Observability in Service Mesh Powered by Envoy and Apache SkyWalking
summary
Demonstrates a working Envoy→SkyWalking pipeline that builds telemetry without code injection or no Istio Mixer by using Envoy L7 metrics and HTTP access logs sent to a SkyWalking metric receiver. Delivers service topology, per-request details and traces while requiring consistent pod naming or a pilot to map IPs to services. Reports about 10–13% CPU overhead on sidecars and no latency impact observed in small-scale tests; tracing via Mixer is avoided because its adapters add complexity and extra hops.
AI-written summary · about
chapters (10)
- 0:00Start
- 1:39What is Service Mesh?
- 2:15Service mesh refresher
- 3:27Envoy architecture
- 4:43Envoy's data plane API
- 5:46Metrics Service
- 10:01Istio Mixer Performance Concerns
- 13:36Options
- 15:01ALS of Sidecar and Ingress
- 18:26Generate Topology Graph Edge Edge of topology map has two detected points, client side and server side
related talks
33mObservability in Service Mesh Powered by Envoy and Apache SkyWalking
40mThe journey of migration between Apache SkyWalking and Service Mesh
10mLightning Talk: Use SkyWalking Satellite to Process Tracing, Logging and Metrics in Serv...
24mObservability for Istio Service Mesh
33mObserving and Troubleshooting your Microservices with Istio
36mIstio: Real world scenarios
20mKeynote: Service Meshes and Observability
36mFine Grained Mesh Metrics for Better Visibility With Native Performance