Efficient Communication Between Python Microservices using LavinMQ
summary
LavinMQ is an open-source Crystal broker used by 84codes and supports AMQP 0-9-1 and MQTT; it writes messages to disk as memory-mapped segment files with an in‑memory queue index so writes avoid heavy fsyncs and the system sustains virtually unbounded queue lengths. Example migration replaces synchronous Slack→ChatGPT HTTP calls with producer/consumer queues so producers keep sending during consumer downtime, enabling hundreds or thousands of messages per second and straightforward horizontal scaling via more worker instances.
AI-written summary · about
chapters (9)
- 0:00Introduction and overview
- 2:34Microservices vs monoliths
- 5:46Synchronous vs asynchronous
- 9:00Introducing LavinMQ
- 14:24Message broker architecture
- 15:59Monolith example
- 19:02Microservice example
- 22:23Scaling with message brokers
- 23:31Q&A and resources
related talks
33mEfficient Communication Between Python Microservices using LavinMQ
41mForget about HTTP
01h 31mIt's the message, stupid: python & amqp
30mA love letter to messaging: Celebrating RabbitMQ's journey
46mReal-world Examples of Message Queue Usage: Insights from RabbitMQ
49mCommunication in a microservice architecture
31mLeveraging an instant messaging protocol to build a scalable cloud architecture
48mA Beginner's Guide to Inter-Service Messaging