KCD Auckland (April 28, 2025)
Kubernetes Scaling Done Right: Probes, Autoscalers, and Resource Management
Scaling applications in Kubernetes is more than just adding pods—it’s about ensuring reliability, efficiency, and resilience. In this talk, I will cover key Kubernetes scaling mechanisms, including:
Liveness & Readiness Probes – Why they matter and how misconfiguration can break scaling.
Horizontal & Vertical Pod Autoscalers (HPA & VPA) – Choosing the right strategy for your workload.
Cluster Autoscaler – Scaling nodes dynamically to match pod demands.
Resource Requests & Limits – Avoiding performance bottlenecks and inefficiencies.
Observability & Metrics – Using Prometheus, KEDA, and custom metrics for smart scaling.
Common Pitfalls – Real-world examples of what happens when scaling is not done right.
I will also demonstrate a live example (or video) showing:
# The difference between properly configured vs. poorly configured health probes.
# How Kubernetes dynamically scales an application based on real traffic load.
This is more important because large-scale organizations do not properly use the autoscaling feature for their applications in k8s.
KCD NZ (Auckland)
Monday April 28th, 2025 - Auckland - 12:00pm to 5:30pm
Venue co-located event: Crowne Plaza Auckland Hotel
