Set resource limits for Dynatrace Operator components

For resource limits, we provide what we think are sensible defaults. Performance tests with environments of up to 25 nodes (e2-standard-32 nodetype on Google Kubernetes Engine), 20 DynaKubes, 2,500 namespaces, and 5,000 pods with default limits performed well.

We identified five different key indicators influencing resource consumption and how each component is impacted:

Indicator

Dynatrace Operator

Webhook

CSI Driver

Namespaces

Nodes

DynaKubes

Pods

Number of OneAgent versions

Customize resource limits

While the default resource limits should be sufficient for most use cases, you can customize them based on your specific needs.

Dynatrace Operator

operator:
  requests:
    cpu: 50m
    memory: 64Mi
  limits:
    cpu: 100m
    memory: 128Mi

WebHook

webhook:
  requests:
    cpu: 300m
    memory: 128Mi
  limits:
    cpu: 300m
    memory: 128Mi

CSI Driver

csidriver:
  csiInit:
    resources:
      requests:
        cpu: 50m
        memory: 100Mi
      limits:
        cpu: 50m
        memory: 100Mi
  server:
    resources:
      requests:
        cpu: 50m
        memory: 100Mi
      limits:
        cpu: 50m
        memory: 100Mi
  provisioner:
    resources:
      requests:
        cpu: 300m
        memory: 100Mi
      limits:
        cpu: 300m
        memory: 100Mi
  registrar:
    resources:
      requests:
        cpu: 20m
        memory: 30Mi
      limits:
        cpu: 20m
        memory: 30Mi
  livenessprobe:
    resources:
      requests:
        cpu: 20m
        memory: 30Mi
      limits:
        cpu: 20m
        memory: 30Mi