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
{{ message }}
This repository has been archived by the owner on Dec 24, 2019. It is now read-only.
For larger clusters (100+ nodes), scaling down (say to 20 nodes) - takes an order of several hours due to a conservative nature of scale-down step in autoscaler. It would be very helpful to be able to scale down by more than 1 instance a the time, by providing a step-down configuration value expresses either in:
count: absolute number of instances to scale down by (for example by 2, 5, etc)
percentage: relative number of instance based on total (for example by 5%, 10%, etc)
The text was updated successfully, but these errors were encountered:
For larger clusters (100+ nodes), scaling down (say to 20 nodes) - takes an order of several hours due to a conservative nature of scale-down step in autoscaler. It would be very helpful to be able to scale down by more than 1 instance a the time, by providing a
step-down
configuration value expresses either in:The text was updated successfully, but these errors were encountered: