How Our Critical Systems Fail, and What We Do About It

Explore a conference talk on ensuring critical systems in the health sector remain stable during changes. Learn about deploying updates using Kubernetes, processing unstructured data, and leveraging Kafka for real-time assessment. Discover how to implement a system that catches subtle deviations in organic data without requiring rollbacks. Gain insights into messaging, traffic management, data normalization, automation, distributed systems, and testing in production environments. Understand the challenges of handling increased traffic and growing systems in mission-critical healthcare applications.

Intro
Welcome
About us
What are we
How we got started
Messaging
Message flow
What was the problem
The problem
The new standard
How the traffic works
Creating a validator
The results
Certification track
Increased traffic
Everything grew
Architecture
Data normalization
Fixing the problem
Automation
Database
Distributed systems
Apache Kafka
Data flow
Testing in production
Testing in API world
Replaying input
Mirror testing
Conclusion
Questions