Building High Resiliency into Digital Solutions
summary
Bank handles ~80 million customers with ~30M digital and ~24M mobile users so designs prioritize keeping critical balance and payment flows available during backend outages. Applications capture user actions as events and queue them in Kafka, process streams with Flink/Storm, and keep a local read model in MongoDB or Cassandra for edge caching. Resiliency hygiene uses precise timeouts, circuit breakers, safe‑mode toggles and global URLs, and validation runs controlled chaos engineering experiments (Gremlin) with expanding blast radius.
AI-written summary · about
chapters (6)
- 0:00Introduction and context
- 3:32Customer expectations vs bank systems
- 7:29Hygiene issues in resiliency
- 19:06Design principles for resiliency
- 27:20Measuring resiliency with chaos engineering
- 32:21Key takeaways and closing
related talks
18mEngineering Resilience at Scale; Lessons from a Distributed Systems Transformation
16mResilient by Design: Architectures for Crisis Response & Recovery
32mHow does your application behaves when everything goes wrong?
47mFault-Tolerant Architecture of a Bank's Front-End System
10mDesigning for Failure using AWS: Resilient Systems
32mSurviving Black Friday - A resilience engineering tale
33mResilient
26mThe tools and techniques behind resilient systems