Add notes to performance comparisons notebook (#13044)

This PR adds a `note` section to the performance comparisons notebook to give users a disclaimer on what they need to do to run this notebook on lower-end hardware.

Authors:
  - GALI PREM SAGAR (https://github.com/galipremsagar)

Approvers:
  - Bradley Dice (https://github.com/bdice)
  - Ray Douglass (https://github.com/raydouglass)
  - Vyas Ramasubramani (https://github.com/vyasr)

URL: #13044

ci/test_notebooks.sh

@@ -34,7 +34,7 @@ pushd notebooks
 
 # Add notebooks that should be skipped here
 # (space-separated list of filenames without paths)
-SKIPNBS="performance_comparisons.ipynb"
+SKIPNBS="performance-comparisons.ipynb"
 
 EXITCODE=0
 trap "EXITCODE=1" ERR

docs/cudf/source/user_guide/index.md

@@ -12,7 +12,7 @@ groupby
 guide-to-udfs
 cupy-interop
 options
-performance-comparisons
+performance-comparisons/index
 PandasCompat
 copy-on-write
 ```

docs/cudf/source/user_guide/performance-comparisons/index.md

@@ -0,0 +1,8 @@
+# Performance comparisons
+
+```{toctree}
+:maxdepth: 2
+
+
+performance-comparisons
+```

docs/cudf/source/user_guide/performance_comparisons.ipynb → docs/cudf/source/user_guide/performance-comparisons/performance-comparisons.ipynb

@@ -8,13 +8,16 @@
    ]
   },
   {
+   "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "This notebook compares the performance of `cuDF` and `pandas`. The comparisons performed are on identical data sizes. This notebook primarily showcases the factor\n",
     "of speedups users can have when the similar `pandas` APIs are run on GPUs using `cudf`.\n",
     "\n",
-    "The hardware details used to run these performance comparisons are at the end of this page."
+    "The hardware details used to run these performance comparisons are at the end of this page.\n",
+    "\n",
+    "**Note**: This notebook is written to measure performance on NVIDIA GPUs with large memory. If running on hardware with lower memory, please consider lowering the `num_rows` values. Performance results may vary by data size, as well as the CPU and GPU used."
    ]
   },
   {
@@ -576,9 +579,10 @@
    },
    "outputs": [],
    "source": [
+    "num_rows = 300_000_000\n",
     "pd_series = pd.Series(\n",
     "    np.random.choice(\n",
-    "        [\"123\", \"56.234\", \"Walmart\", \"Costco\", \"rapids ai\"], size=300_000_000\n",
+    "        [\"123\", \"56.234\", \"Walmart\", \"Costco\", \"rapids ai\"], size=num_rows\n",
     "    )\n",
     ")"
    ]
@@ -1368,10 +1372,10 @@
    },
    "outputs": [],
    "source": [
-    "size = 100_000_000\n",
+    "num_rows = 100_000_000\n",
     "pdf = pd.DataFrame()\n",
-    "pdf[\"key\"] = np.random.randint(0, 2, size)\n",
-    "pdf[\"val\"] = np.random.randint(0, 7, size)\n",
+    "pdf[\"key\"] = np.random.randint(0, 2, num_rows)\n",
+    "pdf[\"val\"] = np.random.randint(0, 7, num_rows)\n",
     "\n",
     "\n",
     "def custom_formula_udf(df):\n",
@@ -1634,7 +1638,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.9"
+   "version": "3.10.10"
   },
   "vscode": {
    "interpreter": {

notebooks/performance-comparisons

@@ -0,0 +1 @@
+../docs/cudf/source/user_guide/performance-comparisons/