spark-excercises.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- https://towardsdatascience.com/six-spark-exercises-to-rule-them-all-242445b24565\n",
    "- https://intellipaat.com/community/6443/how-to-optimize-shuffle-spill-in-apache-spark-application#:~:text=Spark%201.4%20has%20some%20better,available%20for%20the%20shuffle%20buffer.&text=Increase%20the%20memory%20in%20your,executor."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "export SPARK_HOME=/opt/spark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-08T08:16:13.940229Z",
     "start_time": "2020-10-08T08:16:13.935557Z"
    }
   },
   "outputs": [],
   "source": [
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-08T08:16:29.003944Z",
     "start_time": "2020-10-08T08:16:29.000792Z"
    }
   },
   "outputs": [],
   "source": [
    "os.environ['SPARK_HOME'] = '/opt/spark'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-08T08:16:35.109433Z",
     "start_time": "2020-10-08T08:16:35.105615Z"
    }
   },
   "outputs": [],
   "source": [
    "import findspark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-08T08:16:39.974514Z",
     "start_time": "2020-10-08T08:16:39.969567Z"
    }
   },
   "outputs": [],
   "source": [
    "findspark.init()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-08T08:16:41.535677Z",
     "start_time": "2020-10-08T08:16:41.274574Z"
    }
   },
   "outputs": [],
   "source": [
    "import pyspark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-08T08:16:42.051510Z",
     "start_time": "2020-10-08T08:16:42.036682Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'/opt/spark/python/pyspark/__init__.py'"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pyspark.__file__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-03T05:13:29.704606Z",
     "start_time": "2020-09-03T05:13:29.684964Z"
    }
   },
   "outputs": [],
   "source": [
    "from pyspark.sql import SparkSession\n",
    "from pyspark.sql import functions as F\n",
    "from pyspark.sql import types as T\n",
    "\n",
    "c = F.col"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-03T05:13:33.834430Z",
     "start_time": "2020-09-03T05:13:30.506464Z"
    }
   },
   "outputs": [],
   "source": [
    "spark = (\n",
    "    SparkSession\n",
    "    .builder\n",
    "    .master('local[*]')\n",
    "    .config(\n",
    "        'spark.sql.shuffle.partitions',\n",
    "        '400'\n",
    "    )\n",
    "    .config(\"spark.sql.autoBroadcastJoinThreshold\", -1)\n",
    "    .config('spark.driver.memory', '5g')\n",
    "    .appName('hello-spark')\n",
    "    .getOrCreate()\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-03T05:13:33.840581Z",
     "start_time": "2020-09-03T05:13:33.835763Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "            <div>\n",
       "                <p><b>SparkSession - in-memory</b></p>\n",
       "                \n",
       "        <div>\n",
       "            <p><b>SparkContext</b></p>\n",
       "\n",
       "            <p><a href=\"http://localhost:4041\">Spark UI</a></p>\n",
       "\n",
       "            <dl>\n",
       "              <dt>Version</dt>\n",
       "                <dd><code>v3.0.0</code></dd>\n",
       "              <dt>Master</dt>\n",
       "                <dd><code>local[*]</code></dd>\n",
       "              <dt>AppName</dt>\n",
       "                <dd><code>hello-spark</code></dd>\n",
       "            </dl>\n",
       "        </div>\n",
       "        \n",
       "            </div>\n",
       "        "
      ],
      "text/plain": [
       "<pyspark.sql.session.SparkSession at 0x7f04c9971a58>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "spark"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Warmup 1\n",
    "\n",
    "\n",
    "    Find out how many orders, how many products and how many sellers are in the data.\n",
    "    How many products have been sold at least once? Which is the product contained in more orders?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-03T05:13:35.913632Z",
     "start_time": "2020-09-03T05:13:33.842205Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "root\n",
      " |-- product_id: string (nullable = true)\n",
      " |-- product_name: string (nullable = true)\n",
      " |-- price: string (nullable = true)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "sdf_products = spark.read.parquet('/data/dataset/products_parquet')\n",
    "sdf_products.printSchema()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-03T05:13:35.988587Z",
     "start_time": "2020-09-03T05:13:35.914806Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "root\n",
      " |-- seller_id: string (nullable = true)\n",
      " |-- seller_name: string (nullable = true)\n",
      " |-- daily_target: string (nullable = true)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "sdf_seller = spark.read.parquet('/data/dataset/sellers_parquet')\n",
    "sdf_seller.printSchema()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-03T05:13:36.095848Z",
     "start_time": "2020-09-03T05:13:35.989874Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "root\n",
      " |-- order_id: string (nullable = true)\n",
      " |-- product_id: string (nullable = true)\n",
      " |-- seller_id: string (nullable = true)\n",
      " |-- date: string (nullable = true)\n",
      " |-- num_pieces_sold: string (nullable = true)\n",
      " |-- bill_raw_text: string (nullable = true)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "sdf_order = spark.read.parquet(\"/data/dataset/sales_parquet\")\n",
    "sdf_order.printSchema()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Warm Up 2\n",
    "\n",
    "    How many distinct products have been sold in each day?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Excercise 1\n",
    "\n",
    "    What is the average revenue of the orders?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### salting"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Excercise 2\n",
    "\n",
    "    For each seller, what is the average % contribution of an order to the seller's daily quota?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exercise 3\n",
    "\n",
    "    Who are the second most selling and the least selling persons (sellers) for each product?\n",
    "    Who are those for product with `product_id = 0`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- https://databricks.com/blog/2015/07/15/introducing-window-functions-in-spark-sql.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-03T04:30:16.287703Z",
     "start_time": "2020-09-03T04:30:16.283325Z"
    }
   },
   "outputs": [],
   "source": [
    "from pyspark.sql.window import Window"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-03T04:30:18.132027Z",
     "start_time": "2020-09-03T04:30:18.115548Z"
    }
   },
   "source": [
    "```python\n",
    "spec = Window.partitionBy(*cols).orderBy(*cols)\n",
    "windFunc.over(spec)\n",
    "...\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Excercise 4\n",
    "\n",
    "    Create a new column called \"hashed_bill\" defined as follows:\n",
    "\n",
    "        - if the order_id is even: apply MD5 hashing iteratively to the bill_raw_text field,\n",
    "          once for each 'A' (capital 'A') present in the text. E.g. if the bill text is 'nbAAnllA',\n",
    "          you would apply hashing three times iteratively (only if the order number is even)\n",
    "        - if the order_id is odd: apply SHA256 hashing to the bill text\n",
    "\n",
    "    Finally, check if there are any duplicate on the new column"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "spark-excercises",
   "language": "python",
   "name": "spark-excercises"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {},
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}