The following description of using HPA+CA to achieve automatic scaling of workload and node linkage,...

CCE cloud container engine supports the creation of CA policies, scales capacity according to CPU/memory allocation ratio, and periodically scales nodes according to time

HPA expands according to the monitoring standard. and when the cluster resources are insufficient, the newly created pod will be in the Running state, CA All Running states are checked for POD According to the user-configured scaling policy, select the most suitable node pool to scale

Under normal circumstances, the two need to be used together, because HPA requires sufficient resources in the cluster to successfully expand. and when the cluster resources are not enough, the CA needs to expand the nodes so that the cluster has sufficient resources

When the cluster has a large amount of spare resources after HPA scale-in, the CA scale-in node needs to release resources so as not to cause waste