Scale workloads on real demand — queues, streams, custom metrics — and all the way down to zero when idle.

Welcome to KEDA: Event-Driven Autoscaling for Kubernetes.

KEDA (Kubernetes Event-Driven Autoscaling) lets you scale workloads based on the signals that actually drive demand — queue length, stream lag, custom metrics, dozens of sources — and even scale to zero when there's no work. This free book teaches it from the ground up: Kubernetes autoscaling and the HPA's limits, KEDA's architecture and how it extends the HPA, the ScaledObject resource, scalers and the event sources they connect to, scale-to-zero in depth (and the cold-start trade-off), ScaledJobs for batch workloads, authenticating to event sources with TriggerAuthentication, advanced scaling behavior and tuning, observability, and operating KEDA in production. Ten focused chapters with real config and clear diagrams that turn autoscaling from a crude CPU heuristic into a precise, cost-efficient match between capacity and real demand.

This title is part of the ShriIra library and is free to read in full, right here — our small contribution to making world-class knowledge easy to reach.

A note on reading it: open the Contents menu at the top of the reader to jump between chapters, use the Aa menu to set a comfortable text size, theme (light, sepia, or night), and single- or two-page layout. Your place is saved automatically, so you can always pick up where you left off.

We hope it serves you well.

— Shriira Press

Contents

1. Chapter 1 — What KEDA Is
2. Chapter 2 — Kubernetes Autoscaling Fundamentals
3. Chapter 3 — Architecture
4. Chapter 4 — ScaledObjects
5. Chapter 5 — Scalers and Event Sources
6. Chapter 6 — Scale-to-Zero
7. Chapter 7 — ScaledJobs and Batch Workloads
8. Chapter 8 — Authenticating to Event Sources
9. Chapter 9 — Advanced Scaling Behavior
10. Chapter 10 — Operating KEDA and Putting It Together