Unlocking Multi-Cloud Observability: The Case Study of...
summary
EO4EU builds a multi‑cloud observability platform across WEkEO DIAS and CINECA ADA Cloud using Prometheus + Thanos sidecar for metrics, Grafana for dashboards, OpenSearch + Fluent Bit for logs, Jaeger plus OpenTelemetry collectors for traces, and Rancher Fleet with Observer/observee labels and GitOps for automated, reproducible deployment. A three‑person core team moves from single‑cluster prototypes to a running system in about two months of design plus a week of build; the central Observer cluster is ~10 nodes (3 masters, workers ~8 cores/8GB), retains data long‑term with downsampling, and exposes multi‑tenancy and resource‑overhead tradeoffs.
AI-written summary · about
chapters (10)
- 0:00Multi-cloud observability intro
- 1:47Distributed platform challenges
- 3:13Scaling metrics with Prometheus/Thanos
- 5:39Logging standards and EFK stack
- 8:00Tracing with Jaeger and OpenTelemetry
- 9:25Integration challenges and design
- 11:32Open Observability Stack intro
- 12:56Automating with Fleet and GitOps
- 15:26Live demo of platform
- 19:08Q&A
related talks
25mOpen source observability for private cloud: mission impossible or not?
25mMulti-Cluster Observability: What you need to survive with a fleet
32mDesign Observability Solution for Multi-Tenancy Environment
31mCharting New Frontiers: The Evolution of Observability in Cloud Environments
40mBuilding a Scalable Observability Platform for Edge Compute Nodes
31mBest Practices for Effective Multi-Cluster Observability: Lessons Learned
31mCloud Native Observability
46mObservability Ecosystem in Kubernetes