December 2019
Intermediate to advanced
494 pages
11h 41m
English
When autoscaling applications, the cloud computing platform itself has to have the ability to automatically increase or decrease the number of application instances based on the utilization and load.
In Kubernetes, the Horizontal Pod Autoscaler (HPA), a component of Kubernetes, is responsible for performing autoscale-related operations. It is implemented as a control loop that watches the utilization and increases or decreases the number for pods for an application instance.
You can define a threshold such as a percentage of CPU utilization to watch for and then increase/decrease the pods to a maximum/minimum number. Let's use the kubectl command to configure autoscaling for azure-vote-front deployment:
kubectl ...