Kai-Hsun Chen comments

Results 301 comments of


                                            Kai-Hsun Chen

trafficstars

[Core] Enable Scaling Down for Multi-Host TPU Replicas

@ryanaoleary could you also rebase your branch to fix the CI error? Thanks!

[Core] Enable Scaling Down for Multi-Host TPU Replicas

@can-anyscale could you retry the failed test? It is unrelated to this PR. Thanks!

[Core] Enable Scaling Down for Multi-Host TPU Replicas

The RLLib tests fail after retry, but I don't think that is related to this PR because this PR is only for KubeRay. cc @jjyao @can-anyscale

Fix for incorrect channel read behavior after accelerated DAG teardown

> This issue happened because the channel was unregistered with the mutable object manager on DAG teardown Where does `unregister` happen? Are you referring to [SetErrorInternal](https://github.com/ray-project/ray/blob/master/src/ray/core_worker/experimental_mutable_object_manager.cc#L409-L410)? > the core worker...

Fix for incorrect channel read behavior after accelerated DAG teardown

> After this PR is merged, can you add a check for WriterChannelRegistered()? I think this could be a good way to get more familiar with the C++ codebase. Thanks!...

[Docs][Kuberay] Documentation for Using Kuberay with TPUs

![image](https://github.com/user-attachments/assets/a34d1c06-beaa-48d1-a1e8-4f3c585527d9) @ryanaoleary can you fix the CI error?

[Docs][Kuberay] Documentation for Using Kuberay with TPUs

Thanks! I will request our doc team to review this PR.

[Docs][Kuberay] Documentation for Using Kuberay with TPUs

@ryanaoleary would you mind fixing the CI error? Thanks!

[Docs][Kuberay] Documentation for Using Kuberay with TPUs

@can-anyscale @angelinalg Do you have any idea why the build was canceled?

[core][experimental] Accelerated DAG should execute work on actor's main thread

system-level tasks => ex: close channel, resize channel