Skip to content

[serve] Prefer stopping pending replicas instead of running ones #12929

New issue
New issue
Closed
Closed
[serve] Prefer stopping pending replicas instead of running ones#12929
Labels
P2Important issue, but not time-criticalenhancementRequest for new feature and/or capabilityray-team-createdRay Team createdserveRay Serve Related Issue
Milestone
 
[serve] Autoscaling
@edoakes

Description

@edoakes
edoakes
opened on Dec 16, 2020

If a user scales to 100 and then down to 50 immediately, prefer to "dequeue" the pending replicas instead of tear down the running ones.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Assignees

No one assigned

    Labels

    P2Important issue, but not time-criticalenhancementRequest for new feature and/or capabilityray-team-createdRay Team createdserveRay Serve Related Issue

    Type

    No type

    Projects

    No projects

    Milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions