s0nskar opened a new pull request, #47954:
URL: https://github.com/apache/spark/pull/47954
### What changes were proposed in this pull request?
Do not include empty partitions for skew join calculations.
### Why are the changes needed?
Spark currently includes empty partitions while calculating the median size
for the skew join optimization, which does not truly represent the data
distribution required for skew optimization. This makes the median inaccurate
if too many empty partitions are present median will represent a small value or
even 0.
Spark also uses empty partitions while calculating target size for skewed
partitions. It considers empty partitions as non-skewed partitions and affects
the average size of non-skewed partitions required for target size calculation.
```
val nonSkewSizes = sizes.filter(_ <= skewThreshold)
val targetSize = max(nonSkewSizes.sum / nonSkewSizes.length, advisorySize)
```
Example from AdaptiveQueryExecSuite –
```
spark.range(0, 1000, 1, 10).selectExpr("id % 3 as key1", "id as
value1").createOrReplaceTempView("skewData1")
spark.range(0, 1000, 1, 10).selectExpr("id % 1 as key2", "id as
value2").createOrReplaceTempView("skewData2")
"SELECT key1 FROM skewData1 JOIN skewData2 ON key1 = key2"
```
Here for the left side has only 3 non empty partition exists and for right
side has only 1 non empty partition exists. And even though they are not skewed
in size, spark considers them skew and splits them. Which is wrong behaviour.
```
OptimizeSkewedJoin: Left side partition 5 (3 KB) is skewed, split it into 5
parts.
OptimizeSkewedJoin: Left side partition 28 (4 KB) is skewed, split it into 5
parts.
OptimizeSkewedJoin: Left side partition 69 (4 KB) is skewed, split it into 5
parts.
OptimizeSkewedJoin: Right side partition 5 (11 KB) is skewed, split it into
10 parts.
```
### Does this PR introduce _any_ user-facing change?
<!--
Note that it means *any* user-facing change including all aspects such as
the documentation fix.
If yes, please clarify the previous behavior and the change this PR proposes
- provide the console output, description and/or an example to show the
behavior difference if possible.
If possible, please also clarify if this is a user-facing change compared to
the released Spark versions or within the unreleased branches such as master.
If no, write 'No'.
-->
No
### How was this patch tested?
<!--
If tests were added, say they were added here. Please make sure to add some
test cases that check the changes thoroughly including negative and positive
cases if possible.
If it was tested in a way different from regular unit tests, please clarify
how you tested step by step, ideally copy and paste-able, so that other
reviewers can test and check, and descendants can verify in the future.
If tests were not added, please describe why they were not added and/or why
it was difficult to add.
If benchmark tests were added, please run the benchmarks in GitHub Actions
for the consistent environment, and the instructions could accord to:
https://spark.apache.org/developer-tools.html#github-workflow-benchmarks.
-->
WIP
### 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: [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]
