Service Mess to Service Mesh
summary
Shows working demos of Linkerd and Istio: sidecar Envoy proxies provide mutual TLS, the reviews service migrates from v1→v2→v3 with 80/20 and 33/33/33 splits and a user‑specific canary (Jason) with zero downtime, plus circuit breakers and traffic shaping. Uses Prometheus/Grafana, Jaeger traces and Kiali topology to prove observability and diagnose a slow review service. Notes implementation cost as extra proxies/containers and recommends mesh when you need observability and fine‑grained control for multi‑tenant or sensitive workloads.
AI-written summary · about
chapters (22)
- 0:00intro
- 0:27preamble
- 4:02let's talk about service mesh
- 4:44observe
- 5:09control
- 5:51secure
- 6:33from monolithic to microservices
- 7:20from north-south to east-west
- 7:42the challenge with api gateways
- 8:41service mesh: how does it work?
- 11:32more than just a proxy
- 14:37dashboards
- 14:47prevent unexpected traffic patterns
- 15:09service meshes - examples
- 15:52linkerd
- 16:28istio
- 17:03demos
- 29:20in a nutshell
- 30:11service mesh implementation cost
- 30:59benefits of service mesh
- 31:28use a service mesh if
- 33:26questions?
related talks
52mService Mess to Service Mesh
26mAn Introduction to Service Mesh in Action
28mService Mesh: Demystifying Istio
58mUsing Service Meshes and Kubernetes to Solve Service-to-Service Communications...
50mService Mesh and Sidecars with Istio and Envoy
53mService Mesh: What, Why, And How?
52mEmpower Your Microservices with Istio Service Mesh
58mService Mesh and Sidecars with Istio and Envoy