COMPSs_Manuals/source/Sections/09_PyCOMPSs_Notebooks/syntax/8_Integration_with_Numba.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# PyCOMPSs: Integration with Numba\n",
    "\n",
    "In this example we will how to use Numba with PyCOMPSs."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "## Import the PyCOMPSs library"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pycompss.interactive as ipycompss"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Starting runtime  \n",
    "Initialize COMPSs runtime\n",
    "Parameters indicates if the execution will generate task graph, tracefile, monitor interval and debug information."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "********************************************************\n",
      "**************** PyCOMPSs Interactive ******************\n",
      "********************************************************\n",
      "*          .-~~-.--.           ______         ___      *\n",
      "*         :         )         |____  \\       /   |     *\n",
      "*   .~ ~ -.\\       /.- ~~ .      __) |      /_/| |     *\n",
      "*   >       `.   .'       <     |__  |         | |     *\n",
      "*  (         .- -.         )   ____) |   _     | |     *\n",
      "*   `- -.-~  `- -'  ~-.- -'   |______/  |_|    |_|     *\n",
      "*     (        :        )           _ _ .-:            *\n",
      "*      ~--.    :    .--~        .-~  .-~  }            *\n",
      "*          ~-.-^-.-~ \\_      .~  .-~   .~              *\n",
      "*                   \\ \\ '     \\ '_ _ -~                *\n",
      "*                    \\`.\\`.    //                      *\n",
      "*           . - ~ ~-.__\\`.\\`-.//                       *\n",
      "*       .-~   . - ~  }~ ~ ~-.~-.                       *\n",
      "*     .' .-~      .-~       :/~-.~-./:                 *\n",
      "*    /_~_ _ . - ~                 ~-.~-._              *\n",
      "*                                     ~-.<             *\n",
      "********************************************************\n",
      "* - Starting COMPSs runtime...                         *\n",
      "* - Log path : /home/user/.COMPSs/Interactive_12/\n",
      "* - PyCOMPSs Runtime started... Have fun!              *\n",
      "********************************************************\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "if 'BINDER_SERVICE_HOST' in os.environ:\n",
    "    ipycompss.start(graph=True, debug=False,\n",
    "                    project_xml='../xml/project.xml',\n",
    "                    resources_xml='../xml/resources.xml')\n",
    "else:\n",
    "    ipycompss.start(graph=True, monitor=1000, trace=True, debug=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "## Importing task and arguments directionality modules \n",
    "Import task module before annotating functions or methods "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pycompss.api.task import task\n",
    "from pycompss.api.parameter import *\n",
    "from pycompss.api.api import compss_barrier\n",
    "from pycompss.api.api import compss_wait_on"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "## Importing other modules\n",
    "Import the time and numpy modules"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import time\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "## Declaring tasks \n",
    "Declare functions and decorate with @task those that should be tasks  -- Note that they are exactly the same but the \"numba\" parameter in the @task decorator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "@task(returns=1, numba=False)  # Default: numba=False\n",
    "def ident_loops(x):\n",
    "    r = np.empty_like(x)\n",
    "    n = len(x)\n",
    "    for i in range(n):\n",
    "        r[i] = np.cos(x[i]) ** 2 + np.sin(x[i]) ** 2\n",
    "    return r"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "@task(returns=1, numba=True)\n",
    "def ident_loops_jit(x):\n",
    "    r = np.empty_like(x)\n",
    "    n = len(x)\n",
    "    for i in range(n):\n",
    "        r[i] = np.cos(x[i]) ** 2 + np.sin(x[i]) ** 2\n",
    "    return r"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "## Invoking tasks "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found task: ident_loops\n",
      "Found task: ident_loops_jit\n",
      "TIMING RESULTS:\n",
      "* ident_loops      : 10.02781057357788 seconds\n",
      "* ident_loops_jit  : 1.8261973857879639 seconds\n",
      "* SUCCESS: Results match.\n"
     ]
    }
   ],
   "source": [
    "size = 1000000\n",
    "ntasks = 8\n",
    "\n",
    "# Run some tasks without numba jit\n",
    "start = time.time()\n",
    "for i in range(ntasks):\n",
    "    out = ident_loops(np.arange(size))\n",
    "compss_barrier()\n",
    "end = time.time()\n",
    "\n",
    "# Run some tasks with numba jit\n",
    "start_jit = time.time()\n",
    "for i in range(ntasks):\n",
    "    out_jit = ident_loops_jit(np.arange(size))\n",
    "compss_barrier()\n",
    "end_jit = time.time()\n",
    "\n",
    "# Get the last result of each run to compare that the results are ok\n",
    "out = compss_wait_on(out)\n",
    "out_jit = compss_wait_on(out_jit)\n",
    "\n",
    "print(\"TIMING RESULTS:\")\n",
    "print(\"* ident_loops      : %s seconds\" % str(end - start))\n",
    "print(\"* ident_loops_jit  : %s seconds\" % str(end_jit - start_jit))\n",
    "if len(out) == len(out_jit) and list(out) == list(out_jit):\n",
    "    print(\"* SUCCESS: Results match.\")\n",
    "else:\n",
    "    print(\"* FAILURE: Results are different!!!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "## Stop the runtime"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "********************************************************\n",
      "*************** STOPPING PyCOMPSs ******************\n",
      "********************************************************\n",
      "Checking if any issue happened.\n",
      "Warning: some of the variables used with PyCOMPSs may\n",
      "         have not been brought to the master.\n",
      "********************************************************\n"
     ]
    }
   ],
   "source": [
    "ipycompss.stop(sync=False)"
   ]
  }
 ],
 "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"
  },
  "livereveal": {
   "auto_select": "code",
   "overlay": "<div class='myheader'><img height='100' align='right' src='http://www.bsc.es/actris/sites/default/files/logos/BSC-logo%201.jpg'></div>",
   "scroll": true,
   "theme": "simple",
   "transition": "slide"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}