bench-chain

bench-chain

chainable benchmark recording - averages & graphs.

extension of benchmark.js

install

yarn add bench-chain
npm i bench-chain --save

const Bench = require('bench-chain')

documentation

tests

examples

basic

const Bench = require('bench-chain')

Bench
  // location to store benchmarks
  .init(__dirname, 'basic.json')
  // tag current benchmarks with, to mark what changes caused differences
  .tags('v1')
  // actual benchmarks
  .add('1 * 1', () => 1 * 1)
  .add('1 + 1', () => 1 + 1)
  .run()

async

const Bench = require('bench-chain')

const sleep = sleepDuration => new Promise(resolve => setTimeout(resolve, sleepDuration))

Bench
  .init().dir(__dirname).filename('asyncs.json').setup()
  .name('sleepy')
  .tags('v1,v2')

  // can also use .add, and then .runAsync()
  .addAsync('sleep1', async done => {
    await sleep(1000)
    done()
  })
  .addAsync('sleep2', async done => {
    await sleep(2000)
    done()
  })
  .run()

flags

using funwithflags

• --graph will show only the graph reporting, rather than run the benchmarks
• --run-times=10 will run the test 10 times

metadata

battery parsing when available

- will be used for comparing more benchmark results with averages - amperage (number) - currentCapacity (number) - percent (number) - charging (boolean) - temp (number)

• mem: operating system memory, nodejs memory
• num: operations a second from benchmarkjs hertz
• sampled: total runs samples from benchmarkjs
• variation: variation from benchmarkjs
• timesFor: microtime | performance.now times for beginning & end of each run
• now: Date.now for changes over time

graphs for trends and variation

grouped by tags

progress bars