majin1102

1. executing spark sql through ams dashboard is not a good choice, arctic shall connect kyuubi or trino in terminal module in future work 2. there should be a memory...

visual design for overview page:  including: - overview statistics：table/catalog count, space size, optimizing resources - optimizing brief：resource usage curve, optimizing table list - table space brief: space size curve,...

There's context for this. Amoro name transacions on snapshots which are not produced by amoro optimizing.... You could see transactions and optimized two tabs. Both of them consists all snapshots....

This feature would quite benefit for Spark because cache could be shared among tasks in a single executor(which would not previously), and tasks here are belonged to one task type....

> @shendanfengg I saw you trying to change the default value of `base.hive.refresh-interval` in order to refresh the data in Hive for all Mixed-Hive format tables every half an hour...

Glad to see this topic!

Thanks for feeding back, this seems a scheduing issue. Did you truncate data files between 18:09 and 24:09? or it happened before all these failures in the picture. It would...

Thanks so much for the further details. The log really narrows situations. Like you said, apparently the task has lost when optimizer completes the task(the tasks not encounter corrupted file)....

> WDYT If TaskNotFoundException occurs, that should mean the process lifecyle has been ended. That's the design and a incosistency problem apparently occurs here. I think we'd better figure out...

Hey, I checked the stack and found that the code here has been deeply refactored. I wonder if 0.5.1 has encountered similar inconsistency feed back? @zhoujinsong