$10K per Minute of Downtime: Architecture, Queues, Streaming and Fintech
summary
High-load fintech architecture focused on avoiding costly downtime by using queues and streaming to handle large real-time data volumes with low latency. Covers architectural patterns for microservices and event-driven systems that deliver scalability, fault tolerance, and consistency for financial applications.
AI-written summary · about
chapters (16)
- 0:00Solidgate intro
- 5:22Payment reliability
- 7:08RabbitMQ fails fintech
- 14:36RabbitMQ risks
- 16:05Kafka mental model
- 20:30Q&A
- 23:02Fault tolerance
- 25:32Debezium problems
- 29:02Transactional Outbox
- 36:06Order Streamer
- 39:38Event order solution
- 42:55WAL future
- 45:06CDC flexibility
- 49:05Scalability load
- 51:13Kafka rebalancing
- 54:00Summary
related talks
15mPlatform Engineering for FinTech
38mMaintaining highly distributed services with Erlang: 20y experiences in fintech
47mFault-Tolerant Architecture of a Bank's Front-End System
57mDSS Keynote — Back to the Future: Prepare Your Data Infrastructure for Anything
29mDelivering Always-On Financial Services with
40mCAPtivating architecture: Navigating Distributed Systems and Microservices
24mBuilding Planet-Scale Payment Platforms
10mBuilding High-Performance Financial Data Pipelines