You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by GitBox <gi...@apache.org> on 2022/12/06 00:32:37 UTC
[GitHub] [spark] sunchao opened a new pull request, #38924: [SPARK-41398][SQL] Relax constraints on Storage-Partitioned Join when partition keys after runtime filtering do not match
sunchao opened a new pull request, #38924:
URL: https://github.com/apache/spark/pull/38924
<!--
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?
<!--
Please clarify what changes you are proposing. The purpose of this section is to outline the changes and how this PR fixes the issue.
If possible, please consider writing useful notes for better and faster reviews in your PR. See the examples below.
1. If you refactor some codes with changing classes, showing the class hierarchy will help reviewers.
2. If you fix some SQL features, you can provide some references of other DBMSes.
3. If there is design documentation, please add the link.
4. If there is a discussion in the mailing list, please add the link.
-->
This PR relaxes the current constraint of Storage-Partitioned Join which requires that the partition keys after runtime filtering to be exact the same as the partition keys before the filtering.
### Why are the changes needed?
<!--
Please clarify why the changes are needed. For instance,
1. If you propose a new API, clarify the use case for a new API.
2. If you fix a bug, you can clarify why it is a bug.
-->
At the moment, Spark requires that when Storage-Partitioned Join is used together with runtime filtering, the partition keys before and after the filtering shall exact match. If not, a `SparkException` is thrown.
However, this is not strictly necessary in the case where the partition keys after the filtering is a subset of the original keys. In this scenario, we can use empty partitions for those missing keys in the latter.
### 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.
-->
Modified an existing test case to match the change.
--
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-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] dongjoon-hyun commented on pull request #38924: [SPARK-41398][SQL] Relax constraints on Storage-Partitioned Join when partition keys after runtime filtering do not match
Posted by GitBox <gi...@apache.org>.
dongjoon-hyun commented on PR #38924:
URL: https://github.com/apache/spark/pull/38924#issuecomment-1338661559
Also, cc @wangyum , @MaxGekk , @yaooqinn , @c21
--
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-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] cloud-fan commented on a diff in pull request #38924: [SPARK-41398][SQL] Relax constraints on Storage-Partitioned Join when partition keys after runtime filtering do not match
Posted by GitBox <gi...@apache.org>.
cloud-fan commented on code in PR #38924:
URL: https://github.com/apache/spark/pull/38924#discussion_r1041775153
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/BatchScanExec.scala:
##########
@@ -114,8 +117,21 @@ case class BatchScanExec(
// return an empty RDD with 1 partition if dynamic filtering removed the only split
sparkContext.parallelize(Array.empty[InternalRow], 1)
} else {
+ var finalPartitions = filteredPartitions
+
+ outputPartitioning match {
+ case p: KeyGroupedPartitioning =>
+ val partitionMapping = finalPartitions.map(s =>
+ s.head.asInstanceOf[HasPartitionKey].partitionKey() -> s).toMap
+ finalPartitions = p.partitionValuesOpt.get.map { partKey =>
Review Comment:
It's better to centralize the related code.
This empty filling is quite related to the `if (oldRows.size < newRows.size)` check. We should put them together if possible
--
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-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] dongjoon-hyun closed pull request #38924: [SPARK-41398][SQL] Relax constraints on Storage-Partitioned Join when partition keys after runtime filtering do not match
Posted by GitBox <gi...@apache.org>.
dongjoon-hyun closed pull request #38924: [SPARK-41398][SQL] Relax constraints on Storage-Partitioned Join when partition keys after runtime filtering do not match
URL: https://github.com/apache/spark/pull/38924
--
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-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] cloud-fan commented on a diff in pull request #38924: [SPARK-41398][SQL] Relax constraints on Storage-Partitioned Join when partition keys after runtime filtering do not match
Posted by GitBox <gi...@apache.org>.
cloud-fan commented on code in PR #38924:
URL: https://github.com/apache/spark/pull/38924#discussion_r1041721325
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/BatchScanExec.scala:
##########
@@ -114,8 +117,21 @@ case class BatchScanExec(
// return an empty RDD with 1 partition if dynamic filtering removed the only split
sparkContext.parallelize(Array.empty[InternalRow], 1)
} else {
+ var finalPartitions = filteredPartitions
+
+ outputPartitioning match {
+ case p: KeyGroupedPartitioning =>
+ val partitionMapping = finalPartitions.map(s =>
+ s.head.asInstanceOf[HasPartitionKey].partitionKey() -> s).toMap
+ finalPartitions = p.partitionValuesOpt.get.map { partKey =>
Review Comment:
shall we move this logic to `lazy val filteredPartitions`?
--
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-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] cloud-fan commented on a diff in pull request #38924: [SPARK-41398][SQL] Relax constraints on Storage-Partitioned Join when partition keys after runtime filtering do not match
Posted by GitBox <gi...@apache.org>.
cloud-fan commented on code in PR #38924:
URL: https://github.com/apache/spark/pull/38924#discussion_r1041775153
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/BatchScanExec.scala:
##########
@@ -114,8 +117,21 @@ case class BatchScanExec(
// return an empty RDD with 1 partition if dynamic filtering removed the only split
sparkContext.parallelize(Array.empty[InternalRow], 1)
} else {
+ var finalPartitions = filteredPartitions
+
+ outputPartitioning match {
+ case p: KeyGroupedPartitioning =>
+ val partitionMapping = finalPartitions.map(s =>
+ s.head.asInstanceOf[HasPartitionKey].partitionKey() -> s).toMap
+ finalPartitions = p.partitionValuesOpt.get.map { partKey =>
Review Comment:
It's better to centralize the related code.
--
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-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] dongjoon-hyun commented on pull request #38924: [SPARK-41398][SQL] Relax constraints on Storage-Partitioned Join when partition keys after runtime filtering do not match
Posted by GitBox <gi...@apache.org>.
dongjoon-hyun commented on PR #38924:
URL: https://github.com/apache/spark/pull/38924#issuecomment-1339580946
Merged to master for Apache Spark 3.4.0.
--
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-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] sunchao commented on a diff in pull request #38924: [SPARK-41398][SQL] Relax constraints on Storage-Partitioned Join when partition keys after runtime filtering do not match
Posted by GitBox <gi...@apache.org>.
sunchao commented on code in PR #38924:
URL: https://github.com/apache/spark/pull/38924#discussion_r1041745736
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/BatchScanExec.scala:
##########
@@ -114,8 +117,21 @@ case class BatchScanExec(
// return an empty RDD with 1 partition if dynamic filtering removed the only split
sparkContext.parallelize(Array.empty[InternalRow], 1)
} else {
+ var finalPartitions = filteredPartitions
+
+ outputPartitioning match {
+ case p: KeyGroupedPartitioning =>
+ val partitionMapping = finalPartitions.map(s =>
+ s.head.asInstanceOf[HasPartitionKey].partitionKey() -> s).toMap
+ finalPartitions = p.partitionValuesOpt.get.map { partKey =>
Review Comment:
Hmm any obvious advantage of this? It looks the same to me 😅
--
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-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] cloud-fan commented on a diff in pull request #38924: [SPARK-41398][SQL] Relax constraints on Storage-Partitioned Join when partition keys after runtime filtering do not match
Posted by GitBox <gi...@apache.org>.
cloud-fan commented on code in PR #38924:
URL: https://github.com/apache/spark/pull/38924#discussion_r1041671515
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/BatchScanExec.scala:
##########
@@ -81,18 +81,21 @@ case class BatchScanExec(
val newRows = new InternalRowSet(p.expressions.map(_.dataType))
newRows ++= newPartitions.map(_.asInstanceOf[HasPartitionKey].partitionKey())
- val oldRows = p.partitionValuesOpt.get
- if (oldRows.size != newRows.size) {
- throw new SparkException("Data source must have preserved the original partitioning " +
- "during runtime filtering: the number of unique partition values obtained " +
- s"through HasPartitionKey changed: before ${oldRows.size}, after ${newRows.size}")
+ val oldRows = p.partitionValuesOpt.get.toSet
Review Comment:
shall we use `InternalRowSet`?
--
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-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] sunchao commented on a diff in pull request #38924: [SPARK-41398][SQL] Relax constraints on Storage-Partitioned Join when partition keys after runtime filtering do not match
Posted by GitBox <gi...@apache.org>.
sunchao commented on code in PR #38924:
URL: https://github.com/apache/spark/pull/38924#discussion_r1041789809
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/BatchScanExec.scala:
##########
@@ -114,8 +117,21 @@ case class BatchScanExec(
// return an empty RDD with 1 partition if dynamic filtering removed the only split
sparkContext.parallelize(Array.empty[InternalRow], 1)
} else {
+ var finalPartitions = filteredPartitions
+
+ outputPartitioning match {
+ case p: KeyGroupedPartitioning =>
+ val partitionMapping = finalPartitions.map(s =>
+ s.head.asInstanceOf[HasPartitionKey].partitionKey() -> s).toMap
+ finalPartitions = p.partitionValuesOpt.get.map { partKey =>
Review Comment:
I was hoping to re-use the logic code in #38950 too, which is not related to runtime filtering. If moving this close to the `if (oldRows.size < newRows.size)` check then I may have to duplicate it somewhere else?
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/BatchScanExec.scala:
##########
@@ -114,8 +117,21 @@ case class BatchScanExec(
// return an empty RDD with 1 partition if dynamic filtering removed the only split
sparkContext.parallelize(Array.empty[InternalRow], 1)
} else {
+ var finalPartitions = filteredPartitions
+
+ outputPartitioning match {
+ case p: KeyGroupedPartitioning =>
+ val partitionMapping = finalPartitions.map(s =>
+ s.head.asInstanceOf[HasPartitionKey].partitionKey() -> s).toMap
+ finalPartitions = p.partitionValuesOpt.get.map { partKey =>
Review Comment:
I was hoping to re-use the logic in #38950 too, which is not related to runtime filtering. If moving this close to the `if (oldRows.size < newRows.size)` check then I may have to duplicate it somewhere else?
--
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-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] sunchao commented on pull request #38924: [SPARK-41398][SQL] Relax constraints on Storage-Partitioned Join when partition keys after runtime filtering do not match
Posted by GitBox <gi...@apache.org>.
sunchao commented on PR #38924:
URL: https://github.com/apache/spark/pull/38924#issuecomment-1338488876
cc @cloud-fan @dongjoon-hyun @viirya @aokolnychyi @huaxingao
--
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-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] sunchao commented on a diff in pull request #38924: [SPARK-41398][SQL] Relax constraints on Storage-Partitioned Join when partition keys after runtime filtering do not match
Posted by GitBox <gi...@apache.org>.
sunchao commented on code in PR #38924:
URL: https://github.com/apache/spark/pull/38924#discussion_r1041716400
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/BatchScanExec.scala:
##########
@@ -81,18 +81,21 @@ case class BatchScanExec(
val newRows = new InternalRowSet(p.expressions.map(_.dataType))
newRows ++= newPartitions.map(_.asInstanceOf[HasPartitionKey].partitionKey())
- val oldRows = p.partitionValuesOpt.get
- if (oldRows.size != newRows.size) {
- throw new SparkException("Data source must have preserved the original partitioning " +
- "during runtime filtering: the number of unique partition values obtained " +
- s"through HasPartitionKey changed: before ${oldRows.size}, after ${newRows.size}")
+ val oldRows = p.partitionValuesOpt.get.toSet
Review Comment:
Oops you're right! I forgot about `InternalRowSet`. Let me create a follow-up.
--
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-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] [spark] cloud-fan commented on a diff in pull request #38924: [SPARK-41398][SQL] Relax constraints on Storage-Partitioned Join when partition keys after runtime filtering do not match
Posted by GitBox <gi...@apache.org>.
cloud-fan commented on code in PR #38924:
URL: https://github.com/apache/spark/pull/38924#discussion_r1041876441
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/BatchScanExec.scala:
##########
@@ -114,8 +117,21 @@ case class BatchScanExec(
// return an empty RDD with 1 partition if dynamic filtering removed the only split
sparkContext.parallelize(Array.empty[InternalRow], 1)
} else {
+ var finalPartitions = filteredPartitions
+
+ outputPartitioning match {
+ case p: KeyGroupedPartitioning =>
+ val partitionMapping = finalPartitions.map(s =>
+ s.head.asInstanceOf[HasPartitionKey].partitionKey() -> s).toMap
+ finalPartitions = p.partitionValuesOpt.get.map { partKey =>
Review Comment:
ah then it makes sense to keep it here.
--
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-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org