Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
77 changes: 77 additions & 0 deletions benchmarks/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,77 @@
<!-- Licensed to the Apache Software Foundation (ASF) under one
or more contributor license agreements. See the NOTICE file
distributed with this work for additional information
regarding copyright ownership. The ASF licenses this file
to you under the Apache License, Version 2.0 (the
"License"); you may not use this file except in compliance
with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing,
software distributed under the License is distributed on an
"AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
KIND, either express or implied. See the License for the
specific language governing permissions and limitations
under the License. -->

# Running Benchmarks

## pytest-benchmark

These benchmarks provide a convenient way to compare the results of running queries on sedona-db to other engines like DuckDB and postgis.

### Setup

Install pytest-benchmark:
```bash
pip install pytest-benchmark
```

### Running benchmarks

The below commands assume your working directory is in `benchmarks`.

```bash
cd benchmarks/
```

To run a benchmark, simply run the corresponding test function. For example, to run the benchmarks for st_buffer, you can run

```bash
pytest test_functions.py::TestBenchFunctions::test_st_buffer
```

Most of the time, you'll also want to group by `param:table` or `func` (function) by using the `--benchmark-group-by=param:table` flag. pytest-benchmark will highlight the "best" value in green (e.g fastest for median, lowest for stddev) and "worse" value in red for each column per each group.

```bash
pytest --benchmark-group-by=param:table test_functions.py::TestBenchFunctions::test_st_buffer
```

You can also reduce the number of columns that display by using the `--benchmark-columns` flag.

```bash
pytest --benchmark-group-by=param:table --benchmark-columns=median,mean,stddev test_functions.py::TestBenchFunctions::test_st_buffer
```

Example output of the last command:

```
----------------------------- benchmark 'table=collections_complex': 3 tests -----------------------------
Name (time in ms) Median Mean StdDev
----------------------------------------------------------------------------------------------------------
test_st_buffer[collections_complex-SedonaDB] 87.0095 (1.0) 87.7874 (1.0) 3.7269 (1.0)
test_st_buffer[collections_complex-DuckDB] 440.4810 (5.06) 444.6948 (5.07) 12.1143 (3.25)
test_st_buffer[collections_complex-PostGIS] 864.5841 (9.94) 883.3661 (10.06) 50.4996 (13.55)
----------------------------------------------------------------------------------------------------------

---------------------------- benchmark 'table=collections_simple': 3 tests -----------------------------
Name (time in ms) Median Mean StdDev
--------------------------------------------------------------------------------------------------------
test_st_buffer[collections_simple-SedonaDB] 85.8510 (1.0) 86.5050 (1.0) 3.8481 (1.0)
test_st_buffer[collections_simple-DuckDB] 442.6664 (5.16) 444.5187 (5.14) 5.6186 (1.46)
test_st_buffer[collections_simple-PostGIS] 855.3329 (9.96) 854.7194 (9.88) 7.6190 (1.98)
--------------------------------------------------------------------------------------------------------
```

For more details and command line options, refer to the official [pytest-benchmark documentation](https://pytest-benchmark.readthedocs.io/en/latest/usage.html)
94 changes: 94 additions & 0 deletions benchmarks/test_bench_base.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,94 @@
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements. See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership. The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.
import json
from sedonadb.testing import DuckDB, PostGIS, SedonaDB


class TestBenchBase:
def setup_class(self):
self.sedonadb = SedonaDB.create_or_skip()
self.postgis = PostGIS.create_or_skip()
self.duckdb = DuckDB.create_or_skip()

num_geoms = 100_000

# Setup tables
for name, options in [
(
"segments_large",
{
"geom_type": "LineString",
"target_rows": num_geoms,
"vertices_per_linestring_range": [2, 2],
},
),
(
"polygons_simple",
{
"geom_type": "Polygon",
"target_rows": num_geoms,
"vertices_per_linestring_range": [10, 10],
},
),
(
"polygons_complex",
{
"geom_type": "Polygon",
"target_rows": num_geoms,
"vertices_per_linestring_range": [500, 500],
},
),
(
"collections_simple",
{
"geom_type": "GeometryCollection",
"target_rows": num_geoms,
"vertices_per_linestring_range": [10, 10],
},
),
(
"collections_complex",
{
"geom_type": "GeometryCollection",
"target_rows": num_geoms,
"vertices_per_linestring_range": [500, 500],
},
),
]:
# Generate synthetic data
query = f"""
SELECT
geometry as geom1,
geometry as geom2,
round(random() * 100) as integer
FROM sd_random_geometry('{json.dumps(options)}')
"""
tab = self.sedonadb.execute_and_collect(query)

self.sedonadb.create_table_arrow(name, tab)
self.postgis.create_table_arrow(name, tab)
self.duckdb.create_table_arrow(name, tab)

def _get_eng(self, eng):
if eng == SedonaDB:
return self.sedonadb
elif eng == PostGIS:
return self.postgis
elif eng == DuckDB:
return self.duckdb
else:
raise ValueError(f"Unsupported engine: {eng}")
37 changes: 37 additions & 0 deletions benchmarks/test_distance.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,37 @@
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements. See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership. The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.
import pytest
from test_bench_base import TestBenchBase
from sedonadb.testing import DuckDB, PostGIS, SedonaDB


class TestBenchPredicates(TestBenchBase):
@pytest.mark.parametrize("eng", [SedonaDB, PostGIS, DuckDB])
@pytest.mark.parametrize(
"table",
[
"polygons_simple",
"polygons_complex",
],
)
def test_st_distance(self, benchmark, eng, table):
eng = self._get_eng(eng)

def queries():
eng.execute_and_collect(f"SELECT ST_Distance(geom1, geom2) from {table}")

benchmark(queries)
117 changes: 117 additions & 0 deletions benchmarks/test_functions.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,117 @@
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements. See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership. The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License. You may obtain a copy of the License at

# http://www.apache.org/licenses/LICENSE-2.0

# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.
import pytest
from test_bench_base import TestBenchBase
from sedonadb.testing import DuckDB, PostGIS, SedonaDB


class TestBenchFunctions(TestBenchBase):
@pytest.mark.parametrize("eng", [SedonaDB, PostGIS, DuckDB])
@pytest.mark.parametrize(
"table",
[
"polygons_simple",
"polygons_complex",
],
)
def test_st_area(self, benchmark, eng, table):
eng = self._get_eng(eng)

def queries():
eng.execute_and_collect(f"SELECT ST_Area(geom1) from {table}")

benchmark(queries)

@pytest.mark.parametrize("eng", [SedonaDB, PostGIS, DuckDB])
@pytest.mark.parametrize(
"table",
[
"collections_simple",
"collections_complex",
],
)
def test_st_buffer(self, benchmark, eng, table):
eng = self._get_eng(eng)

def queries():
eng.execute_and_collect(f"SELECT ST_Buffer(geom1, 2.0) from {table}")

benchmark(queries)

@pytest.mark.parametrize("eng", [SedonaDB, PostGIS, DuckDB])
@pytest.mark.parametrize(
"table",
[
"polygons_simple",
"polygons_complex",
],
)
def test_st_centroid(self, benchmark, eng, table):
eng = self._get_eng(eng)

def queries():
eng.execute_and_collect(f"SELECT ST_Centroid(geom1) from {table}")

benchmark(queries)

@pytest.mark.parametrize("eng", [SedonaDB, PostGIS, DuckDB])
@pytest.mark.parametrize(
"table",
[
"collections_simple",
"collections_complex",
],
)
def test_st_dimension(self, benchmark, eng, table):
eng = self._get_eng(eng)

def queries():
eng.execute_and_collect(f"SELECT ST_Dimension(geom1) from {table}")

benchmark(queries)

@pytest.mark.parametrize("eng", [SedonaDB, PostGIS, DuckDB])
@pytest.mark.parametrize(
"table",
[
"collections_simple",
"collections_complex",
],
)
def test_st_envelope(self, benchmark, eng, table):
eng = self._get_eng(eng)

def queries():
eng.execute_and_collect(f"SELECT ST_Envelope(geom1) from {table}")

benchmark(queries)

@pytest.mark.parametrize("eng", [SedonaDB, PostGIS, DuckDB])
@pytest.mark.parametrize(
"table",
[
"collections_simple",
"collections_complex",
],
)
def test_st_geometrytype(self, benchmark, eng, table):
eng = self._get_eng(eng)

def queries():
eng.execute_and_collect(f"SELECT ST_GeometryType(geom1) from {table}")

benchmark(queries)
37 changes: 37 additions & 0 deletions benchmarks/test_overlay.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,37 @@
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements. See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership. The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.
import pytest
from test_bench_base import TestBenchBase
from sedonadb.testing import DuckDB, PostGIS, SedonaDB


class TestBenchPredicates(TestBenchBase):
@pytest.mark.parametrize("eng", [SedonaDB, PostGIS, DuckDB])
@pytest.mark.parametrize(
"table",
[
"polygons_simple",
"polygons_complex",
],
)
def test_st_difference(self, benchmark, eng, table):
eng = self._get_eng(eng)

def queries():
eng.execute_and_collect(f"SELECT ST_Difference(geom1, geom2) from {table}")

benchmark(queries)
Loading