Dominik Marciński

Not sure for how long we usually wait to spot potential issues, so not certain what to suggest here. As a side note, the preemption process itself has a few...

Thanks Kensei for the pointer. IIUC the plugin tries the brute force approach, which would impact performance a lot, so I'd think about something more performance oriented. Regarding the interest,...

@sanposhiho Is there also an issue describing the problem and considered solutions?

Don't we need to start consuming NNN on pod updates? I think they are currently ignored.

The views looks great! Thanks @helayoty @sanposhiho and @macsko > What I'd like to see in the kanban board: > - Prioritization between triaged tasks > - Easy way to...

Shouldn't we also revert any actions taken during pre-bind? Moreover, after scheduler restart we should discover non-complete bindings and either continue or revert the actions taken. There is an issue...

The feature was useful to support the initially proposed GPU devices attachment use cases, but assuming the plan is to have an approach that avoids rescheduling (https://github.com/kubernetes/kubernetes/issues/135473) we don't have...

> can the binding failure be pod specific? I am not clear why it will be a resourceslice's condition? Agree that in DRA there is no risk that scheduler allocation...

From what I see we have a few different attempts to bypass the problem of api-server latency overhead and lack of atomicity. Similar ones are [KEP-5055: DRA: admin-controlled attributes](https://github.com/kubernetes/enhancements/pull/5277) and...

Edited: Previously I skimmed the proposal too fast, but I still think we could allow defining the implicit rules similarly how it's proposed in [KEP-5055: DRA: admin-controlled attributes](https://github.com/kubernetes/enhancements/pull/5277) and define...