wankunde opened a new pull request, #43009:
URL: https://github.com/apache/spark/pull/43009
<!--
Thanks for sending a pull request! Here are some tips for you:
1. If this is your first time, please read our contributor guidelines:
https://spark.apache.org/contributing.html
2. Ensure you have added or run the appropriate tests for your PR:
https://spark.apache.org/developer-tools.html
3. If the PR is unfinished, add '[WIP]' in your PR title, e.g.,
'[WIP][SPARK-XXXX] Your PR title ...'.
4. Be sure to keep the PR description updated to reflect all changes.
5. Please write your PR title to summarize what this PR proposes.
6. If possible, provide a concise example to reproduce the issue for a
faster review.
7. If you want to add a new configuration, please read the guideline first
for naming configurations in
'core/src/main/scala/org/apache/spark/internal/config/ConfigEntry.scala'.
8. If you want to add or modify an error type or message, please read the
guideline first in
'core/src/main/resources/error/README.md'.
-->
### What changes were proposed in this pull request?
This PR could be a followup of https://github.com/apache/spark/pull/42488
and https://github.com/apache/spark/pull/42557.
If there is an aggregate operator after the SMJ and the grouping expressions
of aggregate operator contain all the join keys of the streamed side, we can
add a sorter on the streamed side of the SMJ, so that the aggregate can be
convert to a SortAggregate which will be faster than HashAggregate.
For example, with table t1(a, b, c) and t2(x, y, z):
```
SELECT a, b, sum(c)
FROM t1
JOIN t2
ON t1.b = t2.y
GROUP BY a, b
```
The optimized plan:
```
Scan(t1) Scan(t2)
| |
| |
Exchange 1 Exchange 2
\ /
\ /
\ /
SMJ (t1.b = t2.y)
|
|
Aggregate
```
Before this PR, spark EnsureReqirement will add Sorter(t1.b) to the left
side of SMJ.
```
Scan(t1) Scan(t2)
| |
| |
Exchange 1 Exchange 2
\ /
Sort(t1.b) Sort(t2.y)
\ /
SMJ (t1.b = t2.y)
|
|
HashAggregate
```
If we add a Sort(t1.b, t1.a) to the left side of the SMJ, the following
aggregate could be convert to SortAggregate, will be faster.
```
Scan(t1) Scan(t2)
| |
| |
Exchange 1 Exchange 2
\ /
Sort(t1.b, t1.a) Sort(t2.y)
\ /
SMJ (t1.b = t2.y)
|
|
SortAggregate
```
### Why are the changes needed?
Optimize HashAggregate
### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
Added UT
### Was this patch authored or co-authored using generative AI tooling?
No
--
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: reviews-unsubscr...@spark.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
