Unlocking the Power of Kuma Service Mesh: Enhancing Reliability and D...
summary
Zomato adopts Kuma (Envoy-based) on AWS ECS to manage ~500 microservices and ~10,000 dataplanes, replacing brittle static Envoy configs. Control plane serves ~700–1,000 dataplanes each, Kuma enforces mTLS with built‑in CA and GitOps policy rollout, collects Envoy metrics into Victoria/Prometheus + Grafana, and enables canary traffic split (e.g., 5%). Engineering builds ephemeral preview environments and a local preview via SSH reverse tunnel that routes requests by a preview-id header to a developer’s process; system reduces debugging friction and scales reliably.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 1:47Kuma service mesh need at Zummetto
- 3:54Initial mesh setup challenges
- 6:28Why Kuma over others
- 7:13Kuma architecture overview
- 8:41Kuma multi-zone support
- 11:55Reliability with Kuma
- 16:16Developer productivity with Kuma
- 23:21Future plans & Q&A
related talks
31mTidy Up Microservices Connectivity with Apache Kafka® and Kuma
56mService Meshes - Powering the next wave of microservice architectures
56mHow to get a grip on your microservices system using a service-mesh
22mBuilding + operating a service mesh at a mid-size company
11mWelcome & Opening Remarks
39mGrow with Less Pains - Meshing From Monolith to Microservices
58mUsing Service Meshes and Kubernetes to Solve Service-to-Service Communications...
49mMaking Microservices Micro with Istio Service Mesh
