Junfan Zhang
Junfan Zhang
Could you help review this? @advancedxy I want to go on to finish this feature.
> Can you elaborate on why the current block id scheme does not work for stage retry given all data of previous stage gets erased? Sorry for the late reply...
> I strongly object to moving back to the original unique task attempt id which limits the number of tasks or partitions, which renders Uniffle unusable in our production environment.我强烈反对回到原来的唯一任务尝试...
This looks a good try to make the AI features be scoped in the gravitino management. I'm not sure whether the features that used in realtime/offline batch/once-time inference should be...
Overall lgtm. I hope some test cases could be added to cover this case.
> Overall lgtm. I hope some test cases could be added to cover this case. Have any update on this? @maobaolong If you have applied in your internal cluster, please...
Nice feature! It will be better if we could add the new doc about develop to describe this feature.
I hope the heartbeat could not be invoked by other thread like the unrelated unregister operations. If you want the latest info, you can decrease the heartbeat interval of server...
> @zuston The affect of decrease interval can make it better but not a fundamental solution, the latest info could be lost also, and the frequency heartbeat could make Coordinator...