https://github.com/ulule/makroud-benchmarks

Benchmarks for makroud
https://github.com/ulule/makroud-benchmarks

Last synced: 11 months ago
JSON representation

Benchmarks for makroud

• Host: GitHub
• URL: https://github.com/ulule/makroud-benchmarks
• Owner: ulule
• License: mit
• Created: 2019-03-03T12:59:41.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2020-06-29T12:20:42.000Z (about 6 years ago)
• Last Synced: 2025-08-15T21:56:25.113Z (12 months ago)
• Language: Go
• Size: 252 KB
• Stars: 0
• Watchers: 8
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Makroud benchmark

> **NOTE:** A benchmark is always an observation, not a measurement of performance.

## Requirements

At least **Go 1.14.0**.

## Setup

First of all, install latest version of others library:

```bash

go get -u -v "github.com/go-gorp/gorp"

go get -u -v "github.com/go-xorm/xorm"

go get -u -v "github.com/jinzhu/gorm"

go get -u -v "github.com/jmoiron/sqlx"

```

> **NOTE:** This benchmark doesn't include **[SQLBoiler](https://github.com/volatiletech/sqlboiler)**

> and **[Kallax](https://github.com/src-d/go-kallax)** since they rely on code generation and not reflection.

## Execute

```bash

go test -run=XXX -bench=SelectAll -benchmem -benchtime=10s

go test -run=XXX -bench=SelectSubset -benchmem -benchtime=10s

go test -run=XXX -bench=SelectComplex -benchmem -benchtime=10s

go test -run=XXX -bench=Insert -benchmem -benchtime=10s

go test -run=XXX -bench=Update -benchmem -benchtime=10s

go test -run=XXX -bench=Delete -benchmem -benchtime=10s

```

## Benchmark graph

```bash

cd graph && python3 graph.py

```

![SelectAll NsOp](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/select_all_nsop.png)

![SelectAll Bop](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/select_all_bop.png)

![SelectAll Aop](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/select_all_aop.png)

![SelectSubset NsOp](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/select_subset_nsop.png)

![SelectSubset Bop](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/select_subset_bop.png)

![SelectSubset Aop](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/select_subset_aop.png)

![SelectComplex NsOp](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/select_complex_nsop.png)

![SelectComplex Bop](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/select_complex_bop.png)

![SelectComplex Aop](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/select_complex_aop.png)

![Insert NsOp](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/insert_nsop.png)

![Insert Bop](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/insert_bop.png)

![Insert Aop](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/insert_aop.png)

![Update NsOp](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/update_nsop.png)

![Update Bop](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/update_bop.png)

![Update Aop](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/update_aop.png)

![Delete NsOp](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/delete_nsop.png)

![Delete Bop](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/delete_bop.png)

![Delete Aop](https://raw.githubusercontent.com/ulule/makroud-benchmarks/master/graph/images/delete_aop.png)

## Acknowledgement

**SQLX** queries in this benchmark could be optimized.

However, since **Makroud** is built on top of **SQLX**, this benchmark highlight the difference of

performance using the same instructions.

**In every case**, using **SQLX** with a optimized workflow is more efficient than **Makroud**.

## Test Machine

```

OS:     Archlinux x86_64 Linux-4.18.8

CPU:    Intel(R) Core(TM) i7-7500U CPU @ 2.70GHz

Memory: 16GB

Go:     go version go1.11.2 linux/amd64

```

## Credits

- [SQLBoiler Benchmark repository](https://github.com/volatiletech/boilbench)