Andrew Sy Kim

For reference some work has been done in this area already but it primarily focuses on memory scalability https://docs.ray.io/en/latest/cluster/kubernetes/benchmarks/memory-scalability-benchmark.html#kuberay-mem-scalability

This week is Google Cloud NEXT, but @kevin85421, @morhidi and I plan to meet some time next week to kick off this work. If you have any ideas or feedback...

> We can add only ReadyWorkerReplicas int32 in this PR, and remove Ready and add a new field in HeadInfo in the follow-up PRs. Is it safe to remove the...

> I suspect that the RayCluster status is rarely used by users because the ClusterState is not well-defined, yet almost no users have complained about it. I agree that it's...

> This PR changes the definition of ready, which is also backward incompatible. Some automation processes may pass, but others may fail and be hard to debug because this change...

No objections to make quay the default and to stop publishing nighlty images through CI. But do we really have to remove dockerhub support entirely? Can we still publish to...

cc @robscott @aojea

Good discussion on retry policy API for RayJob in https://github.com/ray-project/kuberay/pull/2091#discussion_r1573470644

/assign Igor asked me to take over the KEP update since he's on vacation

> KEP readme using the [latest template](https://github.com/kubernetes/enhancements/tree/master/keps/NNNN-kep-template) has been merged into the k/enhancements repo. @aramase the KEP template is up to date as far as I know, is there a...