Self-Healing Systems
summary
Claim: build microservices-based systems that self-heal by detecting faults, reacting automatically and predicting trouble from history. Concrete steps: five prerequisites including deploy anywhere in a cluster, zero‑downtime updates via blue‑green deployment, and relative scaling instead of absolute instance counts. Tools demonstrated and used in production include Mesos, Kubernetes, Docker Swarm, ZooKeeper/Consul/etcd as real-time registries, Elasticsearch as a historical registry, and Jenkins to orchestrate; the stack proves practical and reduces human intervention while enabling proactive scaling and automated recovery.
AI-written summary · about
chapters (4)
- 0:00Introduction and context
- 1:41The myth of perfection
- 4:02Principles of self-healing
- 28:18Tools and future outlook
related talks
26mSelf Healing Systems
14mBuilding Self-Healing Microservices for Modern SRE Practices
34mToward continuous resilience
14mPlatform Engineering with Self-Healing AWS Infra
16mEngineering Failure-Resilient Systems
28mThe Magic of Kubernetes Self-Healing Capabilities
31mThe Path to self-healing systems
36mEvent-based self-healing monitoring