Implementing distributed applications using ZeroMQ, Python and other bad guys...
summary
Demonstrates building distributed apps with ZeroMQ using Python bindings on a C++ core, showing working examples for request-reply, pub-sub, pipeline patterns and the poller/device primitives. Runs a 100-task ventilator with 3 workers to show near-linear speedup and compares one vs three-worker runtimes; shows clients can start before servers, IPC/TCP transports, and fair load‑balancing. Notes tradeoff: ultra-low latency by default, persistence and guaranteed delivery require extra broker/ack logic; C++ core with Python bindings is cross‑language and production‑proven.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:09Why ZeroMQ background
- 3:19Scalability vs complexity
- 6:53ZeroMQ architecture
- 18:21ZeroMQ patterns overview
- 21:14Request-reply demo
- 28:05Publisher-subscriber pattern demonstration
- 35:13Pipeline load balancing
- 40:31Advanced patterns
- 46:23Questions and answers session
related talks
40mUsing ZMQ for Fun and Profit
33mEfficient Communication Between Python Microservices using LavinMQ
29mBuilding production ready distributed task queue management system with celery
28mWorking with multiple hands: Parallel and Distributed Programming
17mFun, interesting and common RabbitMQ use cases
56mDistributed Messaging patterns
52mDIY Async Message Pump: Lessons from the trenches
16mTo queue or not to queue, that is the question.