mirror of
https://github.com/quickwit-oss/tantivy.git
synced 2025-12-23 02:29:57 +00:00
use bingang for agg benchmark (#2378)
* use bingang for agg benchmark use bingang for agg benchmark, which includes memory consumption Output: ``` full histogram Memory: 15.8 KB Avg: 10.9322ms (+5.44%) Median: 10.8790ms (+9.28%) Min: 10.7470ms Max: 11.3263ms histogram_hard_bounds Memory: 15.5 KB Avg: 5.1939ms (+6.61%) Median: 5.1722ms (+10.98%) Min: 5.0432ms Max: 5.3910ms histogram_with_avg_sub_agg Memory: 48.7 KB Avg: 23.8165ms (+4.57%) Median: 23.7264ms (+10.06%) Min: 23.4995ms Max: 24.8107ms dense histogram Memory: 17.3 KB Avg: 15.6810ms (-8.54%) Median: 15.6174ms (-8.89%) Min: 15.4953ms Max: 16.0702ms histogram_hard_bounds Memory: 15.4 KB Avg: 10.0720ms (-7.33%) Median: 10.0572ms (-7.06%) Min: 9.8500ms Max: 10.4819ms histogram_with_avg_sub_agg Memory: 50.1 KB Avg: 33.0993ms (-7.04%) Median: 32.9499ms (-6.86%) Min: 32.8284ms Max: 34.0529ms sparse histogram Memory: 16.3 KB Avg: 19.2325ms (-0.44%) Median: 19.1211ms (-1.26%) Min: 19.0348ms Max: 19.7902ms histogram_hard_bounds Memory: 16.1 KB Avg: 18.5179ms (-0.61%) Median: 18.4552ms (-0.90%) Min: 18.3799ms Max: 19.0535ms histogram_with_avg_sub_agg Memory: 34.7 KB Avg: 21.2589ms (-0.69%) Median: 21.1867ms (-1.05%) Min: 21.0342ms Max: 21.9900ms ``` * add more bench with term as sub agg
This commit is contained in:
PSeitz
GitHub
@@ -67,6 +67,7 @@ fnv = "1.0.7"
|
||||
winapi = "0.3.9"
|
||||
|
||||
[dev-dependencies]
|
||||
binggan = "0.5.1"
|
||||
rand = "0.8.5"
|
||||
maplit = "1.0.2"
|
||||
matches = "0.1.9"
|
||||
@@ -143,3 +144,8 @@ harness = false
|
||||
[[bench]]
|
||||
name = "index-bench"
|
||||
harness = false
|
||||
|
||||
[[bench]]
|
||||
name = "agg_bench"
|
||||
harness = false
|
||||
|
||||
|
||||
413
benches/agg_bench.rs
Normal file
413
benches/agg_bench.rs
Normal file
@@ -0,0 +1,413 @@
|
||||
use binggan::{black_box, InputGroup, PeakMemAlloc, INSTRUMENTED_SYSTEM};
|
||||
use rand::prelude::SliceRandom;
|
||||
use rand::rngs::StdRng;
|
||||
use rand::{Rng, SeedableRng};
|
||||
use rand_distr::Distribution;
|
||||
use serde_json::json;
|
||||
use tantivy::aggregation::agg_req::Aggregations;
|
||||
use tantivy::aggregation::AggregationCollector;
|
||||
use tantivy::query::{AllQuery, TermQuery};
|
||||
use tantivy::schema::{IndexRecordOption, Schema, TextFieldIndexing, FAST, STRING};
|
||||
use tantivy::{doc, Index, Term};
|
||||
|
||||
#[global_allocator]
|
||||
pub static GLOBAL: &PeakMemAlloc<std::alloc::System> = &INSTRUMENTED_SYSTEM;
|
||||
|
||||
/// Mini macro to register a function via its name
|
||||
/// runner.register("average_u64", move |index| average_u64(index));
|
||||
macro_rules! register {
|
||||
($runner:expr, $func:ident) => {
|
||||
$runner.register(stringify!($func), move |index| $func(index))
|
||||
};
|
||||
}
|
||||
|
||||
fn main() {
|
||||
let inputs = vec![
|
||||
("full", get_test_index_bench(Cardinality::Full).unwrap()),
|
||||
(
|
||||
"dense",
|
||||
get_test_index_bench(Cardinality::OptionalDense).unwrap(),
|
||||
),
|
||||
(
|
||||
"sparse",
|
||||
get_test_index_bench(Cardinality::OptionalSparse).unwrap(),
|
||||
),
|
||||
(
|
||||
"multivalue",
|
||||
get_test_index_bench(Cardinality::Multivalued).unwrap(),
|
||||
),
|
||||
];
|
||||
|
||||
bench_agg(InputGroup::new_with_inputs(inputs));
|
||||
}
|
||||
|
||||
fn bench_agg(mut group: InputGroup<Index>) {
|
||||
group.set_alloc(GLOBAL); // Set the peak mem allocator. This will enable peak memory reporting.
|
||||
register!(group, average_u64);
|
||||
register!(group, average_f64);
|
||||
register!(group, average_f64_u64);
|
||||
register!(group, stats_f64);
|
||||
register!(group, percentiles_f64);
|
||||
register!(group, terms_few);
|
||||
register!(group, terms_many);
|
||||
register!(group, terms_many_order_by_term);
|
||||
register!(group, terms_many_with_top_hits);
|
||||
register!(group, terms_many_with_avg_sub_agg);
|
||||
register!(group, terms_many_json_mixed_type_with_sub_agg_card);
|
||||
register!(group, range_agg);
|
||||
register!(group, range_agg_with_avg_sub_agg);
|
||||
register!(group, range_agg_with_term_agg_few);
|
||||
register!(group, range_agg_with_term_agg_many);
|
||||
register!(group, histogram);
|
||||
register!(group, histogram_hard_bounds);
|
||||
register!(group, histogram_with_avg_sub_agg);
|
||||
register!(group, avg_and_range_with_avg_sub_agg);
|
||||
|
||||
group.run();
|
||||
}
|
||||
|
||||
fn exec_term_with_agg(index: &Index, agg_req: serde_json::Value) {
|
||||
let agg_req: Aggregations = serde_json::from_value(agg_req).unwrap();
|
||||
|
||||
let reader = index.reader().unwrap();
|
||||
let text_field = reader.searcher().schema().get_field("text").unwrap();
|
||||
let term_query = TermQuery::new(
|
||||
Term::from_field_text(text_field, "cool"),
|
||||
IndexRecordOption::Basic,
|
||||
);
|
||||
let collector = get_collector(agg_req);
|
||||
let searcher = reader.searcher();
|
||||
black_box(searcher.search(&term_query, &collector).unwrap());
|
||||
}
|
||||
|
||||
fn average_u64(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"average": { "avg": { "field": "score", } }
|
||||
});
|
||||
exec_term_with_agg(index, agg_req)
|
||||
}
|
||||
fn average_f64(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"average": { "avg": { "field": "score_f64", } }
|
||||
});
|
||||
exec_term_with_agg(index, agg_req)
|
||||
}
|
||||
fn average_f64_u64(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"average_f64": { "avg": { "field": "score_f64" } },
|
||||
"average": { "avg": { "field": "score" } },
|
||||
});
|
||||
exec_term_with_agg(index, agg_req)
|
||||
}
|
||||
fn stats_f64(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"average_f64": { "stats": { "field": "score_f64", } }
|
||||
});
|
||||
exec_term_with_agg(index, agg_req)
|
||||
}
|
||||
|
||||
fn percentiles_f64(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"mypercentiles": {
|
||||
"percentiles": {
|
||||
"field": "score_f64",
|
||||
"percents": [ 95, 99, 99.9 ]
|
||||
}
|
||||
}
|
||||
});
|
||||
execute_agg(index, agg_req);
|
||||
}
|
||||
fn terms_few(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"my_texts": { "terms": { "field": "text_few_terms" } },
|
||||
});
|
||||
execute_agg(index, agg_req);
|
||||
}
|
||||
fn terms_many(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"my_texts": { "terms": { "field": "text_many_terms" } },
|
||||
});
|
||||
execute_agg(index, agg_req);
|
||||
}
|
||||
fn terms_many_order_by_term(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"my_texts": { "terms": { "field": "text_many_terms", "order": { "_key": "desc" } } },
|
||||
});
|
||||
execute_agg(index, agg_req);
|
||||
}
|
||||
fn terms_many_with_top_hits(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"my_texts": {
|
||||
"terms": { "field": "text_many_terms" },
|
||||
"aggs": {
|
||||
"top_hits": { "top_hits":
|
||||
{
|
||||
"sort": [
|
||||
{ "score": "desc" }
|
||||
],
|
||||
"size": 2,
|
||||
"doc_value_fields": ["score_f64"]
|
||||
}
|
||||
}
|
||||
}
|
||||
},
|
||||
});
|
||||
execute_agg(index, agg_req);
|
||||
}
|
||||
fn terms_many_with_avg_sub_agg(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"my_texts": {
|
||||
"terms": { "field": "text_many_terms" },
|
||||
"aggs": {
|
||||
"average_f64": { "avg": { "field": "score_f64" } }
|
||||
}
|
||||
},
|
||||
});
|
||||
execute_agg(index, agg_req);
|
||||
}
|
||||
fn terms_many_json_mixed_type_with_sub_agg_card(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"my_texts": {
|
||||
"terms": { "field": "json.mixed_type" },
|
||||
"aggs": {
|
||||
"average_f64": { "avg": { "field": "score_f64" } }
|
||||
}
|
||||
},
|
||||
});
|
||||
execute_agg(index, agg_req);
|
||||
}
|
||||
|
||||
fn execute_agg(index: &Index, agg_req: serde_json::Value) {
|
||||
let agg_req: Aggregations = serde_json::from_value(agg_req).unwrap();
|
||||
let collector = get_collector(agg_req);
|
||||
|
||||
let reader = index.reader().unwrap();
|
||||
let searcher = reader.searcher();
|
||||
black_box(searcher.search(&AllQuery, &collector).unwrap());
|
||||
}
|
||||
fn range_agg(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"range_f64": { "range": { "field": "score_f64", "ranges": [
|
||||
{ "from": 3, "to": 7000 },
|
||||
{ "from": 7000, "to": 20000 },
|
||||
{ "from": 20000, "to": 30000 },
|
||||
{ "from": 30000, "to": 40000 },
|
||||
{ "from": 40000, "to": 50000 },
|
||||
{ "from": 50000, "to": 60000 }
|
||||
] } },
|
||||
});
|
||||
execute_agg(index, agg_req);
|
||||
}
|
||||
fn range_agg_with_avg_sub_agg(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"rangef64": {
|
||||
"range": {
|
||||
"field": "score_f64",
|
||||
"ranges": [
|
||||
{ "from": 3, "to": 7000 },
|
||||
{ "from": 7000, "to": 20000 },
|
||||
{ "from": 20000, "to": 30000 },
|
||||
{ "from": 30000, "to": 40000 },
|
||||
{ "from": 40000, "to": 50000 },
|
||||
{ "from": 50000, "to": 60000 }
|
||||
]
|
||||
},
|
||||
"aggs": {
|
||||
"average_f64": { "avg": { "field": "score_f64" } }
|
||||
}
|
||||
},
|
||||
});
|
||||
execute_agg(index, agg_req);
|
||||
}
|
||||
|
||||
fn range_agg_with_term_agg_few(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"rangef64": {
|
||||
"range": {
|
||||
"field": "score_f64",
|
||||
"ranges": [
|
||||
{ "from": 3, "to": 7000 },
|
||||
{ "from": 7000, "to": 20000 },
|
||||
{ "from": 20000, "to": 30000 },
|
||||
{ "from": 30000, "to": 40000 },
|
||||
{ "from": 40000, "to": 50000 },
|
||||
{ "from": 50000, "to": 60000 }
|
||||
]
|
||||
},
|
||||
"aggs": {
|
||||
"my_texts": { "terms": { "field": "text_few_terms" } },
|
||||
}
|
||||
},
|
||||
});
|
||||
execute_agg(index, agg_req);
|
||||
}
|
||||
fn range_agg_with_term_agg_many(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"rangef64": {
|
||||
"range": {
|
||||
"field": "score_f64",
|
||||
"ranges": [
|
||||
{ "from": 3, "to": 7000 },
|
||||
{ "from": 7000, "to": 20000 },
|
||||
{ "from": 20000, "to": 30000 },
|
||||
{ "from": 30000, "to": 40000 },
|
||||
{ "from": 40000, "to": 50000 },
|
||||
{ "from": 50000, "to": 60000 }
|
||||
]
|
||||
},
|
||||
"aggs": {
|
||||
"my_texts": { "terms": { "field": "text_many_terms" } },
|
||||
}
|
||||
},
|
||||
});
|
||||
execute_agg(index, agg_req);
|
||||
}
|
||||
fn histogram(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"rangef64": {
|
||||
"histogram": {
|
||||
"field": "score_f64",
|
||||
"interval": 100 // 1000 buckets
|
||||
},
|
||||
}
|
||||
});
|
||||
execute_agg(index, agg_req);
|
||||
}
|
||||
fn histogram_hard_bounds(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"rangef64": { "histogram": { "field": "score_f64", "interval": 100, "hard_bounds": { "min": 1000, "max": 300000 } } },
|
||||
});
|
||||
execute_agg(index, agg_req);
|
||||
}
|
||||
fn histogram_with_avg_sub_agg(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"rangef64": {
|
||||
"histogram": { "field": "score_f64", "interval": 100 },
|
||||
"aggs": {
|
||||
"average_f64": { "avg": { "field": "score_f64" } }
|
||||
}
|
||||
}
|
||||
});
|
||||
execute_agg(index, agg_req);
|
||||
}
|
||||
fn avg_and_range_with_avg_sub_agg(index: &Index) {
|
||||
let agg_req = json!({
|
||||
"rangef64": {
|
||||
"range": {
|
||||
"field": "score_f64",
|
||||
"ranges": [
|
||||
{ "from": 3, "to": 7000 },
|
||||
{ "from": 7000, "to": 20000 },
|
||||
{ "from": 20000, "to": 60000 }
|
||||
]
|
||||
},
|
||||
"aggs": {
|
||||
"average_in_range": { "avg": { "field": "score" } }
|
||||
}
|
||||
},
|
||||
"average": { "avg": { "field": "score" } }
|
||||
});
|
||||
execute_agg(index, agg_req);
|
||||
}
|
||||
|
||||
#[derive(Clone, Copy, Hash, Default, Debug, PartialEq, Eq, PartialOrd, Ord)]
|
||||
enum Cardinality {
|
||||
/// All documents contain exactly one value.
|
||||
/// `Full` is the default for auto-detecting the Cardinality, since it is the most strict.
|
||||
#[default]
|
||||
Full = 0,
|
||||
/// All documents contain at most one value.
|
||||
OptionalDense = 1,
|
||||
/// All documents may contain any number of values.
|
||||
Multivalued = 2,
|
||||
/// 1 / 20 documents has a value
|
||||
OptionalSparse = 3,
|
||||
}
|
||||
|
||||
fn get_collector(agg_req: Aggregations) -> AggregationCollector {
|
||||
AggregationCollector::from_aggs(agg_req, Default::default())
|
||||
}
|
||||
|
||||
fn get_test_index_bench(cardinality: Cardinality) -> tantivy::Result<Index> {
|
||||
let mut schema_builder = Schema::builder();
|
||||
let text_fieldtype = tantivy::schema::TextOptions::default()
|
||||
.set_indexing_options(
|
||||
TextFieldIndexing::default().set_index_option(IndexRecordOption::WithFreqs),
|
||||
)
|
||||
.set_stored();
|
||||
let text_field = schema_builder.add_text_field("text", text_fieldtype);
|
||||
let json_field = schema_builder.add_json_field("json", FAST);
|
||||
let text_field_many_terms = schema_builder.add_text_field("text_many_terms", STRING | FAST);
|
||||
let text_field_few_terms = schema_builder.add_text_field("text_few_terms", STRING | FAST);
|
||||
let score_fieldtype = tantivy::schema::NumericOptions::default().set_fast();
|
||||
let score_field = schema_builder.add_u64_field("score", score_fieldtype.clone());
|
||||
let score_field_f64 = schema_builder.add_f64_field("score_f64", score_fieldtype.clone());
|
||||
let score_field_i64 = schema_builder.add_i64_field("score_i64", score_fieldtype);
|
||||
let index = Index::create_from_tempdir(schema_builder.build())?;
|
||||
let few_terms_data = ["INFO", "ERROR", "WARN", "DEBUG"];
|
||||
|
||||
let lg_norm = rand_distr::LogNormal::new(2.996f64, 0.979f64).unwrap();
|
||||
|
||||
let many_terms_data = (0..150_000)
|
||||
.map(|num| format!("author{}", num))
|
||||
.collect::<Vec<_>>();
|
||||
{
|
||||
let mut rng = StdRng::from_seed([1u8; 32]);
|
||||
let mut index_writer = index.writer_with_num_threads(1, 200_000_000)?;
|
||||
// To make the different test cases comparable we just change one doc to force the
|
||||
// cardinality
|
||||
if cardinality == Cardinality::OptionalDense {
|
||||
index_writer.add_document(doc!())?;
|
||||
}
|
||||
if cardinality == Cardinality::Multivalued {
|
||||
index_writer.add_document(doc!(
|
||||
json_field => json!({"mixed_type": 10.0}),
|
||||
json_field => json!({"mixed_type": 10.0}),
|
||||
text_field => "cool",
|
||||
text_field => "cool",
|
||||
text_field_many_terms => "cool",
|
||||
text_field_many_terms => "cool",
|
||||
text_field_few_terms => "cool",
|
||||
text_field_few_terms => "cool",
|
||||
score_field => 1u64,
|
||||
score_field => 1u64,
|
||||
score_field_f64 => lg_norm.sample(&mut rng),
|
||||
score_field_f64 => lg_norm.sample(&mut rng),
|
||||
score_field_i64 => 1i64,
|
||||
score_field_i64 => 1i64,
|
||||
))?;
|
||||
}
|
||||
let mut doc_with_value = 1_000_000;
|
||||
if cardinality == Cardinality::OptionalSparse {
|
||||
doc_with_value /= 20;
|
||||
}
|
||||
let _val_max = 1_000_000.0;
|
||||
for _ in 0..doc_with_value {
|
||||
let val: f64 = rng.gen_range(0.0..1_000_000.0);
|
||||
let json = if rng.gen_bool(0.1) {
|
||||
// 10% are numeric values
|
||||
json!({ "mixed_type": val })
|
||||
} else {
|
||||
json!({"mixed_type": many_terms_data.choose(&mut rng).unwrap().to_string()})
|
||||
};
|
||||
index_writer.add_document(doc!(
|
||||
text_field => "cool",
|
||||
json_field => json,
|
||||
text_field_many_terms => many_terms_data.choose(&mut rng).unwrap().to_string(),
|
||||
text_field_few_terms => few_terms_data.choose(&mut rng).unwrap().to_string(),
|
||||
score_field => val as u64,
|
||||
score_field_f64 => lg_norm.sample(&mut rng),
|
||||
score_field_i64 => val as i64,
|
||||
))?;
|
||||
if cardinality == Cardinality::OptionalSparse {
|
||||
for _ in 0..20 {
|
||||
index_writer.add_document(doc!(text_field => "cool"))?;
|
||||
}
|
||||
}
|
||||
}
|
||||
// writing the segment
|
||||
index_writer.commit()?;
|
||||
}
|
||||
|
||||
Ok(index)
|
||||
}
|
||||
@@ -1,585 +0,0 @@
|
||||
#[cfg(all(test, feature = "unstable"))]
|
||||
mod bench {
|
||||
|
||||
use rand::prelude::SliceRandom;
|
||||
use rand::rngs::StdRng;
|
||||
use rand::{Rng, SeedableRng};
|
||||
use rand_distr::Distribution;
|
||||
use serde_json::json;
|
||||
use test::{self, Bencher};
|
||||
|
||||
use crate::aggregation::agg_req::Aggregations;
|
||||
use crate::aggregation::AggregationCollector;
|
||||
use crate::query::{AllQuery, TermQuery};
|
||||
use crate::schema::{IndexRecordOption, Schema, TextFieldIndexing, FAST, STRING};
|
||||
use crate::{Index, Term};
|
||||
|
||||
#[derive(Clone, Copy, Hash, Default, Debug, PartialEq, Eq, PartialOrd, Ord)]
|
||||
enum Cardinality {
|
||||
/// All documents contain exactly one value.
|
||||
/// `Full` is the default for auto-detecting the Cardinality, since it is the most strict.
|
||||
#[default]
|
||||
Full = 0,
|
||||
/// All documents contain at most one value.
|
||||
Optional = 1,
|
||||
/// All documents may contain any number of values.
|
||||
Multivalued = 2,
|
||||
/// 1 / 20 documents has a value
|
||||
Sparse = 3,
|
||||
}
|
||||
|
||||
fn get_collector(agg_req: Aggregations) -> AggregationCollector {
|
||||
AggregationCollector::from_aggs(agg_req, Default::default())
|
||||
}
|
||||
|
||||
fn get_test_index_bench(cardinality: Cardinality) -> crate::Result<Index> {
|
||||
let mut schema_builder = Schema::builder();
|
||||
let text_fieldtype = crate::schema::TextOptions::default()
|
||||
.set_indexing_options(
|
||||
TextFieldIndexing::default().set_index_option(IndexRecordOption::WithFreqs),
|
||||
)
|
||||
.set_stored();
|
||||
let text_field = schema_builder.add_text_field("text", text_fieldtype);
|
||||
let json_field = schema_builder.add_json_field("json", FAST);
|
||||
let text_field_many_terms = schema_builder.add_text_field("text_many_terms", STRING | FAST);
|
||||
let text_field_few_terms = schema_builder.add_text_field("text_few_terms", STRING | FAST);
|
||||
let score_fieldtype = crate::schema::NumericOptions::default().set_fast();
|
||||
let score_field = schema_builder.add_u64_field("score", score_fieldtype.clone());
|
||||
let score_field_f64 = schema_builder.add_f64_field("score_f64", score_fieldtype.clone());
|
||||
let score_field_i64 = schema_builder.add_i64_field("score_i64", score_fieldtype);
|
||||
let index = Index::create_from_tempdir(schema_builder.build())?;
|
||||
let few_terms_data = ["INFO", "ERROR", "WARN", "DEBUG"];
|
||||
|
||||
let lg_norm = rand_distr::LogNormal::new(2.996f64, 0.979f64).unwrap();
|
||||
|
||||
let many_terms_data = (0..150_000)
|
||||
.map(|num| format!("author{}", num))
|
||||
.collect::<Vec<_>>();
|
||||
{
|
||||
let mut rng = StdRng::from_seed([1u8; 32]);
|
||||
let mut index_writer = index.writer_with_num_threads(1, 200_000_000)?;
|
||||
// To make the different test cases comparable we just change one doc to force the
|
||||
// cardinality
|
||||
if cardinality == Cardinality::Optional {
|
||||
index_writer.add_document(doc!())?;
|
||||
}
|
||||
if cardinality == Cardinality::Multivalued {
|
||||
index_writer.add_document(doc!(
|
||||
json_field => json!({"mixed_type": 10.0}),
|
||||
json_field => json!({"mixed_type": 10.0}),
|
||||
text_field => "cool",
|
||||
text_field => "cool",
|
||||
text_field_many_terms => "cool",
|
||||
text_field_many_terms => "cool",
|
||||
text_field_few_terms => "cool",
|
||||
text_field_few_terms => "cool",
|
||||
score_field => 1u64,
|
||||
score_field => 1u64,
|
||||
score_field_f64 => lg_norm.sample(&mut rng),
|
||||
score_field_f64 => lg_norm.sample(&mut rng),
|
||||
score_field_i64 => 1i64,
|
||||
score_field_i64 => 1i64,
|
||||
))?;
|
||||
}
|
||||
let mut doc_with_value = 1_000_000;
|
||||
if cardinality == Cardinality::Sparse {
|
||||
doc_with_value /= 20;
|
||||
}
|
||||
let _val_max = 1_000_000.0;
|
||||
for _ in 0..doc_with_value {
|
||||
let val: f64 = rng.gen_range(0.0..1_000_000.0);
|
||||
let json = if rng.gen_bool(0.1) {
|
||||
// 10% are numeric values
|
||||
json!({ "mixed_type": val })
|
||||
} else {
|
||||
json!({"mixed_type": many_terms_data.choose(&mut rng).unwrap().to_string()})
|
||||
};
|
||||
index_writer.add_document(doc!(
|
||||
text_field => "cool",
|
||||
json_field => json,
|
||||
text_field_many_terms => many_terms_data.choose(&mut rng).unwrap().to_string(),
|
||||
text_field_few_terms => few_terms_data.choose(&mut rng).unwrap().to_string(),
|
||||
score_field => val as u64,
|
||||
score_field_f64 => lg_norm.sample(&mut rng),
|
||||
score_field_i64 => val as i64,
|
||||
))?;
|
||||
if cardinality == Cardinality::Sparse {
|
||||
for _ in 0..20 {
|
||||
index_writer.add_document(doc!(text_field => "cool"))?;
|
||||
}
|
||||
}
|
||||
}
|
||||
// writing the segment
|
||||
index_writer.commit()?;
|
||||
}
|
||||
|
||||
Ok(index)
|
||||
}
|
||||
|
||||
use paste::paste;
|
||||
#[macro_export]
|
||||
macro_rules! bench_all_cardinalities {
|
||||
( $x:ident ) => {
|
||||
paste! {
|
||||
#[bench]
|
||||
fn $x(b: &mut Bencher) {
|
||||
[<$x _card>](b, Cardinality::Full)
|
||||
}
|
||||
|
||||
#[bench]
|
||||
fn [<$x _opt>](b: &mut Bencher) {
|
||||
[<$x _card>](b, Cardinality::Optional)
|
||||
}
|
||||
|
||||
#[bench]
|
||||
fn [<$x _multi>](b: &mut Bencher) {
|
||||
[<$x _card>](b, Cardinality::Multivalued)
|
||||
}
|
||||
|
||||
#[bench]
|
||||
fn [<$x _sparse>](b: &mut Bencher) {
|
||||
[<$x _card>](b, Cardinality::Sparse)
|
||||
}
|
||||
|
||||
}
|
||||
};
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_average_u64);
|
||||
|
||||
fn bench_aggregation_average_u64_card(b: &mut Bencher, cardinality: Cardinality) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
let text_field = reader.searcher().schema().get_field("text").unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let term_query = TermQuery::new(
|
||||
Term::from_field_text(text_field, "cool"),
|
||||
IndexRecordOption::Basic,
|
||||
);
|
||||
|
||||
let agg_req_1: Aggregations = serde_json::from_value(json!({
|
||||
"average": { "avg": { "field": "score", } }
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req_1);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&term_query, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_stats_f64);
|
||||
|
||||
fn bench_aggregation_stats_f64_card(b: &mut Bencher, cardinality: Cardinality) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
let text_field = reader.searcher().schema().get_field("text").unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let term_query = TermQuery::new(
|
||||
Term::from_field_text(text_field, "cool"),
|
||||
IndexRecordOption::Basic,
|
||||
);
|
||||
|
||||
let agg_req_1: Aggregations = serde_json::from_value(json!({
|
||||
"average_f64": { "stats": { "field": "score_f64", } }
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req_1);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&term_query, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_average_f64);
|
||||
|
||||
fn bench_aggregation_average_f64_card(b: &mut Bencher, cardinality: Cardinality) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
let text_field = reader.searcher().schema().get_field("text").unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let term_query = TermQuery::new(
|
||||
Term::from_field_text(text_field, "cool"),
|
||||
IndexRecordOption::Basic,
|
||||
);
|
||||
|
||||
let agg_req_1: Aggregations = serde_json::from_value(json!({
|
||||
"average_f64": { "avg": { "field": "score_f64", } }
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req_1);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&term_query, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_percentiles_f64);
|
||||
|
||||
fn bench_aggregation_percentiles_f64_card(b: &mut Bencher, cardinality: Cardinality) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let agg_req_str = r#"
|
||||
{
|
||||
"mypercentiles": {
|
||||
"percentiles": {
|
||||
"field": "score_f64",
|
||||
"percents": [ 95, 99, 99.9 ]
|
||||
}
|
||||
}
|
||||
} "#;
|
||||
let agg_req_1: Aggregations = serde_json::from_str(agg_req_str).unwrap();
|
||||
|
||||
let collector = get_collector(agg_req_1);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&AllQuery, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_average_u64_and_f64);
|
||||
|
||||
fn bench_aggregation_average_u64_and_f64_card(b: &mut Bencher, cardinality: Cardinality) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
let text_field = reader.searcher().schema().get_field("text").unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let term_query = TermQuery::new(
|
||||
Term::from_field_text(text_field, "cool"),
|
||||
IndexRecordOption::Basic,
|
||||
);
|
||||
|
||||
let agg_req_1: Aggregations = serde_json::from_value(json!({
|
||||
"average_f64": { "avg": { "field": "score_f64" } },
|
||||
"average": { "avg": { "field": "score" } },
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req_1);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&term_query, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_terms_few);
|
||||
|
||||
fn bench_aggregation_terms_few_card(b: &mut Bencher, cardinality: Cardinality) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let agg_req: Aggregations = serde_json::from_value(json!({
|
||||
"my_texts": { "terms": { "field": "text_few_terms" } },
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&AllQuery, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_terms_many_with_top_hits_agg);
|
||||
|
||||
fn bench_aggregation_terms_many_with_top_hits_agg_card(
|
||||
b: &mut Bencher,
|
||||
cardinality: Cardinality,
|
||||
) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let agg_req: Aggregations = serde_json::from_value(json!({
|
||||
"my_texts": {
|
||||
"terms": { "field": "text_many_terms" },
|
||||
"aggs": {
|
||||
"top_hits": { "top_hits":
|
||||
{
|
||||
"sort": [
|
||||
{ "score": "desc" }
|
||||
],
|
||||
"size": 2,
|
||||
"doc_value_fields": ["score_f64"]
|
||||
}
|
||||
}
|
||||
}
|
||||
},
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&AllQuery, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_terms_many_with_sub_agg);
|
||||
|
||||
fn bench_aggregation_terms_many_with_sub_agg_card(b: &mut Bencher, cardinality: Cardinality) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let agg_req: Aggregations = serde_json::from_value(json!({
|
||||
"my_texts": {
|
||||
"terms": { "field": "text_many_terms" },
|
||||
"aggs": {
|
||||
"average_f64": { "avg": { "field": "score_f64" } }
|
||||
}
|
||||
},
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&AllQuery, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_terms_many_json_mixed_type_with_sub_agg);
|
||||
|
||||
fn bench_aggregation_terms_many_json_mixed_type_with_sub_agg_card(
|
||||
b: &mut Bencher,
|
||||
cardinality: Cardinality,
|
||||
) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let agg_req: Aggregations = serde_json::from_value(json!({
|
||||
"my_texts": {
|
||||
"terms": { "field": "json.mixed_type" },
|
||||
"aggs": {
|
||||
"average_f64": { "avg": { "field": "score_f64" } }
|
||||
}
|
||||
},
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&AllQuery, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_terms_many2);
|
||||
|
||||
fn bench_aggregation_terms_many2_card(b: &mut Bencher, cardinality: Cardinality) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let agg_req: Aggregations = serde_json::from_value(json!({
|
||||
"my_texts": { "terms": { "field": "text_many_terms" } },
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&AllQuery, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_terms_many_order_by_term);
|
||||
|
||||
fn bench_aggregation_terms_many_order_by_term_card(b: &mut Bencher, cardinality: Cardinality) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let agg_req: Aggregations = serde_json::from_value(json!({
|
||||
"my_texts": { "terms": { "field": "text_many_terms", "order": { "_key": "desc" } } },
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&AllQuery, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_range_only);
|
||||
|
||||
fn bench_aggregation_range_only_card(b: &mut Bencher, cardinality: Cardinality) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let agg_req_1: Aggregations = serde_json::from_value(json!({
|
||||
"range_f64": { "range": { "field": "score_f64", "ranges": [
|
||||
{ "from": 3, "to": 7000 },
|
||||
{ "from": 7000, "to": 20000 },
|
||||
{ "from": 20000, "to": 30000 },
|
||||
{ "from": 30000, "to": 40000 },
|
||||
{ "from": 40000, "to": 50000 },
|
||||
{ "from": 50000, "to": 60000 }
|
||||
] } },
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req_1);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&AllQuery, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_range_with_avg);
|
||||
|
||||
fn bench_aggregation_range_with_avg_card(b: &mut Bencher, cardinality: Cardinality) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let agg_req_1: Aggregations = serde_json::from_value(json!({
|
||||
"rangef64": {
|
||||
"range": {
|
||||
"field": "score_f64",
|
||||
"ranges": [
|
||||
{ "from": 3, "to": 7000 },
|
||||
{ "from": 7000, "to": 20000 },
|
||||
{ "from": 20000, "to": 30000 },
|
||||
{ "from": 30000, "to": 40000 },
|
||||
{ "from": 40000, "to": 50000 },
|
||||
{ "from": 50000, "to": 60000 }
|
||||
]
|
||||
},
|
||||
"aggs": {
|
||||
"average_f64": { "avg": { "field": "score_f64" } }
|
||||
}
|
||||
},
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req_1);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&AllQuery, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
// hard bounds has a different algorithm, because it actually limits collection range
|
||||
//
|
||||
bench_all_cardinalities!(bench_aggregation_histogram_only_hard_bounds);
|
||||
|
||||
fn bench_aggregation_histogram_only_hard_bounds_card(
|
||||
b: &mut Bencher,
|
||||
cardinality: Cardinality,
|
||||
) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let agg_req_1: Aggregations = serde_json::from_value(json!({
|
||||
"rangef64": { "histogram": { "field": "score_f64", "interval": 100, "hard_bounds": { "min": 1000, "max": 300000 } } },
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req_1);
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&AllQuery, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_histogram_with_avg);
|
||||
|
||||
fn bench_aggregation_histogram_with_avg_card(b: &mut Bencher, cardinality: Cardinality) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let agg_req_1: Aggregations = serde_json::from_value(json!({
|
||||
"rangef64": {
|
||||
"histogram": { "field": "score_f64", "interval": 100 },
|
||||
"aggs": {
|
||||
"average_f64": { "avg": { "field": "score_f64" } }
|
||||
}
|
||||
}
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req_1);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&AllQuery, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_histogram_only);
|
||||
|
||||
fn bench_aggregation_histogram_only_card(b: &mut Bencher, cardinality: Cardinality) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let agg_req_1: Aggregations = serde_json::from_value(json!({
|
||||
"rangef64": {
|
||||
"histogram": {
|
||||
"field": "score_f64",
|
||||
"interval": 100 // 1000 buckets
|
||||
},
|
||||
}
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req_1);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&AllQuery, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
|
||||
bench_all_cardinalities!(bench_aggregation_avg_and_range_with_avg);
|
||||
|
||||
fn bench_aggregation_avg_and_range_with_avg_card(b: &mut Bencher, cardinality: Cardinality) {
|
||||
let index = get_test_index_bench(cardinality).unwrap();
|
||||
let reader = index.reader().unwrap();
|
||||
let text_field = reader.searcher().schema().get_field("text").unwrap();
|
||||
|
||||
b.iter(|| {
|
||||
let term_query = TermQuery::new(
|
||||
Term::from_field_text(text_field, "cool"),
|
||||
IndexRecordOption::Basic,
|
||||
);
|
||||
|
||||
let agg_req_1: Aggregations = serde_json::from_value(json!({
|
||||
"rangef64": {
|
||||
"range": {
|
||||
"field": "score_f64",
|
||||
"ranges": [
|
||||
{ "from": 3, "to": 7000 },
|
||||
{ "from": 7000, "to": 20000 },
|
||||
{ "from": 20000, "to": 60000 }
|
||||
]
|
||||
},
|
||||
"aggs": {
|
||||
"average_in_range": { "avg": { "field": "score" } }
|
||||
}
|
||||
},
|
||||
"average": { "avg": { "field": "score" } }
|
||||
}))
|
||||
.unwrap();
|
||||
|
||||
let collector = get_collector(agg_req_1);
|
||||
|
||||
let searcher = reader.searcher();
|
||||
searcher.search(&term_query, &collector).unwrap()
|
||||
});
|
||||
}
|
||||
}
|
||||
@@ -143,8 +143,6 @@ use std::fmt::Display;
|
||||
#[cfg(test)]
|
||||
mod agg_tests;
|
||||
|
||||
mod agg_bench;
|
||||
|
||||
use core::fmt;
|
||||
|
||||
pub use agg_limits::AggregationLimits;
|
||||
|
||||
Reference in New Issue
Block a user