Cloud Native resiliency patterns from the ground up
summary
Two small Quarkus microservices (activity + UI) demo application-level circuit breaker tuned to open at 75%, timeout 250 ms with retries every 100 ms and half-open retry after 1s; a fallback returns an empty activity. Infrastructure demos use istio destination rules, outlier detection, kiali, and scooper for cross-namespace failover and bucket4j token-bucket rate limiting. Result: without controls one scenario takes ~23.2s and yields 504s; with circuit breaking responses drop to ~2.47s and surface 503s, showing fail-fast recovery and measurable improvements.
AI-written summary · about
chapters (16)
- 0:00Introduction
- 1:42Who am I
- 3:13What is resilience
- 5:42Application level
- 13:04Sequence diagram
- 13:27Infrastructure patterns
- 14:43Locality load balancing and failover
- 15:52Application fallbacks
- 24:33Scooper
- 31:31Load Chatting
- 31:49Rate Limiting
- 33:30Request Priority
- 34:33Autoscaling
- 34:50Autoscaling with load chatting
- 36:41Thank you
- 36:54Questions
related talks
43mFault tolerance – look, it’s possible!
35mDesign patterns for application resiliency
38mResilience patterns in practice
50mSurviving Scale: Patterns for Evolving Failure-Resilient Microservices
01h 03mPatterns for Resilient Architecture
53mResiliency and Availability Design Patterns for the Cloud
24mCloud Native Hacks for the Real World: Solve your daily struggles with tech
18mDistributed Systems Resiliency 101: Circuit Breakers