Build Reliable Applications with Kubernetes and Istio Using Application Archetypes
summary
Introduces Application Archetypes and shows concrete Kubernetes + Istio implementations (GKE clusters, Global Load Balancer, Cloud DNS, Spanner/Cloud SQL) for zonal, regional, multi‑regional and global deployments, with math and measured SLOs. Demonstrates that splitting services across archetypes raises aggregate availability (example: single zonal → two zonal clusters + HA DB yields nearly two extra nines), explains fail‑ops, tradeoffs in cost/complexity, and that the patterns successfully produce predictable reliability gains.
AI-written summary · about
chapters (10)
- 0:00Reliable systems on cloud
- 1:05Pyramids of reliability
- 3:54Application archetypes
- 10:33Hot/cold/multi-zonal archetypes
- 13:01Active/passive region models
- 15:49Global high-availability archetype
- 18:12Reliability with SLIs/SLOs
- 22:26Kubernetes and Istio architectures
- 28:42Graceful degradation trade-offs
- 31:51Designing reliable systems
related talks
31mReliable multi-cluster application architectures with Istio
39mRunning Resilient Workloads with Istio
47mIstio on Kubernetes: Chaos, Canaries, Dark Launches
40mReimagining Application Networking and Security
54mEasy Microservices with Kubernetes and Istio
50mDeveloping Microservices with Istio
34mSpreading Apps, Controlling Traffic & Optimising Costs in Kubernetes
51mBuilding a Resilient Application Infrastructure on Kubernetes