Queue output items and reuse them. Remove the unneeded type system in sql and just use the Go type system. In best case this is more than an order of magnitude speedup: ``` BenchmarkSelectAll_1M-12 1 1841049400 ns/op 274299728 B/op 4198522 allocs/op BenchmarkSelectAll_1M-12 14 84833400 ns/op 169228346 B/op 3146541 allocs/op ```
197 lines
5 KiB
Go
197 lines
5 KiB
Go
/*
|
|
* MinIO Cloud Storage, (C) 2019 MinIO, Inc.
|
|
*
|
|
* Licensed 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.
|
|
*/
|
|
|
|
package s3select
|
|
|
|
import (
|
|
"bytes"
|
|
"encoding/csv"
|
|
"io"
|
|
"io/ioutil"
|
|
"math/rand"
|
|
"net/http"
|
|
"strconv"
|
|
"testing"
|
|
"time"
|
|
|
|
humanize "github.com/dustin/go-humanize"
|
|
)
|
|
|
|
var randSrc = rand.New(rand.NewSource(time.Now().UnixNano()))
|
|
|
|
const charset = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789"
|
|
|
|
func newRandString(length int) string {
|
|
b := make([]byte, length)
|
|
for i := range b {
|
|
b[i] = charset[randSrc.Intn(len(charset))]
|
|
}
|
|
return string(b)
|
|
}
|
|
|
|
func genSampleCSVData(count int) []byte {
|
|
buf := &bytes.Buffer{}
|
|
csvWriter := csv.NewWriter(buf)
|
|
csvWriter.Write([]string{"id", "name", "age", "city"})
|
|
|
|
for i := 0; i < count; i++ {
|
|
csvWriter.Write([]string{
|
|
strconv.Itoa(i),
|
|
newRandString(10),
|
|
newRandString(5),
|
|
newRandString(10),
|
|
})
|
|
}
|
|
|
|
csvWriter.Flush()
|
|
return buf.Bytes()
|
|
}
|
|
|
|
type nullResponseWriter struct {
|
|
}
|
|
|
|
func (w *nullResponseWriter) Header() http.Header {
|
|
return nil
|
|
}
|
|
|
|
func (w *nullResponseWriter) Write(p []byte) (int, error) {
|
|
return len(p), nil
|
|
}
|
|
|
|
func (w *nullResponseWriter) WriteHeader(statusCode int) {
|
|
}
|
|
|
|
func (w *nullResponseWriter) Flush() {
|
|
}
|
|
|
|
func benchmarkSelect(b *testing.B, count int, query string) {
|
|
var requestXML = []byte(`
|
|
<?xml version="1.0" encoding="UTF-8"?>
|
|
<SelectObjectContentRequest>
|
|
<Expression>` + query + `</Expression>
|
|
<ExpressionType>SQL</ExpressionType>
|
|
<InputSerialization>
|
|
<CompressionType>NONE</CompressionType>
|
|
<CSV>
|
|
<FileHeaderInfo>USE</FileHeaderInfo>
|
|
</CSV>
|
|
</InputSerialization>
|
|
<OutputSerialization>
|
|
<CSV>
|
|
</CSV>
|
|
</OutputSerialization>
|
|
<RequestProgress>
|
|
<Enabled>FALSE</Enabled>
|
|
</RequestProgress>
|
|
</SelectObjectContentRequest>
|
|
`)
|
|
|
|
csvData := genSampleCSVData(count)
|
|
|
|
b.ResetTimer()
|
|
b.ReportAllocs()
|
|
b.SetBytes(int64(count))
|
|
|
|
b.RunParallel(func(pb *testing.PB) {
|
|
for pb.Next() {
|
|
s3Select, err := NewS3Select(bytes.NewReader(requestXML))
|
|
if err != nil {
|
|
b.Fatal(err)
|
|
}
|
|
|
|
if err = s3Select.Open(func(offset, length int64) (io.ReadCloser, error) {
|
|
return ioutil.NopCloser(bytes.NewReader(csvData)), nil
|
|
}); err != nil {
|
|
b.Fatal(err)
|
|
}
|
|
|
|
s3Select.Evaluate(&nullResponseWriter{})
|
|
s3Select.Close()
|
|
}
|
|
})
|
|
}
|
|
|
|
func benchmarkSelectAll(b *testing.B, count int) {
|
|
benchmarkSelect(b, count, "select * from S3Object")
|
|
}
|
|
|
|
// BenchmarkSelectAll_100K - benchmark * function with 100k records.
|
|
func BenchmarkSelectAll_100K(b *testing.B) {
|
|
benchmarkSelectAll(b, 100*humanize.KiByte)
|
|
}
|
|
|
|
// BenchmarkSelectAll_1M - benchmark * function with 1m records.
|
|
func BenchmarkSelectAll_1M(b *testing.B) {
|
|
benchmarkSelectAll(b, 1*humanize.MiByte)
|
|
}
|
|
|
|
// BenchmarkSelectAll_2M - benchmark * function with 2m records.
|
|
func BenchmarkSelectAll_2M(b *testing.B) {
|
|
benchmarkSelectAll(b, 2*humanize.MiByte)
|
|
}
|
|
|
|
// BenchmarkSelectAll_10M - benchmark * function with 10m records.
|
|
func BenchmarkSelectAll_10M(b *testing.B) {
|
|
benchmarkSelectAll(b, 10*humanize.MiByte)
|
|
}
|
|
|
|
func benchmarkSingleCol(b *testing.B, count int) {
|
|
benchmarkSelect(b, count, "select id from S3Object")
|
|
}
|
|
|
|
// BenchmarkSingleRow_100K - benchmark SELECT column function with 100k records.
|
|
func BenchmarkSingleCol_100K(b *testing.B) {
|
|
benchmarkSingleCol(b, 1e5)
|
|
}
|
|
|
|
// BenchmarkSelectAll_1M - benchmark * function with 1m records.
|
|
func BenchmarkSingleCol_1M(b *testing.B) {
|
|
benchmarkSingleCol(b, 1e6)
|
|
}
|
|
|
|
// BenchmarkSelectAll_2M - benchmark * function with 2m records.
|
|
func BenchmarkSingleCol_2M(b *testing.B) {
|
|
benchmarkSingleCol(b, 2e6)
|
|
}
|
|
|
|
// BenchmarkSelectAll_10M - benchmark * function with 10m records.
|
|
func BenchmarkSingleCol_10M(b *testing.B) {
|
|
benchmarkSingleCol(b, 1e7)
|
|
}
|
|
|
|
func benchmarkAggregateCount(b *testing.B, count int) {
|
|
benchmarkSelect(b, count, "select count(*) from S3Object")
|
|
}
|
|
|
|
// BenchmarkAggregateCount_100K - benchmark count(*) function with 100k records.
|
|
func BenchmarkAggregateCount_100K(b *testing.B) {
|
|
benchmarkAggregateCount(b, 100*humanize.KiByte)
|
|
}
|
|
|
|
// BenchmarkAggregateCount_1M - benchmark count(*) function with 1m records.
|
|
func BenchmarkAggregateCount_1M(b *testing.B) {
|
|
benchmarkAggregateCount(b, 1*humanize.MiByte)
|
|
}
|
|
|
|
// BenchmarkAggregateCount_2M - benchmark count(*) function with 2m records.
|
|
func BenchmarkAggregateCount_2M(b *testing.B) {
|
|
benchmarkAggregateCount(b, 2*humanize.MiByte)
|
|
}
|
|
|
|
// BenchmarkAggregateCount_10M - benchmark count(*) function with 10m records.
|
|
func BenchmarkAggregateCount_10M(b *testing.B) {
|
|
benchmarkAggregateCount(b, 10*humanize.MiByte)
|
|
}
|