The autoscaling of the services released on the platform both in cloud and on premise is entrusted to the resource of Kubernetes called Horizontal Pod Autoscaler. This resource uses the metrics collected by Kubernetes itself to evaluate the need for instantiation new pods to divide the work up to a maximum set.
The resources that are taken into consideration are the use of cpu and memory. If one of these values exceeds one threshold expressed as a percentage then the system will create a new instance of the pod to divide the load that will come deleted after a period of time when all available instances remain below the set threshold. This percentage is calculated with this expression:
The value of the desired metric for the service in question will be taken from the value set in the block
request contained in
resources of the
Deployment of Kubernetes.