Kubernetes pod autoscaler using custom metrics
Blog post from Sysdig
The blog post by Mateo Burillo explains how to deploy a Kubernetes pod autoscaler using custom metrics provided by third-party monitoring services, such as Sysdig Monitor, to enhance dynamic scaling capabilities. It outlines the benefits of using the Kubernetes Horizontal Pod Autoscaler (HPA) for reliable and predictable scaling, and discusses the integration of custom metrics through the Kubernetes API using an extension service. The post also details the process of enabling the Kubernetes aggregation layer and creating APIService objects to facilitate the retrieval of custom metrics. Additionally, it highlights the use of tools like KEDA to simplify the configuration of custom metrics, allowing for the triggering of autoscalers with Prometheus and Sysdig metrics. The piece concludes by emphasizing the extensibility of Kubernetes and invites readers to engage with the project through various platforms and try out a Sysdig Monitor free trial.