Diving into Event-Driven Architectures (EDA) with Python - Solving complexity at scale
summary
Event-driven architectures use asynchronous messages, brokers, and small typed payloads to decouple components so parts can be replaced and scaled independently; brokers (examples: Kafka, AMQP, MQTT, cloud SNS) must guarantee delivery and support once‑only processing and replay. Payloads use Avro/Protobuf/MessagePack or JSON, often compressed with Snappy; keep messages ≲1 packet to reach hundreds of thousands of messages/sec. Python has asyncio clients and high‑level libs like broadcaster but lacks strong async API tooling and schema/versioning ecosystems.
AI-written summary · about
chapters (20)
- 0:00Intro
- 0:45Motivation
- 1:08Idea
- 1:55REST vs GraphQL
- 3:17EventDriven Architecture
- 5:22Communication
- 7:29Encoding
- 9:07Topics
- 10:53Streaming
- 12:06Message Brokers
- 13:00Important Considerations
- 15:12API Standards
- 16:27Recap
- 17:14Example
- 19:40Python
- 20:03Async API
- 20:32PiPi
- 21:28Broadcaster
- 22:36Conclusion
- 23:01Questions
related talks
53mPatterns of Event Driven Architecture
44mBuilding Distributed Applications with Event-driven Architecture
51mComplex Event Flows in Distributed Systems
42mDesign Principles for an Event-Driven Architecture
46mManaging Event Driven Architectures
51mSimplifying Event-Driven Services with FastStream: A How-To Guide
31mThe Power of Event-Driven Systems without Burning your Hands or Budgets
38mEvent-driven applications: Apache Kafka and Python