Building a Multi-tenant Monitoring System with Grafana, Prometheus and Cortex
summary
A practical setup for multi-tenant metrics monitoring using Prometheus for scraping, Prometheus remote_write to Cortex with object-bucket long-term storage, and Grafana providing per-organization dashboards. Includes a tenant-labeling sidecar that tags series so metrics are routed to the correct tenant, operational choices like caching layers to reduce object-store operations, Helm charts for Kubernetes deployment, and sizing notes from a production run (~2 million time series across 23 tenants) showing scalable, cost-controlled performance.
AI-written summary · about
chapters (11)
- 0:00Start
- 3:53Scenario
- 5:31Prometheus
- 7:29Multitenancy
- 10:01Architecture - Micro
- 17:21Grafana Configuration
- 21:03Dashboards
- 23:20Monitoring Cortex with Cortex
- 25:19Resources
- 26:22Cortex Deployment
- 28:45Conclusion
related talks
32mDesign Observability Solution for Multi-Tenancy Environment
20mIntro to Scaling Prometheus with Cortex
24mBanking Observability at Scale
37mIntro: Cortex
35mUnlocking Kubernetes Observability: Secure, Tenant-Cen...
24mPractical Kubernetes Monitoring with Prometheus
32m“Gluing it all together” aka building a platform around Prometheus
36mObservability at Scale With Monitoring as Code...