1. Severity of the issue: (select one)
None: I’m just curious or want clarification.
I am looking at the configuration for spilling storage and i came across
type: ray_storage which is implemented via: ExternalStorageRayStorageImpl found in file: python\ray_private\external_storage.py
i was wondering what it represents as there is no documentation for this
Hi Shiranbi, this was an experimental API that is in the process of being deprecated. The idea was to configure external storage across the entire Ray cluster, but it was never carried through and I wouldn’t recommend using it.
@eoakes is there a way to disable spilling? and get an error if i am out of memory in the object store?
There is not currently a stable API to disable spilling but we’re in the process of adding one: [Core] Expose object_spilling_enabled Option to Public API by MengjinYan · Pull Request #51851 · ray-project/ray · GitHub