The Hare Wins the Race: Getting the most out of RabbitMQ in distributed applications
summary
Overview of concrete takeaways and demos: RabbitMQ basics with direct/topic/fanout routing, AMQP libraries shown (pika, kombu, librabbitmq), and Celery’s default single-queue problem with a case study that routes tasks into two queues via a custom router and a topic exchange to keep short queries responsive while fitting neural nets on heavier workers. Live code is on GitHub and slides are linked, the management UI is reachable on RabbitMQ management UI port 15672, and practical gotchas include unrouted messages fixed by an alternate exchange and the inequivalent arg error when redeclaring immutable objects.
AI-written summary · about
chapters (24)
- 0:00Intro
- 0:49predata
- 2:09Message Brokers
- 2:49RabbitMQ Architecture
- 4:54Bindings
- 5:18Exchange Types
- 5:36Direct Exchanges
- 6:27Topic Exchanges
- 9:01Fanout Exchanges
- 10:38Python Libraries
- 11:45Code Example: Pika
- 12:17Code Example: Producer
- 12:42Code Example: Consumer
- 15:19Celery Tasks
- 16:17Celery Defaults
- 16:56Case Study
- 17:53Initial Configuration
- 18:14Sample Tasks
- 19:50Celery Router Functions
- 21:36New Configuration
- 23:08Launching Workers
- 24:55Alternate Exchanges
- 27:17Managing Complexity
- 28:09Outline
related talks
02h 59mPractical RabbitMq workshop
39mReal World RabbitMQ: Exploring Practical Applications
41mDecouple all the things: Asynchronous messaging keeps it simple
30mA love letter to messaging: Celebrating RabbitMQ's journey
32mRabbitMQ in Real Life - Tips and Tricks
47mRabbitMQ Hands On
03mBuilding a Distributed Data Ingestion System with RabbitMQ
48mBuilding a Distributed Data Ingestion System with RabbitMQ