2_compare_results.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "89613f27-a847-4fbb-b3c1-d7dea25ecbd4",
   "metadata": {},
   "source": [
    "# Comparing data loading speeds with `kvikIO` and `Zarr` engine\n",
    "\n",
    "Benchmark details:\n",
    "- Loading a 1 year subset of the WeatherBench2 (~18.2GB)\n",
    "- Calculate median time to load entire dataset in 23 mini-batches (of batch size 32),\n",
    "  across 10 epochs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "b553b3d2-71ac-44e7-b6e4-da564bfdaf5e",
   "metadata": {},
   "outputs": [],
   "source": [
    "import importlib\n",
    "import time\n",
    "\n",
    "import lightning as L\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import torch\n",
    "from tqdm.auto import tqdm, trange\n",
    "\n",
    "module = importlib.import_module(name=\"1_benchmark_kvikIOzarr\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "53799aa7-3da5-49d3-a152-cc07cf82fedd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Empty dataframe to store results\n",
    "df = pd.DataFrame()\n",
    "# Optimize torch performance\n",
    "torch.set_float32_matmul_precision(precision=\"medium\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "4b36c1a7-8dd3-439a-9689-533f1eaa5960",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Training loop data loading function\n",
    "def training_loop(train_dataloader) -> list[float]:\n",
    "    epoch_timings: list = []\n",
    "    for epoch in trange(10):\n",
    "        # Start timing\n",
    "        tic: float = time.perf_counter()\n",
    "\n",
    "        # Mini-batch processing\n",
    "        for i, batch in tqdm(iterable=enumerate(train_dataloader), total=23):\n",
    "            input, target, metadata = batch\n",
    "            # Compute Mean Squared Error loss between t=0 and t=1, just for fun\n",
    "            loss: torch.Tensor = torch.functional.F.mse_loss(input=input, target=target)\n",
    "            # print(f\"Batch {i}, MSE Loss: {loss}\")\n",
    "\n",
    "        # Stop timing\n",
    "        toc: float = time.perf_counter()\n",
    "        epoch_timings.append(toc - tic)\n",
    "    return epoch_timings"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01793895-d093-4098-b793-fd333ed0e31c",
   "metadata": {},
   "source": [
    "## kvikIO engine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "867ada60-17cf-4fec-84c3-55ef6a727f79",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading data using kvikio engine\n"
     ]
    }
   ],
   "source": [
    "# Setup data\n",
    "datamodule: L.LightningDataModule = module.WeatherBench2DataModule(engine=\"kvikio\")\n",
    "datamodule.setup()\n",
    "train_dataloader = datamodule.train_dataloader()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f1209931-6bf7-4543-bc94-59e7382bb2ae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "375fcb11738d4528ae53edb29451cc59",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2352ab599366497292e5d1b894a51281",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "15ad8191ead24937ac85d770b73c7633",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9c6c0f6e891f479787a7e09420d771c1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4434a7d73eb54cd4817adc1248e9c711",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6598a5974fec44d9802dff5c13cb2e6d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6a6465b12679457ba09b6761ce2b983e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7d8fb9d3fb3146a18035a4041752d335",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "122dd90475f04e318f54a96cdd5a97bf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0051541caa3a49b899befd0587d933c4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4a4792893e44474bbccaeed4bcd8ad19",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Training loop\n",
    "epoch_timings: list[float] = training_loop(train_dataloader=train_dataloader)\n",
    "df[\"kvikio\"] = epoch_timings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "eb6a1389-cd13-41cf-86c7-695255a0f82a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total: 121.5782 seconds, Median: 11.8606 seconds/epoch, Mean: 12.1578 ± 0.8311 seconds/epoch\n"
     ]
    }
   ],
   "source": [
    "# Report timings\n",
    "total_time: float = df.kvikio.sum()\n",
    "median_time: float = df.kvikio.median()\n",
    "mean_time: float = df.kvikio.mean()\n",
    "std_time: float = df.kvikio.std()\n",
    "print(\n",
    "    f\"Total: {total_time:0.4f} seconds, \"\n",
    "    f\"Median: {median_time:0.4f} seconds/epoch, \"\n",
    "    f\"Mean: {mean_time:0.4f} ± {std_time:0.4f} seconds/epoch\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "03a6b722-1873-4a5b-a0bb-d231cbd86751",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "4e42e8b1-6b68-4392-8cf7-a2b7db222280",
   "metadata": {},
   "source": [
    "## Zarr engine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0ee14720-6e3e-4f98-8499-558d5c12b54a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading data using zarr engine\n"
     ]
    }
   ],
   "source": [
    "# Setup data\n",
    "datamodule: L.LightningDataModule = module.WeatherBench2DataModule(engine=\"zarr\")\n",
    "datamodule.setup()\n",
    "train_dataloader = datamodule.train_dataloader()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "b9b0efa9-b7e7-4817-9e6c-a16422424c69",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3a6706c2b54e429da8366abe832363ad",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9681e387d314430888bc874f3af9e51a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c1eece7b62f94b11b2e87bbf1e56dd17",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "307b0e0b5676442a9be53e8d037d70e9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e7ffad649cae4644a7871f1050923cfe",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b6bef0db7fb7496cb60ab96577fd47e8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e682263f3742438cab56a3015fd9bf83",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "331ee74722504b66a3ca84a2356ab1a6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d4139c00f6de4cb899b5e990f6eb38ae",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e39db6c413d24cd5bea9edd9ffc2f92e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "88fbea224f1b4f369eab8bc2f8d0ccd6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/23 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Training loop\n",
    "epoch_timings: list[float] = training_loop(train_dataloader=train_dataloader)\n",
    "df[\"zarr\"] = epoch_timings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "10024607-209b-47e9-aa34-e264638add84",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total: 159.2952 seconds, Median: 16.0240 seconds/epoch, Mean: 15.9295 ± 0.3761 seconds/epoch\n"
     ]
    }
   ],
   "source": [
    "# Report timings\n",
    "total_time: float = df.zarr.sum()\n",
    "median_time: float = df.zarr.median()\n",
    "mean_time: float = df.zarr.mean()\n",
    "std_time: float = df.zarr.std()\n",
    "print(\n",
    "    f\"Total: {total_time:0.4f} seconds, \"\n",
    "    f\"Median: {median_time:0.4f} seconds/epoch, \"\n",
    "    f\"Mean: {mean_time:0.4f} ± {std_time:0.4f} seconds/epoch\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1101537c-809c-4963-a5d5-301226a581c8",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "5d14428e-eebc-49e8-877b-0e002a416dd4",
   "metadata": {},
   "source": [
    "## Summary results\n",
    "\n",
    "Show time difference of using GPUDirect Storage `kvikio` engine vs CPU-based `zarr` engine."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "ca1dcb1c-41a7-42d1-b6e3-c823f6febf57",
   "metadata": {},
   "outputs": [],
   "source": [
    "# kivkIO engine should take less time than Zarr engine\n",
    "assert df.kvikio.median() < df.zarr.median()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "b10e6ce0-eaa5-4219-a7bd-dcc1d5bb34b0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "kvikIO engine uses 25.98% less time than Zarr engine\n"
     ]
    }
   ],
   "source": [
    "time_diff: float = (df.zarr.median() - df.kvikio.median()) / df.zarr.median()\n",
    "print(f\"kvikIO engine uses {time_diff * 100:0.2f}% less time than Zarr engine\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "6cc383e1-a15e-46bf-8e0d-ce323206c5d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_theme(context=\"talk\", palette=[\"#7400ff\", \"#e01073\"])\n",
    "ax = sns.barplot(data=df, estimator=\"median\")\n",
    "for container in ax.containers:\n",
    "    ax.bar_label(container=container, fontsize=11, fmt=\"%.1fs\", label_type=\"center\")\n",
    "ax.set_ylabel(ylabel=\"Median data load time per epoch\\n ◀ seconds, lower is better\")\n",
    "ax.set_xlabel(\n",
    "    xlabel=\"   (with GDS)            (without GDS) \\n\\n xarray backend engine\"\n",
    ")\n",
    "ax.set_title(label=\"Reading ERA5 data with/without GPUDirect Storage\")\n",
    "fig = ax.get_figure()\n",
    "fig.savefig(fname=\"figures/compare_kvikio_zarr.svg\", bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e91222bf-8619-4388-9f7d-5cd9609849b0",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "foss4g2023oceania",
   "language": "python",
   "name": "foss4g2023oceania"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}