Influxdata
/
influxdb
mirror of https://github.com/influxdata/influxdb.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

fix: read_buffer benchmarks array type creation (#5401) 

Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
pull/24376/head
Andrew Lamb 2022-08-16 06:28:27 -04:00 committed by GitHub
parent 49ab568ca8
commit b60e0beee2
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 12 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
read_buffer/benches/database.rs
View File

@ -1,5 +1,6 @@
use arrow::{
array::{ArrayRef, Int64Array, StringArray},
array::{ArrayRef, DictionaryArray, TimestampNanosecondArray},
datatypes::Int32Type,
record_batch::RecordBatch,
};
use criterion::{criterion_group, criterion_main, BatchSize, Criterion};
@ -89,21 +90,27 @@ fn generate_row_group(rows: usize) -> RecordBatch {
let data: Vec<ArrayRef> = vec![
// sorted 2 cardinality column
Arc::new(StringArray::from(
Arc::new(DictionaryArray::<Int32Type>::from_iter(
(0..rows)
.into_iter()
.map(|i| if i < rows / 2 { "prod" } else { "dev" })
.map(|i| {
if i < rows / 2 {
Some("prod")
} else {
Some("dev")
}
})
.collect::<Vec<_>>(),
)),
// completely unique cardinality column
Arc::new(StringArray::from(
Arc::new(DictionaryArray::<Int32Type>::from_iter(
container_ids
.iter()
.map(|id| id.as_str())
.collect::<Vec<_>>(),
)),
// ms increasing time column;
Arc::new(Int64Array::from(
Arc::new(TimestampNanosecondArray::from(
(0..rows)
.into_iter()
.map(|i| BASE_TIME + (i as i64 * ONE_MS))