You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Which component are you using?:
/area vertical-pod-autoscaler
Is your feature request designed to solve a problem? If so describe the problem this feature should solve.:
Currently, the VPA recommender provides CPU recommendations in raw values without rounding, making it difficult to interpret the recommendations.
Describe the solution you'd like.:
Implement a "round CPU" feature that provides CPU recommendations rounded to the nearest unit. For example, users should be able to specify a rounding factor (e.g., --rounding-factor-cpu=50) so that a recommendation like 79m gets rounded to 100m, and a recommendation like 33m could be rounded to 35m or 50m, depending on the rounding factor. The goal is to make the CPU values more user-friendly and easier to configure in Kubernetes specs.
Describe any alternative solutions you've considered.:
Additional context.:
/assign
The text was updated successfully, but these errors were encountered:
/retitle Feature Request: Implement Rounding for CPU Recommendations in VPA
k8s-ci-robot
changed the title
Feature Request: Add Humanize Feature for CPU Recommendations
Feature Request: Implement Rounding for CPU Recommendations in VPA
Jan 9, 2025
Which component are you using?:
/area vertical-pod-autoscaler
Is your feature request designed to solve a problem? If so describe the problem this feature should solve.:
Currently, the VPA recommender provides CPU recommendations in raw values without rounding, making it difficult to interpret the recommendations.
Describe the solution you'd like.:
Implement a "round CPU" feature that provides CPU recommendations rounded to the nearest unit. For example, users should be able to specify a rounding factor (e.g., --rounding-factor-cpu=50) so that a recommendation like 79m gets rounded to 100m, and a recommendation like 33m could be rounded to 35m or 50m, depending on the rounding factor. The goal is to make the CPU values more user-friendly and easier to configure in Kubernetes specs.
Describe any alternative solutions you've considered.:
Additional context.:
/assign
The text was updated successfully, but these errors were encountered: