jayzhan211 commented on code in PR #11777: URL: https://github.com/apache/datafusion/pull/11777#discussion_r1701548962
########## datafusion/core/benches/high_cardinality.rs: ########## @@ -0,0 +1,129 @@ +// 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. + +use arrow::{ + datatypes::{DataType, Field, Schema}, + record_batch::RecordBatch, +}; +use arrow_array::{Int64Array, StringArray}; +use criterion::{criterion_group, criterion_main, Criterion}; +use datafusion::prelude::SessionContext; +use datafusion::{datasource::MemTable, error::Result}; +use futures::executor::block_on; +use std::sync::Arc; +use tokio::runtime::Runtime; + +async fn query(ctx: &mut SessionContext, sql: &str) { + let rt = Runtime::new().unwrap(); + + // execute the query + let df = rt.block_on(ctx.sql(sql)).unwrap(); + criterion::black_box(rt.block_on(df.collect()).unwrap()); +} + +fn create_context(array_len: usize, batch_size: usize) -> Result<SessionContext> { + // define a schema. + let schema = Arc::new(Schema::new(vec![ + Field::new("a", DataType::Int64, false), + Field::new("b", DataType::Utf8, false), + ])); + + // define data. + let batches = (0..array_len / batch_size) + .map(|_i| { + let data1 = (0..batch_size) + .into_iter() + .map(|x| x as i64) + .collect::<Vec<_>>(); + let data2 = (0..batch_size) + .into_iter() + .map(|j| format!("a{j}")) + .collect::<Vec<_>>(); + + RecordBatch::try_new( + schema.clone(), + vec![ + Arc::new(Int64Array::from(data1)), + Arc::new(StringArray::from(data2)), + ], + ) + .unwrap() + }) + .collect::<Vec<_>>(); + + let ctx = SessionContext::new(); + + // declare a table in memory. In spark API, this corresponds to createDataFrame(...). + let provider = MemTable::try_new(schema, vec![batches])?; + ctx.register_table("t", Arc::new(provider))?; + + Ok(ctx) +} + +fn criterion_benchmark(c: &mut Criterion) { + let array_len = 2000000; // 2M rows + let batch_size = array_len; + + c.bench_function("benchmark", |b| { + let mut ctx = create_context(array_len, batch_size).unwrap(); + b.iter(|| block_on(query(&mut ctx, "select a, b, count(*) from t group by a, b order by count(*) desc limit 10"))) + }); +} + +criterion_group! { + name = benches; + // This can be any expression that returns a `Criterion` object. + config = Criterion::default().sample_size(10); + targets = criterion_benchmark +} +criterion_main!(benches); + +// Previous result +// reuse-hash +// benchmark time: [2.5999 s 6.3132 s 11.062 s] +// Found 1 outliers among 10 measurements (10.00%) + +// main +// benchmark time: [4.1404 s 8.4601 s 13.226 s] + +// Latest reuslt +// single-multi-groupby +// Gnuplot not found, using plotters backend +// benchmark time: [82.970 ms 98.723 ms 111.32 ms] +// change: [-99.328% -98.932% -97.777%] (p = 0.00 < 0.05) +// Performance has improved. +// Found 2 outliers among 10 measurements (20.00%) +// 2 (20.00%) high mild + +// main (I guess this improves because of the change Check hashes first during probing the aggr hash table #11718, but I didn't verify) +// Gnuplot not found, using plotters backend +// Benchmarking benchmark: Warming up for 3.0000 s +// Warning: Unable to complete 10 samples in 5.0s. You may wish to increase target time to 23.3s. +// benchmark time: [660.82 ms 1.3354 s 2.1247 s] +// change: [+675.04% +1377.8% +2344.1%] (p = 0.00 < 0.05) +// Performance has regressed. +// Found 1 outliers among 10 measurements (10.00%) +// 1 (10.00%) high mild + + +// Gnuplot not found, using plotters backend +// Benchmarking benchmark: Warming up for 3.0000 s +// Warning: Unable to complete 10 samples in 5.0s. You may wish to increase target time to 8.2s or enable flat sampling. +// benchmark time: [150.22 ms 154.31 ms 158.06 ms] +// change: [+254.76% +261.92% +270.28%] (p = 0.00 < 0.05) +// Performance has regressed. +// Found 1 outliers among 10 measurements (10.00%) Review Comment: ``` // Gnuplot not found, using plotters backend // Benchmarking benchmark: Warming up for 3.0000 s // Warning: Unable to complete 10 samples in 5.0s. You may wish to increase target time to 8.2s or enable flat sampling. // benchmark time: [150.22 ms 154.31 ms 158.06 ms] // change: [+254.76% +261.92% +270.28%] (p = 0.00 < 0.05) // Performance has regressed. // Found 1 outliers among 10 measurements (10.00%) // Gnuplot not found, using plotters backend // Benchmarking benchmark: Warming up for 3.0000 s // Warning: Unable to complete 10 samples in 5.0s. You may wish to increase target time to 29.0s. // benchmark time: [2.0650 s 3.5149 s 5.1064 s] // change: [+1182.7% +2229.6% +3244.4%] (p = 0.00 < 0.05) // Performance has regressed. ``` -- 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. To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For additional commands, e-mail: github-h...@datafusion.apache.org
