You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@arrow.apache.org by GitBox <gi...@apache.org> on 2020/09/05 11:12:46 UTC
[GitHub] [arrow] alamb commented on a change in pull request #8116: ARROW-9919: [Rust][DataFusion] Speedup math operations by 15%+
alamb commented on a change in pull request #8116:
URL: https://github.com/apache/arrow/pull/8116#discussion_r483940437
##########
File path: rust/datafusion/benches/math_query_sql.rs
##########
@@ -0,0 +1,100 @@
+// 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.
+
+#[macro_use]
+extern crate criterion;
+use criterion::Criterion;
+
+use std::sync::Arc;
+
+extern crate arrow;
+extern crate datafusion;
+
+use arrow::{
+ array::{Float32Array, Float64Array},
+ datatypes::{DataType, Field, Schema},
+ record_batch::RecordBatch,
+};
+use datafusion::error::Result;
+
+use datafusion::datasource::MemTable;
+use datafusion::execution::context::ExecutionContext;
+
+fn query(ctx: &mut ExecutionContext, sql: &str) {
+ // execute the query
+ let df = ctx.sql(&sql).unwrap();
+ let results = df.collect().unwrap();
+
+ // display the relation
+ for _batch in results {}
+}
+
+fn create_context(array_len: usize, batch_size: usize) -> Result<ExecutionContext> {
+ // define a schema.
+ let schema = Arc::new(Schema::new(vec![
+ Field::new("f32", DataType::Float32, false),
+ Field::new("f64", DataType::Float64, false),
+ ]));
+
+ // define data.
+ let batches = (0..array_len / batch_size)
+ .map(|i| {
+ RecordBatch::try_new(
+ schema.clone(),
+ vec![
+ Arc::new(Float32Array::from(vec![i as f32; batch_size])),
+ Arc::new(Float64Array::from(vec![i as f64; batch_size])),
+ ],
+ )
+ .unwrap()
+ })
+ .collect::<Vec<_>>();
+
+ let mut ctx = ExecutionContext::new();
+
+ // declare a table in memory. In spark API, this corresponds to createDataFrame(...).
+ let provider = MemTable::new(schema, vec![batches])?;
+ ctx.register_table("t", Box::new(provider));
+
+ Ok(ctx)
+}
+
+fn criterion_benchmark(c: &mut Criterion) {
+ c.bench_function("sqrt_20_12", |b| {
Review comment:
BTW these benchmarks were taking ~ 200s on my laptop -- I think you can probably show a similar improvement with smaller dataset sizes (aka I think we should reduce the number of iterations and/or datasize so the benchmarks run in ~10sec - 20sec)
##########
File path: rust/datafusion/benches/math_query_sql.rs
##########
@@ -0,0 +1,100 @@
+// 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.
+
+#[macro_use]
+extern crate criterion;
+use criterion::Criterion;
+
+use std::sync::Arc;
+
+extern crate arrow;
+extern crate datafusion;
+
+use arrow::{
+ array::{Float32Array, Float64Array},
+ datatypes::{DataType, Field, Schema},
+ record_batch::RecordBatch,
+};
+use datafusion::error::Result;
+
+use datafusion::datasource::MemTable;
+use datafusion::execution::context::ExecutionContext;
+
+fn query(ctx: &mut ExecutionContext, sql: &str) {
+ // execute the query
+ let df = ctx.sql(&sql).unwrap();
+ let results = df.collect().unwrap();
+
+ // display the relation
+ for _batch in results {}
+}
+
+fn create_context(array_len: usize, batch_size: usize) -> Result<ExecutionContext> {
+ // define a schema.
+ let schema = Arc::new(Schema::new(vec![
+ Field::new("f32", DataType::Float32, false),
+ Field::new("f64", DataType::Float64, false),
+ ]));
+
+ // define data.
+ let batches = (0..array_len / batch_size)
+ .map(|i| {
+ RecordBatch::try_new(
+ schema.clone(),
+ vec![
+ Arc::new(Float32Array::from(vec![i as f32; batch_size])),
+ Arc::new(Float64Array::from(vec![i as f64; batch_size])),
+ ],
+ )
+ .unwrap()
+ })
+ .collect::<Vec<_>>();
+
+ let mut ctx = ExecutionContext::new();
+
+ // declare a table in memory. In spark API, this corresponds to createDataFrame(...).
+ let provider = MemTable::new(schema, vec![batches])?;
+ ctx.register_table("t", Box::new(provider));
+
+ Ok(ctx)
+}
+
+fn criterion_benchmark(c: &mut Criterion) {
+ c.bench_function("sqrt_20_12", |b| {
Review comment:
```
alamb@ip-192-168-1-129:~/Software/arrow/rust$ ARROW_TEST_DATA=`pwd`/../testing/data cargo bench --bench math_query_sql
Finished bench [optimized] target(s) in 0.15s
Running target/release/deps/math_query_sql-8428878a7376985f
Gnuplot not found, using plotters backend
Benchmarking sqrt_20_12: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 179.8s or reduce sample count to 10.
Benchmarking sqrt_20_12: Collecting 100 samples in estimated 179.78 s (5050 iterations)
...
```
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org