Open source observability for private cloud: mission impossible or not?
summary
Company runs hybrid private/public clouds (Swisscom, OVH) with a five‑person SRE team and legacy apps, using Prometheus + Grafana for metrics, Filebeat → Elastic Stack for logs, ArgoCD/Helm for deployments, and a commercial APM that is costly and vendor‑locked. A distributed‑tracing POC pairs Jaeger operator with OpenTelemetry SDKs but stalls: docs and profiling gaps force shelving for production. Observability stays fragmented, alerting stays deliberately small (10 alerting rules); goal is a single pane but migration needs engineering time, capacity planning and cost analysis.
AI-written summary · about
chapters (10)
- 0:00Introduction and thanks
- 1:04About EQS Group profile
- 4:40Private cloud and its use
- 7:08Observability challenges
- 10:17Current observability architecture
- 14:14Current challenges and APM migration
- 16:24OpenTelemetry POC feedback
- 19:35Future plans and open source strategy
- 22:43Summary and call for feedback
- 23:49Q&A and closing
related talks
28mOpen Source for Better Observability
42mOpen Source for Better Observability
28mUnlocking Multi-Cloud Observability: The Case Study of...
01h 02mDesigning an open-source observability stack for Kubernetes
12mOpen-source observability is the way forward, and here’s why
56mFull-stack observability and application monitoring with
23mObservability in OpenStack Learnings and building blocks from the SIG Monitoring at SCS
45mOpen Source for Better Observability