Module 6: Production Operations - Monitoring with Prometheus & Grafana
Production Operations: Monitoring with Prometheus & Grafana
Duration: 60 minutes
You’ve deployed Edera. Your workloads are running in isolated microVMs with prevention-based security. Now comes the question every operator asks: “How do I know what’s actually happening?”
Traditional container monitoring gives you visibility into processes sharing a kernel. But Edera’s architecture is fundamentally different—you’re now running a hypervisor layer with isolated zones, each with its own kernel and resources. You need visibility at three distinct layers:
- Zone level - Individual microVM resource consumption and health
- Hypervisor level - Xen resource management and allocation
- Host level - Physical node (Dom0) metrics and capacity
Without proper observability, you’re flying blind. You can’t optimize resource allocation, troubleshoot performance issues, or prove compliance with SLAs.
What You’ll Learn
In this module, you’ll learn how to:
- Deploy a production-ready Prometheus and Grafana monitoring stack on Amazon EKS
- Configure automatic discovery of Edera metrics endpoints across your cluster
- Understand the three-tier metric hierarchy: zones, hypervisor, and host
- Navigate the 18-panel Grafana dashboard for complete observability
- Interpret CPU, memory, and health metrics to optimize your deployment
- Troubleshoot common issues using the monitoring stack
This is hands-on operational knowledge. By the end, you’ll have a complete monitoring solution running and know exactly what every metric means.
Ready to begin? Let’s start by setting up the monitoring stack →