Configuring HorizontalPodAutoscaler Target Average Utilization Limits

You can configure the HorizontalPodAutoscaler (HPA) target average utilization limits in the FTM custom resource (CR). This field is optional. If you do not configure it, the default utilization limits are used.

The following table shows the parameters used to configure the HPA average utilization limits for the FTM pods.
Table 1. HPA average utilization configuration
Property name Description
hpa.averageUtilizationCPU Indicates the HPA’s target average CPU utilization.
hpa.averageUtilizationMemory Indicates the HPA’s target average memory utilization.

Finding HPA average utilization

You can check the average CPU and memory utilization settings of an HPA by using the following steps:
  1. To view the current HPA values, run the following command. Replace <hpa_name> with the actual HPA resource name.
    oc get hpa <hpa_name> -o yaml
  2. Locate the CPU and memory target values in the spec.metrics.resource list.
    spec:
  maxReplicas: 5
  metrics:
    - resource:
        name: cpu
        target:
          averageUtilization: 50
          type: Utilization
      type: Resource
    - resource:
        name: memory
        target:
          averageUtilization: 60
          type: Utilization
      type: Resource

The following sections show examples of how to set the HPA average CPU and memory utilization values for the FTM pods.

Example YAML 1 - Setting HPA for a specific pod

In this example, the business-rules-manager pod’s HPA average CPU utilization is set to 50%, and average memory utilization is set to 60%.
spec:
  resources:
    - name: business-rules-manager
      maxReplicas: 5
      hpa:
        averageUtilizationCPU: 50
        averageUtilizationMemory: 60

Example YAML 2 - Setting default HPA for all scalable pods

In this example, every horizontally scaling pod’s HPA average CPU utilization is set to 50%, and average memory utilization is set to 60% by using the resource name default 
spec:
  resources:
    - name: business-rules-manager
      maxReplicas: 5
    - name: default
      hpa:
        averageUtilizationCPU: 50
        averageUtilizationMemory: 60