beliefer opened a new pull request, #41899:
URL: https://github.com/apache/spark/pull/41899
### What changes were proposed in this pull request?
`WindowGroupLimitExec` is updated to use the PartitionEvaluator API to do
execution.
### Why are the changes needed?
To define the computing logic and requires the caller side to explicitly
list what needs to be serialized and sent to executors
### Does this PR introduce _any_ user-facing change?
'No'.
Just update the inner implementation.
### How was this patch tested?
Test cases updated & running benchmark manually.
```
[info] Java HotSpot(TM) 64-Bit Server VM 1.8.0_311-b11 on Mac OS X 10.16
[info] Intel(R) Core(TM) i7-9750H CPU @ 2.60GHz
[info] Benchmark Top-K: Best
Time(ms) Avg Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
[info]
-----------------------------------------------------------------------------------------------------------------------------------------------
[info] ROW_NUMBER (PARTITION: , WindowGroupLimit: false)
10622 11266 617 2.0 506.5 1.0X
[info] ROW_NUMBER (PARTITION: , WindowGroupLimit: true)
1712 1744 19 12.2 81.6 6.2X
[info] ROW_NUMBER (PARTITION: PARTITION BY b, WindowGroupLimit: false)
23679 25107 NaN 0.9 1129.1 0.4X
[info] ROW_NUMBER (PARTITION: PARTITION BY b, WindowGroupLimit: true)
6381 6527 95 3.3 304.3 1.7X
[info] RANK (PARTITION: , WindowGroupLimit: false)
11492 11631 106 1.8 548.0 0.9X
[info] RANK (PARTITION: , WindowGroupLimit: true)
2675 2920 118 7.8 127.5 4.0X
[info] RANK (PARTITION: PARTITION BY b, WindowGroupLimit: false)
24208 24299 95 0.9 1154.3 0.4X
[info] RANK (PARTITION: PARTITION BY b, WindowGroupLimit: true)
6347 6478 85 3.3 302.6 1.7X
[info] DENSE_RANK (PARTITION: , WindowGroupLimit: false)
11288 11959 458 1.9 538.2 0.9X
[info] DENSE_RANK (PARTITION: , WindowGroupLimit: true)
2684 2945 144 7.8 128.0 4.0X
[info] DENSE_RANK (PARTITION: PARTITION BY b, WindowGroupLimit: false)
24316 25130 711 0.9 1159.5 0.4X
[info] DENSE_RANK (PARTITION: PARTITION BY b, WindowGroupLimit: true)
6589 6925 383 3.2 314.2 1.6X
```
--
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]
