zhengruifeng opened a new pull request #33298:
URL: https://github.com/apache/spark/pull/33298


   ### What changes were proposed in this pull request?
   1, introduce `ShuffleExecAccumulator` in `ShuffleExchangeExec` to support 
arbitrary statistics;
   
   2, impl a key sampling `ShuffleExecAccumulator` to detect skew keys and show 
debug info on SparkUI;
   
   3, in `OptimizeSkewedJoin`, estimate the joined size of each partition based 
on the sampled keys, and split a partition if it is not split yet and its 
estimated joined size is too larger.
   
   
   ### Why are the changes needed?
   1, make it easy to add a use-defined statistics which can be used in AQE 
rules;
   2, showing skew info on sparkUI is usefully;
   3, spliting partitions based on joined size can resolve data inflation;
   
   
   ### Does this PR introduce _any_ user-facing change?
   Yes, new features are added
   
   
   ### How was this patch tested?
   added testsuites
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to