Technology · Ebook

Cortex: Scalable Prometheus Metrics

by Shriira Press

Cortex is a horizontally scalable, highly available, multi-tenant, long-term storage system for Prometheus metrics — turning Prometheus from a single-server tool into a clustered metrics platform. This free book teaches it from the ground up: the Prometheus scaling problem and what Cortex is, the Prometheus data model and where it hits limits, Cortex's microservices architecture, the write path (remote write, distributors, ingesters), storage (blocks in object storage, the compactor), the read path (query-frontend, querier, store-gateway), multi-tenancy (isolation, per-tenant limits, the platform model), scalability and high availability (the hash ring, sharding, replication), alerting and rules and the ecosystem (Grafana, Thanos, Mimir), and operating Cortex in practice. Ten focused chapters with clear diagrams that demystify how a scalable, long-term, multi-tenant metrics platform works — Prometheus monitoring, delivered as an organization-wide service.

Contents

1. 1Preface
2. 2Chapter 1 — What Cortex Is
3. 3Chapter 2 — Prometheus and the Scaling Problem
4. 4Chapter 3 — Cortex's Architecture
5. 5Chapter 4 — The Write Path
6. 6Chapter 5 — Storage: Blocks and Object Storage
7. 7Chapter 6 — The Read Path
8. 8Chapter 7 — Multi-Tenancy
9. 9Chapter 8 — Scalability and High Availability
10. 10Chapter 9 — Alerting, Rules, and the Ecosystem
11. 11Chapter 10 — Operating Cortex in Practice