notebooks/00000000--empty-template.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import boto3\n",
    "import datetime\n",
    "import json\n",
    "from dateutil.tz import tzlocal\n",
    "\n",
    "import get_test_details\n",
    "import query_experiment_details\n",
    "import aggregate_statistics\n",
    "import plot\n",
    "\n",
    "stepfunctions = boto3.client('stepfunctions')\n",
    "cloudwatch_logs = boto3.client('logs')\n",
    "cloudformation = boto3.client('cloudformation')\n",
    "cloudwatch = boto3.client('cloudwatch')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Sample test specification. Submit is as the input of the StepFunction workflow that ends with `-main` to start a test.\n",
    "```\n",
    "{\n",
    "  \"test_specification\": {\n",
    "    \"parameters\": {\n",
    "      \"cluster_throughput_mb_per_sec\": [ 8, 16, 24, 32, 40, 44, 48, 52, 56 ],\n",
    "      \"num_producers\": [ 6 ],\n",
    "      \"consumer_groups\": [ { \"num_groups\": 2, \"size\": 6 }\n",
    "      ],\n",
    "      \"client_props\": [\n",
    "        {\n",
    "          \"producer\": \"acks=all linger.ms=5 batch.size=65536 buffer.memory=2147483648 security.protocol=PLAINTEXT\",\n",
    "          \"consumer\": \"security.protocol=PLAINTEXT\"\n",
    "        }\n",
    "      ],\n",
    "      \"num_partitions\": [ 36 ],\n",
    "      \"record_size_byte\": [ 1024 ],\n",
    "      \"replication_factor\": [ 3 ],\n",
    "      \"duration_sec\": [ 3600 ]\n",
    "    },\n",
    "    \"skip_remaining_throughput\": {\n",
    "      \"less-than\": [\n",
    "        \"sent_div_requested_mb_per_sec\",\n",
    "        0.995\n",
    "      ]\n",
    "    },\n",
    "    \"depletion_configuration\": {\n",
    "      \"upper_threshold\": {\n",
    "        \"mb_per_sec\": 200\n",
    "      },\n",
    "      \"approximate_timeout_hours\": 0.5\n",
    "    }\n",
    "  }\n",
    "}\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "test_params = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "test_params.extend([\n",
    "    {'execution_arn': '' }\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_details = get_test_details.get_test_details(test_params, stepfunctions, cloudformation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "(producer_stats, consumer_stats) = query_experiment_details.query_cw_logs(test_details, cloudwatch_logs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "partitions = {\n",
    "    'ignore_keys': [ 'topic_id', 'cluster_name', 'test_id', 'client_props.consumer', 'cluster_id', 'duration_sec', 'throughput_series_id', 'brokers_type_numeric', ],\n",
    "    'title_keys': [ 'kafka_version', 'broker_storage', 'provisioned_throughput', 'in_cluster_encryption', 'producer.security.protocol', ],\n",
    "    'row_keys': [ 'num_producers', 'consumer_groups.num_groups',  ],\n",
    "    'column_keys': [ 'producer.acks', 'producer.batch.size', 'num_partitions', ],\n",
    "    'metric_color_keys': [ 'brokers_type_numeric', 'brokers', ],\n",
    "}\n",
    "\n",
    "filter_fn = lambda x: True\n",
    "filter_agg_fn = lambda x: True\n",
    "\n",
    "filtered_producer_stats = list(filter(filter_fn, producer_stats))\n",
    "filtered_consumer_stats = filter(filter_fn, consumer_stats)\n",
    "\n",
    "(producer_aggregated_stats, consumer_aggregated_stats, combined_stats) = aggregate_statistics.aggregate_cw_logs(filtered_producer_stats, filtered_consumer_stats, partitions)\n",
    "filtered_producer_aggregated_stats = list(filter(filter_agg_fn, producer_aggregated_stats))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot.plot_measurements(filtered_producer_aggregated_stats, ['latency_ms_p50_mean', 'latency_ms_p99_mean', ], 'producer put latency (ms)', **partitions, )\n",
    "plot.plot_measurements(filtered_producer_aggregated_stats, ['latency_ms_p50_stdev', 'latency_ms_p99_stdev', ], 'producer put latency stdev (ms)', **partitions, )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot.plot_measurements(producer_aggregated_stats, ['sent_div_requested_mb_per_sec'], 'sent / requested throughput (mb/sec)', **partitions, ymin=0.990, ymax=1.01)\n",
    "plot.plot_measurements(producer_aggregated_stats, ['actual_duration_div_requested_duration_sec_max'], 'actual test duration/requested test duration (ratio)', **partitions)\n",
    "plot.plot_measurements(producer_aggregated_stats, ['num_tests'], 'number of tests (count)', **partitions)"
   ]
  },
  {
   "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": "3.10.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}