What the hell is service mesh anyway !?
summary
Explains what a service mesh actually wires into your stack and what it costs: deploys sidecar proxies managed by a control plane, implements mutual TLS, canary and blue‑green routing, and exposes metrics/traces via Prometheus, Grafana and Jaeger while running the Sock Shop demo. Demonstrates injecting 5s delays for chaos testing and live routing changes; initial demo crashes under load before finally working. Argues service mesh helps modernize legacy apps and debug microservices but adds latency, CPU/memory overhead, learning curve and potential architecture lock‑in.
AI-written summary · about
chapters (9)
- 0:00Start
- 7:39Distributed Systems are hard!
- 17:51Service Mesh Architecture
- 21:05Service Mesh Ecosystem
- 21:23Service Mesh features
- 21:56Resiliency
- 22:57Observability
- 36:56Use cases
- 40:28Service Mesh is not Free!
related talks
01h 03mJust what is a "service mesh", and if I get one will it make everything OK?
43mWTF Do We Need a Service Mesh?
10mWhat the heck is a Service Mesh ?
43mService Mesh and the future of networking
34mService Mesh for Developers
11mWelcome & Opening Remarks
56mService Meshes - Powering the next wave of microservice architectures
13mService Mesh is a pipeline’s best friend