The path to self-healing: Re-architecting for massive scale on kubernetes
summary
A complex monolith is dismantled and re-architected into a Kubernetes microservices platform in 75 days, with a six‑phase plan: capacity planning, containerization, observability (Prometheus, Grafana, Loki), autoscaling (HPA/VPA and cluster node autoscaler), multi‑zone Kafka redundancy, CI/CD and devsecops pipelines, and GDPR/NIS2 data‑sovereignty controls. The platform ingests telemetry from 100+ device types, handles thousands of tickets daily, adds an agentic AI layer for realtime remediation, and cuts per‑device cost from $10–$12 to $2–$2.5, running on managed Kubernetes with multi‑cluster HA.
AI-written summary · about
chapters (8)
- 0:00Introduction and context
- 3:38Problem with monolithic architecture
- 7:09Migration to microservices plan
- 11:26Kubernetes migration execution
- 13:56Observability and scalability
- 16:44Security and compliance
- 17:27Outcomes and cost reduction
- 18:31Future with CO and Q&A
related talks
18mRe-Architecting Monoliths Into Kubernetes Microservices at Million-User Scale: Hard...
38mBuilding the Perfect Infrastructure with Kubernetes
40mOperating a Self-Healing Bare-Metal Kubernetes Platform at Global Scale
36mOptimizing Resource Usage in Kubernetes
28mOvercoming the Challenges of Centralizing Container and Kubernetes Operations
35mKubernetes journey
45mHow Kubernetes Was the Secret Sauce in Our Globally Distributed Team’s Transition to Microservices
37mRunning Kubernetes in a Manufacturing Line