Idioms for Building Distributed Fault-tolerant Applications with Elixir
summary
Elixir running on the Erlang VM claims concurrency, distribution and fault tolerance as design consequences by using lightweight processes and supervisors, message passing, supervision trees and a let it crash mentality. Examples: WhatsApp reaches about 2 million connections on one node; mix, tasks, Ecto and the macro system are shown in demos. A tiny supervised task is killed and automatically restarted, remote shells spawn processes on other nodes, and live code replacement is done with load_binary to patch running modules.
AI-written summary · about
chapters (36)
- 0:00Intro
- 1:38History
- 3:28Functional Programming Languages
- 5:37Airlink
- 11:45Interrogatory
- 12:08Idioms
- 14:25Processes
- 16:21Process Examples
- 17:25Write Need
- 18:42Fail Fast
- 21:03Supervisors
- 22:27Concurrency
- 23:18Release Process
- 24:44Virtual Machine Properties
- 25:28Compatibility
- 26:05Accessibility
- 27:00Macros
- 28:26Unit testing
- 29:55ECTO
- 31:10Productivity
- 31:24Tools
- 32:06Package Manager
- 32:21Demo
- 32:57Mix
- 34:45Module
- 35:12Children
- 39:27Interactive Shell
- 41:59Getting Started Guide
- 42:50Elixir Conference
- 43:24Poker Game
- 44:26Books
- 44:48Thank you
- 45:12What happens when the supervisor acts
- 46:54What about debugging
- 48:52Logging statements
- 49:28Hot code loading
related talks
44mIdioms for Building Fault-tolerant Applications with Elixir
41mIdioms for building fault-tolerant applications with Elixir
37mElixir - functional, concurrent, distributed programming
03h 19mWorkshop - All Aboard The Elixir Express!
01h 06mWhat every Node.js developer needs to know about Elixir
04mInterview
01h 03mThinking Erlang in a connected world
53mElixir: a potion to mix complexity and concurrency in real‑time