The DevOpsBeast Blog

Production engineering notes.

Field notes on Kubernetes, GPUs, Linux, and the rest of the production stack, from engineers who run real infrastructure.

RSS·subscribe in your feed reader

All (39)DevOpsBeast (2)GPU (1)GPU Cost Optimization (1)GPU Infrastructure (3)Kubernetes (1)Kubernetes Debugging (4)Kubernetes Networking (2)Kubernetes Operations (2)Kubernetes Performance (1)Kubernetes Security (2)Linux (2)LLM Infrastructure (7)Networking (2)Observability (1)Security (8)

Observability·Jun 23, 2026·10 min read

One Label Added Four Million Series to Your Prometheus. Here Is the Math.

A developer adds user_id to one counter to debug a support ticket. A week later Prometheus is eating 60 GB of RAM and queries time out. This is cardinality, the hidden cost center of metrics, and the math that predicts the disaster before it happens.

Read post