Illya Chekrygin
Illya Chekrygin
> What happens if the old workload is preempted by another workload? Then we’ll end up with two Pending workloads. I’m wondering if it would be better to mark the...
Thank you, @mimowo, for the detailed review and insights. I’ve addressed and/or responded to all the comments in your review. Could you please take another look when you have a...
> I'm thinking about renaming the feature (and associated things like feature gate or KEP title) from DynamicallySizedJobs to ElasticJobs. It is shorter to pronounce and captures the very same...
> But the agreement was a conditional one since when we design the WokrloadSlice mechanism, we wanted to evaludate the approach only by scale down case for evaluation I suspect...
> When they enable the PartialAdmission, the scale-up result might be weird. I’m not 100% certain, but my expectation is that scale-up should reflect in the **minimum required pod count**...
> The Kueue supports PartialAdmission only for batch/v1 Job Great callout, @tenzen-y, thank you for raising this, you're absolutely right that this highlights a gap in the current KEP. After...
> could you mention that in the Beta graduation criteria? done.
Hello @mimowo @tenzen-y thank you for the review and feedback. I have reviewed all comments and updated KEP per provided feed back. Note: WorkloadSliceReplacement -> WorkloadSliceReplaced change was discussed in...