Getting the most out of service mesh
summary
Service mesh claims to let teams offload traffic management (routing, retries, circuit breakers), security via mutual TLS, and observability (metrics plus distributed tracing), reducing app code changes in most cases. Practical caveats: tracing requires propagating headers, external services may need app changes, and added latency is single-digit milliseconds. Persistent client‑state and event‑driven systems like Kafka need special configuration. Success requires explicit coordination between dev and ops about app requirements and mesh configuration.
AI-written summary · about
chapters (4)
related talks
10mWhat the heck is a Service Mesh ?
20mService Mesh: A Hole in the Pocket?
11mWelcome & Opening Remarks
53mDo you need a Service Mesh?
20mKeynote: Service Meshes and Observability
23mPanel: The State Of Service Mesh
20mThe Unexplored World of OpenTelemetry and Service Mesh
33mPlatform Engineering meets Service Mesh: an SRE love story