Microservices: Awareness and Discovery Pattern on K8s
summary
Kubernetes native Services and Endpoints centralize service discovery and remove the need for client-side registries like Consul or Eureka by shifting load‑balancing out of application code. The talk covers using Services+Endpoints, CoreDNS and kube-proxy (noting IPVS limitations on some public clouds), and optional gateways/sidecars (NGINX, Envoy, Istio) for advanced routing; demos show listing instances via K8s primitives and applying round‑robin, least‑connections, hashing or never‑queue policies while relying on correct readiness/liveness probes.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 1:51Service discovery history
- 4:23Microservices origins
- 7:35Client-side limits
- 10:03K8s discovery
- 14:44K8s Services discovery
- 16:33Service types and load balancing
- 19:43Cloud load balancer limits
- 20:47CoreDNS and gateways
- 23:40Takeaways for K8s microservices
- 24:46Q&A
related talks
46mExperiences from using discovery services in a microservice landscape
49mDeveloping microservices applications on the Kubernetes platform
37mDistributed applications and Kubernetes: Frameworks or service meshes or both
28mRunning Microservices on Kubernetes – Best Practices
38mThe Untapped Power of Services - L7 Load Balancing Without a Service Mesh
51mHow to get a grip on your microservices system using a service mesh
28mDeploying hybrid topologies with k8s & Envoy: a look at service discovery
44mScaling Microservices with Kubernetes