Summary
In this module, you learned how to:
- Manually scale workload replicas or AKS node pools
- Use the horizontal pod autoscaler (HPA) to scale pod replicas based on metrics such as CPU utilization
- Use the cluster autoscaler to adjust node pool size when pods, including replicas created by HPA, can't be scheduled within existing capacity
- Integrate with Azure Container Instances (ACI) through virtual nodes for burst capacity
You also saw where Kubernetes Event-driven Autoscaling (KEDA) and node auto-provisioning (NAP) fit as related options.
Learn more
To go deeper on the topics covered in this module, see the following Microsoft Learn articles:
- Scaling options for applications in AKS
- Manually scale the node count in an AKS cluster
- Scale node pools in AKS
- Autoscale pods in AKS with the horizontal pod autoscaler
- Cluster autoscaling in AKS overview
- Use the cluster autoscaler in AKS
- Virtual nodes on Azure Container Instances
- Create and configure an AKS cluster to use virtual nodes
- Kubernetes Event-driven Autoscaling (KEDA) overview
- Node auto-provisioning (NAP) overview
If you don't have an Azure subscription yet, you can create one with the following options: