VPA Recommendations
Vertical Pod Autoscalar Recommendations vs. Utilization vs. Requests & Limits
What’s it all about?
A dashboard to visualize metrics about the Vertical Pod Autoscalar (https://github.com/kubernetes/autoscaler/tree/master/vertical-pod-autoscaler) using the generated Prometheus metrics of the Kube-State-Metrics (https://github.com/kubernetes/kube-state-metrics/blob/master/docs/verticalpodautoscaler-metrics.md)
What do I need as a perquisite?
- Prometheus installed => ultimately using the kube-prometheus stack (https://github.com/prometheus-community/helm-charts/tree/main/charts/kube-prometheus-stack)
- The kube-state-metrics must be configured to enable Vertical Pod Autoscalar metrics. Example for the Helm Chart above:
kube-state-metrics:
namespaceOverride: ""
rbac:
create: true
...
collectors:
verticalpodautoscalers: true
- VPA Installed and at least one
VerticalPodAutoscalar
CRD is configured against a certain target (check the documentation of the VPA for more info) - Optionally: VPA is using Prometheus as a historical query backend to monitor changes over time.
My installation & configurations for the VPA
I am using the Helm chart at https://github.com/FairwindsOps/charts/tree/master/stable/vpa for the installing the VPA. There are other ways to do so. Just thought of sharing my values.yml
for customizing this chart as the Prometheus part wasn’t that trivial. You can check my GitHub Gist here https://gist.github.com/sherifkayad/1b4e4df408e1be357168a38e1980b9a5
Dashboard revisions
Upload an updated version of an exported dashboard.json file from Grafana
Revision | Decscription | Created | |
---|---|---|---|
Download |
Sign up for Grafana Cloud
Get this dashboard
Data source:
Dependencies: