https://github.com/gitbookio/bipbip
Benchmark runner for performance regression test suites
https://github.com/gitbookio/bipbip
benchmark node-js performance
Last synced: over 1 year ago
JSON representation
Benchmark runner for performance regression test suites
- Host: GitHub
- URL: https://github.com/gitbookio/bipbip
- Owner: GitbookIO
- Created: 2018-05-18T09:07:54.000Z (about 8 years ago)
- Default Branch: master
- Last Pushed: 2023-08-31T09:51:32.000Z (almost 3 years ago)
- Last Synced: 2025-04-13T22:49:35.973Z (over 1 year ago)
- Topics: benchmark, node-js, performance
- Language: JavaScript
- Homepage:
- Size: 635 KB
- Stars: 17
- Watchers: 3
- Forks: 5
- Open Issues: 12
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# `bipbip`
Easy to setup, benchmark tool for Node.js. This module is inspired by [jest](https://github.com/facebook/jest).
BipBip is intended to be run as a part of a performance regression test suite. It is intended to help answer questions like "have performance characteristics changed between releases" or "does this change have an impact on performance?"
# Installation
```
$ npm install -g bipbip
```
# Usage
Create a file `__benchmarks__/hello.js` ([example](__benchmarks__/fibonacci.js)):
```js
suite('Hello world', () => {
scenario('hello()', () => {
doSomething();
});
scenario('world()', () => {
doSomething();
})
});
```
then run:
```
$ bipbip
```
Results can be saved to a JSON file, to be compared later on:
```
$ bipbip --save ./results.json
$ bipbip --compare ./results.json
```
# Benchmarks API
`bipbip` defines some global variables in the executed JS files:
- `suite(name: string, fn: () => void)`: define a suite of scenarios
- `scenario(name: string, run: () => void)`: define a scenario
# CLI options
```
$ bipbip [options]
```
The command line accepts globs as arguments: `benchmark *.js`, `benchmark module1/*.js module2/*.js`
| Option | Description |
| ------ | ----------- |
| `-s, --save [file]` | Save the results of benchmarks |
| `-c, --compare [file]` | Compare the results to previously saved results |
| `-d, --duration [ms]` | Maximum duration of each scenario (default is 5sec) |
| `-e, --executions [count]` | Maximum executions per scenario (default is 1M) |
# Usage in a CI
When using `bipbip` in a CI service (like Travis), the results can be preserved in the [CI cache](https://docs.travis-ci.com/user/caching/#Fetching-and-storing-caches).
```yaml
jobs:
include:
- if: branch = master
script:
- bipbip --save .cache/benchmarks.json --compare .cache/benchmarks.json
- if: branch != master
script:
- bipbip --compare .cache/benchmarks.json
```
# Usage with babel
When running benchmarks on JS files not compiled for the current node version. You can run the benchmark CLI using `babel-node`:
```
$ babel-node node_modules/.bin/bipbip
```