Philly ETE 2023 — Lightning Talk: Getting Creative with HPA Scaling in Kubernetes — Drew Rogers
Abstract Scaling workloads have gotten faster and faster throughout the years, but what about the metrics that drive scaling? From virtual machines to pods in Kubernetes, the rules for scaling haven’t changed all that much. Your app’s CPU/Memory increases, scale up. What if we had a way to fetch any metrics we want from internal or external sources? What if instead of waiting for increased utilization to scale our app, we can predict it? Using custom and external metrics, we…