Parallel loadz using concurrent.futures

parallel_loadz.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "9e0e063f",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import concurrent.futures"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "bc48edbe",
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_fake_data():\n",
    "    for i in range(20):\n",
    "        x = np.random.uniform(size=100_000_000)\n",
    "        np.savez(f\"./{i}.npz\", x)\n",
    "\n",
    "make_fake_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "7f1e0797",
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_one_file(i):\n",
    "    return np.load(f\"./{i}.npz\")['arr_0']\n",
    " \n",
    "def load_serial():\n",
    "    return [load_one_file(i) for i in range(20)]\n",
    "    \n",
    "def load_parallel(max_workers):\n",
    "    with concurrent.futures.ThreadPoolExecutor(max_workers=max_workers) as executor:\n",
    "        X = executor.map(load_one_file, range(20))\n",
    "    return list(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "957dc96e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 6.98 s, sys: 10.4 s, total: 17.3 s\n",
      "Wall time: 19 s\n"
     ]
    }
   ],
   "source": [
    "%time X = load_serial()\n",
    "del X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "560710dd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 6.71 s, sys: 10.1 s, total: 16.8 s\n",
      "Wall time: 17.6 s\n"
     ]
    }
   ],
   "source": [
    "%time X = load_parallel(1)\n",
    "del X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "ec5b9be6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 8.9 s, sys: 13.7 s, total: 22.6 s\n",
      "Wall time: 14.7 s\n"
     ]
    }
   ],
   "source": [
    "%time X = load_parallel(2)\n",
    "del X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "836cc182",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 11.6 s, sys: 19.6 s, total: 31.2 s\n",
      "Wall time: 13.9 s\n"
     ]
    }
   ],
   "source": [
    "%time X = load_parallel(4)\n",
    "del X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "53497a5c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 11.8 s, sys: 21.4 s, total: 33.2 s\n",
      "Wall time: 15.6 s\n"
     ]
    }
   ],
   "source": [
    "%time X = load_parallel(6)\n",
    "del X"
   ]
  }
 ],
 "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}