to_csv benchmark

to_csv.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<xarray.DataArray 'mul-a39f522da57bff7c1e435f9a642d1171' (r: 25000, c: 750)>\n",
       "dask.array<shape=(25000, 750), dtype=float64, chunksize=(3125, 750)>\n",
       "Coordinates:\n",
       "  * r        (r) <U9 'row 0' 'row 1' 'row 2' 'row 3' 'row 4' 'row 5' 'row 6' ...\n",
       "  * c        (c) <U7 'col 0' 'col 1' 'col 2' 'col 3' 'col 4' 'col 5' 'col 6' ..."
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import xarray\n",
    "import dask.array as da\n",
    "from xarray_extras.csv import to_csv\n",
    "\n",
    "# Production sizes\n",
    "# ROWS = 500000\n",
    "# Quick test\n",
    "ROWS = 25000\n",
    "\n",
    "a = xarray.DataArray(da.random.random((ROWS, 750), chunks=(ROWS // 8, -1)) * 1e6,\n",
    "                     dims=['r', 'c'],\n",
    "                     coords={'r': ['row %d' % i for i in range(25000)],\n",
    "                             'c': ['col %d' % i for i in range(750)]})\n",
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "!rm -rf /dev/shm/{v1,v2}\n",
    "!mkdir /dev/shm/v2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 6 ms, sys: 5 ms, total: 11 ms\n",
      "Wall time: 10.9 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "future1 = to_csv(a, '/dev/shm/v1.csv.gz', compression='gzip', float_format='%.2f')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 205 ms, sys: 2.89 s, total: 3.1 s\n",
      "Wall time: 10 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "future1.compute()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 5.72 s, sys: 0 ns, total: 5.72 s\n",
      "Wall time: 5.77 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "future2 = a.to_dataset('c').to_dask_dataframe().to_csv(\n",
    "    '/dev/shm/v2', compression='gzip', float_format='%.2f', index=False, compute=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 2min 7s, sys: 18.4 s, total: 2min 25s\n",
      "Wall time: 1min 57s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "import dask\n",
    "dask.compute(*future2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}