Micro-Runner

a CLI playground for benchmarking your JavaScript code

The main aim behind Micro-Runner is making microbenchmarking accessible and useful.
Often, web, mobile or desktop applications are suffering of poor performances and developers are struggling to analyze their applications performance and improve them. Micro-Runner aims to run some code in isolation like you would do in a playground, directly from your CLI, providing an immediate feedback when you want to compare the most performant implementation across multiple of them.

Installation

npm install -g micro-runner

In this way, Micro-Runner is always available inside your terminal!

Basic usage

micro-runner ./mybenchmark.js

Micro-Runner runs your test retrieving the benchmark of specific parts of your code displaying them in a table.
The best Micro-Runner usage is for comparing similar alghortims with different implementations.
Micro-Runner is a playground for learning how fast some code may run in isolation.

CLI arguments

micro-runner ./mybenchmark.js

Micro-Runner requires at least a file to run a benchmark on.
When you need to specify multiple file please use the folder argument as specified in this guide.

micro-runner -i 10 ./mybenchmark.js

-i corresponds to the number of iterations you wanna run per benchmark, it has to be an integer and can be a number from 1 to 100. When the parameter provided is larger than 100, the iterations per benchmark will be 100.
When this parameter is not specified the default value is 10.

If you have many tests and you set 100 iterations per benchmark be aware it could take a while.

micro-runner -f ./benchmarks

-f is used for specifying a folder with JavaScript files. Every single JavaScript file in that folder will run sequentially one after the other.

micro-runner -f ./benchmarks -m mjs

-m command followed by mjs is used when the files to benchmark are ES6 modules. By default all the modules will be run as CommonJS modules, you can also explicitely add the argument cjs to -m parameter.
ES6 modules are supported only from Node 14 onwards, Node 12 and 13 with experimental-modules flag will not work

micro-runner --verbose ./benchmark.js

-verbose command shows every iteration time for each benchmark

micro-runner -x ./results ./benchmark.js

-x command followed by a directory exports the Micro-Runner results to an Excel file

micro-runner -v

-v command shows the Micro-Runner version installed

Micro-Runner-Metrics

Micro-Runner-Metrics is the library used by Micro-Runner CLI for gathering and parsing metrics.

For more information on how to setup a micro-benchmark for micro-runner, please install micro-runner-metrics library and follow the guidelines.