sunchao opened a new pull request, #36068:
URL: https://github.com/apache/spark/pull/36068
This is a backport of #35657 to `branch-3.3`
<!--
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 introduces the initial implementation of Storage-Partitioned Join
([SPIP](https://docs.google.com/document/d/1foTkDSM91VxKgkEcBMsuAvEjNybjja-uHk-r3vtXWFE)).
Changes:
- `org.apache.spark.sql.connector.read.partitioning.Partitioning` currently
is very limited (as mentioned in the SPIP), and cannot be extended to handle
join cases. This PR completely replace it following the catalyst `Partitioning`
interface, and added two concrete sub-classes: `KeyGroupedPartitioning` and
`UnknownPartitioning`. This allows a V2 data source to report to Spark it's
partition transform expressions, via `SupportsReportPartitioning` interface.
- with the above change,
`org.apache.spark.sql.connector.read.partitioning.Distribution` and
`org.apache.spark.sql.connector.read.partitioning.ClusteredDistribution` now
are replaced by classes with the same name in
`org.apache.spark.sql.connector.distributions` package. Therefore, this PR
marks the former two as deprecated.
- `DataSourcePartitioning` used to be in
`org.apache.spark.sql.execution.datasources.v2`. This moves it into package
`org.apache.spark.sql.catalyst.plans.physical` and renames it to
`KeyGroupedPartitioning`, so that it can be extended for more non-V2 use cases,
such as Hive bucketing. In addition, it is also changed to accommodate the
Storage-Partitioned Join feature.
- a new expression type: `TransformExpression`, is introduced to bind
syntactic partition transforms with their semantic meaning, represented by a V2
function. This expression is un-evaluable for now, and is used later in
`EnsureRequirements` to check whether join children are compatible with each
other.
- a new optimizer rule: `V2ScanPartitioning`, is added to recognize `Scan`s
implement `SupportsReportPartitioning`. If they do, this rule converts V2
partition transform expressions into their counterparts in catalyst, and
annotate `DataSourceV2ScanRelation` with the result. These are later propagated
into `DataSourceV2ScanExecBase`.
- changes are made in `DataSourceV2ScanExecBase` to create
`KeyGroupedPartitioning` for scan if 1) the scan is annotated with catalyst
partition transform expressions, and 2) if all input splits implement
`HasPartitionKey`.
- A new config: `spark.sql.sources.v2.bucketing.enabled` is introduced to
turn on or off the behavior. By default it is false.
### 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.
-->
Spark currently support bucketing in DataSource V1, but not in V2. This is
the first step to support bucket join, and is general form, storage-partitioned
join, for V2 data sources. In addition, the work here can potentially used to
support Hive bucketing as well. Please check the SPIP for details.
### 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'.
-->
With the changes, a user can now:
- have V2 data sources to report distribution and ordering to Spark on read
path
- Spark will recognize the distribution property and eliminate shuffle in
join/aggregate/window, etc, when the source distribution matches the required
distribution from these.
- a new config `spark.sql.sources.v2.bucketing.enabled` is introduced to
turn on/off the above behavior
### 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.
-->
- Added a new test suite `KeyGroupedPartitioningSuite` covers end-to-end
tests on the new feature
- Extended `EnsureRequirementsSuite` to cover `DataSourcePartitioning`
- Some existing test classes, such as `InMemoryTable` are extended to cover
the changes
--
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]
