Supercharge Your Kubernetes Autoscaling with Custom Metrics

Learn how to extend Kubernetes' autoscaling capabilities beyond basic CPU and memory metrics in this conference talk from American Airlines engineers. Explore the implementation of custom metrics-based scaling using Horizontal Pod Autoscaler (HPA), enabling applications to scale dynamically based on business-specific indicators like queue depths and throughput volumes. Through live demonstrations, discover the process of deploying custom metrics providers, configuring HPA to utilize these metrics, and implementing intelligent scaling decisions. Gain practical insights into developing tailored scaling policies and see real-world examples of applications scaling automatically based on custom metrics such as queue length and requests per second.