Istio Tells me my Service has Slow Response Time, Now What?
summary
Istio telemetry feeds Prometheus and Grafana so Turbonomic builds a realtime model of apps, containers, nodes and clouds and then takes actions: Istio telemetry plus Kubernetes API lets it horizontal scale pods, change CPU/memory allocations, reschedule pods to other nodes or scale nodes themselves. Demo runs on Kubernetes 1.9 with Prometheus and shows automated provisioning of an extra instance and live pod moves that reduce response time, claiming continuous closed‑loop control across the full stack: Turbonomic closes the loop.
AI-written summary · about
chapters (16)
- 0:00Introduction
- 1:00Who is Istio
- 1:57What is Istio
- 3:28Telemetry
- 3:59Application Components
- 4:11What do I do
- 4:26Challenges
- 5:43EastWest Traffic
- 7:24Other Factors
- 8:16Resource Fragmentation
- 8:57kubernetes
- 9:43Autoscaling
- 10:13Continuously scaling
- 11:02Do I provision
- 11:51The complete stack
- 12:59Demo
related talks
25mThe Lean Observability Stack: Quick and Native Telemetry for Service...
49mComprehensive Container-Based Service Monitoring with Kubernetes and Istio
56mObservability in Action for your Service Mesh
24mObservability for Istio Service Mesh
25mUsing Istio to Build Better Service Level Objectives
24mIstio in Production: Expected Value, Results, and Effort
36mIstio: Real world scenarios
12mThe Road Ahead: Istio Observability