Github user cloud-fan commented on the issue:
    In theory, this should be done in a cost-based style. Changing the way how 
union combines data will reduce the parallelism.
    For example, if we union 2 tables each has 5 partitions. Without this PR we 
will launch 10 tasks to process the data, and locality should be easy to 
satisfy. After this PR, we only launch 5 tasks, and locality is hard to meet, 
we may have extra data transfer.
    We should move statistics to physical plan first. cc @wzhfy 


To unsubscribe, e-mail:
For additional commands, e-mail:

Reply via email to