[Feature] [API Server] Allow autoscaling in python api server client

ray-project / kuberay

A toolkit to run Ray applications on Kubernetes

Apache License 2.0

990 stars 330 forks source link

[Feature] [API Server] Allow autoscaling in python api server client #2029

Open smit-kiri opened 3 months ago

smit-kiri commented 3 months ago

Search before asking

[X] I had searched in the issues and found no similar feature requirement.

Description

I see that API Server does support autoscaling, but that is unavailable in python api server client.

It would be very useful to add this capability

Use case

No response

Related issues

No response

Are you willing to submit a PR?

[ ] Yes I am willing to submit a PR!

kevin85421 commented 3 months ago

cc @tedhtchang @blublinsky @z103cb

blublinsky commented 3 months ago

Oh yes, Cluster spec is missing:

    "enableInTreeAutoscaling": true,
    "autoscalerOptions": {
        "upscalingMode": "Default",
        "idleTimeoutSeconds": 30,
        "cpu": "500m",
        "memory": "512Mi"
    },

parameters. Can get to it in a couple of weeks

anyscalesam commented 3 months ago

OK to assign to you @blublinsky to track next step?

blublinsky commented 3 months ago

Sure, but I need my current PR to be merged first. It has been sitting there for several months now