victor-homyakov

Приёмы оптимизации кода по скорости

Организационные

• Культура разработки performance-first: https://tonsky.me/blog/performance-first/
• Бюджет скорости: https://wp-rocket.me/blog/performance-budgets/
• Performance mantra: http://www.brendangregg.com/blog/2018-06-30/benchmarking-checklist.html

Не зависящие от языка и его реализации

• Rosetta Code - реализации задач на разных языках: http://rosettacode.org/wiki/Collections
• Статьи по основным структурам данных:

matt-peters

"""
An example of running both pytorch and tensorflow in the same network,
while pasing weights and gradients between the two.

In this example, we run a simple 2-layer feed-forward network,
with the first layer size (5, 2) and the second (2, 3).
The code contains an implementation of forward/backward passes with
three versions:
    * tensorflow only
    * pytorch only

dannguyen

This is a guide that basically combines protolium's very helpful ffmpeg cheatsheet with the spleeter library.

Here's a tweet thread that shows a video snippet, with separate bass, vocals, and drums track:

https://twitter.com/dancow/status/1191068710968209408

Installing stuff

fgilio

/*
 * Handling Errors using async/await
 * Has to be used inside an async function
 */
try {
    const response = await axios.get('https://your.site/api/v1/bla/ble/bli');
    // Success 🎉
    console.log(response);
} catch (error) {
    // Error 😨

branneman

// :: (String, String) => String
const spawn = require('child_process').spawnSync;

// :: String => [String]
const getRules = raw => raw
    .split('\n')
    .map(line => line.trim())
    .filter(line => !!line)
    .filter(line => line[0] !== '/' && line[0] !== '✖')
    .map(line => line.match(/[a-z-]+$/)[0]);

jjmu15

// Determine if an element is in the visible viewport
function isInViewport(element) {
  var rect = element.getBoundingClientRect();
  var html = document.documentElement;
  return (
    rect.top >= 0 &&
    rect.left >= 0 &&
    rect.bottom <= (window.innerHeight || html.clientHeight) &&
    rect.right <= (window.innerWidth || html.clientWidth)
  );

denji

Moved to git repository: https://github.com/denji/nginx-tuning

NGINX Tuning For Best Performance

For this configuration you can use web server you like, i decided, because i work mostly with it to use nginx.

Generally, properly configured nginx can handle up to 400K to 500K requests per second (clustered), most what i saw is 50K to 80K (non-clustered) requests per second and 30% CPU load, course, this was 2 x Intel Xeon with HyperThreading enabled, but it can work without problem on slower machines.

You must understand that this config is used in testing environment and not in production so you will need to find a way to implement most of those features best possible for your servers.

kiesia

default['nginx']['version'] = "1.2.0"
default['nginx']['passenger']['version'] = "3.0.12"

predakanga

Requirements

For libcodegen

• Boost >= 1.35
• zlib

Additional requirements for the codegen binary

• TagLib