())\n",
    "            l.set_ydata(new_ydata)\n",
    "            l.set_xdata(l.get_xdata()[:len(new_ydata)])\n",
    "\n",
    "    ax2.set_ylabel('[\\%]')\n",
    "    ax2.set_xlabel('')\n",
    "    ax2.xaxis.set_ticks([0, 1])\n",
    "    if i == (len(cols) - 1):\n",
    "        ax2.set_xticklabels(['CPU-only', 'GPU'])\n",
    "        ax2.set_xticklabels(ax2.get_xticklabels(), rotation=45, ha='right')\n",
    "    else:\n",
    "        ax2.set_xticklabels([])\n",
    "    ax2.set_yticks(np.arange(0, 101, 25))\n",
    "    ax2.set_ylim(0, 115)\n",
    "    ax2.legend().remove()\n",
    "\n",
    "plt.legend(handles=[mpatches.Patch(color=non_gpu_color, label='CPU-only Nodes'), mpatches.Patch(color=gpu_color, label='GPU Nodes')], loc='upper center', bbox_to_anchor=(-1.5, 13.6), ncol=2)\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "fig.savefig('plots/section_4/utilization_regular_vs_ml_distributions_density_box.png', dpi=900, bbox_inches='tight')\n",
    "fig.savefig('plots/section_4/utilization_regular_vs_ml_distributions_density_box.pdf', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>node_load_1_ratio</th>\n",
       "      <th>cpu_power_sum_ratio</th>\n",
       "      <th>cpu_mem_ratio</th>\n",
       "      <th>cpu_temp_max_ratio</th>\n",
       "      <th>cpu_core_temp_max_ratio</th>\n",
       "      <th>gpu_power_sum_ratio</th>\n",
       "      <th>gpu_mem_sum_ratio</th>\n",
       "      <th>gpu_temp_max_ratio</th>\n",
       "      <th>gpu_node</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.278277e+08</td>\n",
       "      <td>1.278277e+08</td>\n",
       "      <td>1.278277e+08</td>\n",
       "      <td>1.278277e+08</td>\n",
       "      <td>1.278277e+08</td>\n",
       "      <td>1.278277e+08</td>\n",
       "      <td>1.278277e+08</td>\n",
       "      <td>1.278277e+08</td>\n",
       "      <td>1.278277e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4.493632e+01</td>\n",
       "      <td>4.595084e+01</td>\n",
       "      <td>1.008437e+01</td>\n",
       "      <td>3.002018e+01</td>\n",
       "      <td>2.305366e+01</td>\n",
       "      <td>2.629734e+00</td>\n",
       "      <td>1.832744e+00</td>\n",
       "      <td>4.146844e+00</td>\n",
       "      <td>9.822657e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4.443622e+01</td>\n",
       "      <td>3.031956e+01</td>\n",
       "      <td>1.436649e+01</td>\n",
       "      <td>2.111640e+01</td>\n",
       "      <td>1.600227e+01</td>\n",
       "      <td>1.135784e+01</td>\n",
       "      <td>9.259290e+00</td>\n",
       "      <td>1.646279e+01</td>\n",
       "      <td>2.976207e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1.200000e+01</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>6.000000e+00</td>\n",
       "      <td>4.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2.100000e+01</td>\n",
       "      <td>4.400000e+01</td>\n",
       "      <td>4.000000e+00</td>\n",
       "      <td>3.300000e+01</td>\n",
       "      <td>2.500000e+01</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.000000e+02</td>\n",
       "      <td>6.900000e+01</td>\n",
       "      <td>1.300000e+01</td>\n",
       "      <td>4.800000e+01</td>\n",
       "      <td>3.700000e+01</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000e+02</td>\n",
       "      <td>1.000000e+02</td>\n",
       "      <td>9.200000e+01</td>\n",
       "      <td>9.700000e+01</td>\n",
       "      <td>7.600000e+01</td>\n",
       "      <td>1.050000e+02</td>\n",
       "      <td>9.600000e+01</td>\n",
       "      <td>1.020000e+02</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       node_load_1_ratio  cpu_power_sum_ratio  cpu_mem_ratio  \\\n",
       "count       1.278277e+08         1.278277e+08   1.278277e+08   \n",
       "mean        4.493632e+01         4.595084e+01   1.008437e+01   \n",
       "std         4.443622e+01         3.031956e+01   1.436649e+01   \n",
       "min         0.000000e+00         0.000000e+00   0.000000e+00   \n",
       "25%         0.000000e+00         1.200000e+01   1.000000e+00   \n",
       "50%         2.100000e+01         4.400000e+01   4.000000e+00   \n",
       "75%         1.000000e+02         6.900000e+01   1.300000e+01   \n",
       "max         1.000000e+02         1.000000e+02   9.200000e+01   \n",
       "\n",
       "       cpu_temp_max_ratio  cpu_core_temp_max_ratio  gpu_power_sum_ratio  \\\n",
       "count        1.278277e+08             1.278277e+08         1.278277e+08   \n",
       "mean         3.002018e+01             2.305366e+01         2.629734e+00   \n",
       "std          2.111640e+01             1.600227e+01         1.135784e+01   \n",
       "min          0.000000e+00             0.000000e+00         0.000000e+00   \n",
       "25%          6.000000e+00             4.000000e+00         0.000000e+00   \n",
       "50%          3.300000e+01             2.500000e+01         0.000000e+00   \n",
       "75%          4.800000e+01             3.700000e+01         0.000000e+00   \n",
       "max          9.700000e+01             7.600000e+01         1.050000e+02   \n",
       "\n",
       "       gpu_mem_sum_ratio  gpu_temp_max_ratio      gpu_node  \n",
       "count       1.278277e+08        1.278277e+08  1.278277e+08  \n",
       "mean        1.832744e+00        4.146844e+00  9.822657e-02  \n",
       "std         9.259290e+00        1.646279e+01  2.976207e-01  \n",
       "min         0.000000e+00        0.000000e+00  0.000000e+00  \n",
       "25%         0.000000e+00        0.000000e+00  0.000000e+00  \n",
       "50%         0.000000e+00        0.000000e+00  0.000000e+00  \n",
       "75%         0.000000e+00        0.000000e+00  0.000000e+00  \n",
       "max         9.600000e+01        1.020000e+02  1.000000e+00  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_double.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>node_load_1_ratio</th>\n",
       "      <th>cpu_power_sum_ratio</th>\n",
       "      <th>cpu_mem_ratio</th>\n",
       "      <th>cpu_temp_max_ratio</th>\n",
       "      <th>cpu_core_temp_max_ratio</th>\n",
       "      <th>gpu_power_sum_ratio</th>\n",
       "      <th>gpu_mem_sum_ratio</th>\n",
       "      <th>gpu_temp_max_ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.152716e+08</td>\n",
       "      <td>1.152716e+08</td>\n",
       "      <td>1.152716e+08</td>\n",
       "      <td>1.152716e+08</td>\n",
       "      <td>1.152716e+08</td>\n",
       "      <td>115271641.0</td>\n",
       "      <td>115271641.0</td>\n",
       "      <td>115271641.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4.756557e+01</td>\n",
       "      <td>4.783780e+01</td>\n",
       "      <td>1.031743e+01</td>\n",
       "      <td>3.214011e+01</td>\n",
       "      <td>2.474143e+01</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4.499856e+01</td>\n",
       "      <td>3.089782e+01</td>\n",
       "      <td>1.460478e+01</td>\n",
       "      <td>2.098012e+01</td>\n",
       "      <td>1.585960e+01</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>5.298489e-03</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.000000e-01</td>\n",
       "      <td>1.143680e+01</td>\n",
       "      <td>1.516998e+00</td>\n",
       "      <td>6.349206e+00</td>\n",
       "      <td>5.194805e+00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>3.137500e+01</td>\n",
       "      <td>4.911255e+01</td>\n",
       "      <td>3.597796e+00</td>\n",
       "      <td>3.548387e+01</td>\n",
       "      <td>2.727273e+01</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.000000e+02</td>\n",
       "      <td>7.165413e+01</td>\n",
       "      <td>1.310875e+01</td>\n",
       "      <td>4.920635e+01</td>\n",
       "      <td>3.815789e+01</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000e+02</td>\n",
       "      <td>9.981036e+01</td>\n",
       "      <td>9.174352e+01</td>\n",
       "      <td>9.666667e+01</td>\n",
       "      <td>7.631579e+01</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       node_load_1_ratio  cpu_power_sum_ratio  cpu_mem_ratio  \\\n",
       "count       1.152716e+08         1.152716e+08   1.152716e+08   \n",
       "mean        4.756557e+01         4.783780e+01   1.031743e+01   \n",
       "std         4.499856e+01         3.089782e+01   1.460478e+01   \n",
       "min         0.000000e+00         0.000000e+00   5.298489e-03   \n",
       "25%         5.000000e-01         1.143680e+01   1.516998e+00   \n",
       "50%         3.137500e+01         4.911255e+01   3.597796e+00   \n",
       "75%         1.000000e+02         7.165413e+01   1.310875e+01   \n",
       "max         1.000000e+02         9.981036e+01   9.174352e+01   \n",
       "\n",
       "       cpu_temp_max_ratio  cpu_core_temp_max_ratio  gpu_power_sum_ratio  \\\n",
       "count        1.152716e+08             1.152716e+08          115271641.0   \n",
       "mean         3.214011e+01             2.474143e+01                  0.0   \n",
       "std          2.098012e+01             1.585960e+01                  0.0   \n",
       "min          0.000000e+00             0.000000e+00                  0.0   \n",
       "25%          6.349206e+00             5.194805e+00                  0.0   \n",
       "50%          3.548387e+01             2.727273e+01                  0.0   \n",
       "75%          4.920635e+01             3.815789e+01                  0.0   \n",
       "max          9.666667e+01             7.631579e+01                  0.0   \n",
       "\n",
       "       gpu_mem_sum_ratio  gpu_temp_max_ratio  \n",
       "count        115271641.0         115271641.0  \n",
       "mean                 0.0                 0.0  \n",
       "std                  0.0                 0.0  \n",
       "min                  0.0                 0.0  \n",
       "25%                  0.0                 0.0  \n",
       "50%                  0.0                 0.0  \n",
       "75%                  0.0                 0.0  \n",
       "max                  0.0                 0.0  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_double_regular.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>node_load_1_ratio</th>\n",
       "      <th>cpu_power_sum_ratio</th>\n",
       "      <th>cpu_mem_ratio</th>\n",
       "      <th>cpu_temp_max_ratio</th>\n",
       "      <th>cpu_core_temp_max_ratio</th>\n",
       "      <th>gpu_power_sum_ratio</th>\n",
       "      <th>gpu_mem_sum_ratio</th>\n",
       "      <th>gpu_temp_max_ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.255608e+07</td>\n",
       "      <td>1.255608e+07</td>\n",
       "      <td>1.255608e+07</td>\n",
       "      <td>1.255608e+07</td>\n",
       "      <td>1.255608e+07</td>\n",
       "      <td>1.255608e+07</td>\n",
       "      <td>1.255608e+07</td>\n",
       "      <td>1.255608e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2.110471e+01</td>\n",
       "      <td>2.857754e+01</td>\n",
       "      <td>8.061655e+00</td>\n",
       "      <td>1.021246e+01</td>\n",
       "      <td>7.358645e+00</td>\n",
       "      <td>2.681078e+01</td>\n",
       "      <td>1.866584e+01</td>\n",
       "      <td>4.217956e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.918326e+01</td>\n",
       "      <td>1.586820e+01</td>\n",
       "      <td>1.160606e+01</td>\n",
       "      <td>8.578851e+00</td>\n",
       "      <td>6.093781e+00</td>\n",
       "      <td>2.578864e+01</td>\n",
       "      <td>2.363560e+01</td>\n",
       "      <td>3.401318e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>7.844452e-03</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1.250000e-01</td>\n",
       "      <td>1.177314e+01</td>\n",
       "      <td>9.004034e-01</td>\n",
       "      <td>1.818182e+00</td>\n",
       "      <td>1.315789e+00</td>\n",
       "      <td>8.331696e+00</td>\n",
       "      <td>1.720082e-02</td>\n",
       "      <td>1.562500e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>9.583333e+00</td>\n",
       "      <td>3.261302e+01</td>\n",
       "      <td>3.485287e+00</td>\n",
       "      <td>9.433962e+00</td>\n",
       "      <td>6.578947e+00</td>\n",
       "      <td>1.624875e+01</td>\n",
       "      <td>7.391866e+00</td>\n",
       "      <td>3.030303e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2.675000e+01</td>\n",
       "      <td>4.145952e+01</td>\n",
       "      <td>1.054370e+01</td>\n",
       "      <td>1.785714e+01</td>\n",
       "      <td>1.184211e+01</td>\n",
       "      <td>3.887250e+01</td>\n",
       "      <td>3.142064e+01</td>\n",
       "      <td>7.187500e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000e+02</td>\n",
       "      <td>8.468707e+01</td>\n",
       "      <td>9.008024e+01</td>\n",
       "      <td>3.928571e+01</td>\n",
       "      <td>2.894737e+01</td>\n",
       "      <td>1.054422e+02</td>\n",
       "      <td>9.628448e+01</td>\n",
       "      <td>1.015873e+02</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       node_load_1_ratio  cpu_power_sum_ratio  cpu_mem_ratio  \\\n",
       "count       1.255608e+07         1.255608e+07   1.255608e+07   \n",
       "mean        2.110471e+01         2.857754e+01   8.061655e+00   \n",
       "std         2.918326e+01         1.586820e+01   1.160606e+01   \n",
       "min         0.000000e+00         0.000000e+00   7.844452e-03   \n",
       "25%         1.250000e-01         1.177314e+01   9.004034e-01   \n",
       "50%         9.583333e+00         3.261302e+01   3.485287e+00   \n",
       "75%         2.675000e+01         4.145952e+01   1.054370e+01   \n",
       "max         1.000000e+02         8.468707e+01   9.008024e+01   \n",
       "\n",
       "       cpu_temp_max_ratio  cpu_core_temp_max_ratio  gpu_power_sum_ratio  \\\n",
       "count        1.255608e+07             1.255608e+07         1.255608e+07   \n",
       "mean         1.021246e+01             7.358645e+00         2.681078e+01   \n",
       "std          8.578851e+00             6.093781e+00         2.578864e+01   \n",
       "min          0.000000e+00             0.000000e+00         0.000000e+00   \n",
       "25%          1.818182e+00             1.315789e+00         8.331696e+00   \n",
       "50%          9.433962e+00             6.578947e+00         1.624875e+01   \n",
       "75%          1.785714e+01             1.184211e+01         3.887250e+01   \n",
       "max          3.928571e+01             2.894737e+01         1.054422e+02   \n",
       "\n",
       "       gpu_mem_sum_ratio  gpu_temp_max_ratio  \n",
       "count       1.255608e+07        1.255608e+07  \n",
       "mean        1.866584e+01        4.217956e+01  \n",
       "std         2.363560e+01        3.401318e+01  \n",
       "min         0.000000e+00        0.000000e+00  \n",
       "25%         1.720082e-02        1.562500e+01  \n",
       "50%         7.391866e+00        3.030303e+01  \n",
       "75%         3.142064e+01        7.187500e+01  \n",
       "max         9.628448e+01        1.015873e+02  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_double_ml.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "np.float64(17.396570808177522)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# how much in percent of time is gpu temp above 90%\n",
    "df_double_ml[df_double_ml['gpu_temp_max_ratio'] > 90]['gpu_temp_max_ratio'].count() / df_double_ml['gpu_temp_max_ratio'].count() * 100"
   ]
  }
 ],
 "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": "3.12.5"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "3a8149eb17ef945abfdf8e383acce9168fd9c2ac41a1124cb85eb67776d7d0c8"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}