[SPARK-57723][CORE] Enable spark.rdd.compress by default like PySpark#56824
Closed
dongjoon-hyun wants to merge 2 commits into
Closed
[SPARK-57723][CORE] Enable spark.rdd.compress by default like PySpark#56824dongjoon-hyun wants to merge 2 commits into
spark.rdd.compress by default like PySpark#56824dongjoon-hyun wants to merge 2 commits into
Conversation
Member
Author
|
cc @HyukjinKwon , @cloud-fan , @viirya , @MaxGekk |
MaxGekk
approved these changes
Jun 26, 2026
Member
Author
|
Thank you, @MaxGekk . I'll make it sure that all CIs pass in this PR. |
viirya
approved these changes
Jun 26, 2026
Member
Author
|
Thank you, @viirya ! |
dongjoon-hyun
added a commit
that referenced
this pull request
Jun 27, 2026
### What changes were proposed in this pull request? This PR aims to enable `spark.rdd.compress` by default for **non-Python** languages because we are supporting more and more languages via `Spark Connect`, we had better match the behavior with PySpark for the feature parity across all languages. Currently, only `Python` language has this advantage unfairly in the Apache Spark language eco-system. https://github.com/apache/spark/blob/c3e9ea2299c0fb064e913b2c8463ec777ef6cdb4/python/pyspark/core/context.py#L88 - Scala/Java (main repository) - https://github.com/apache/spark-connect-swift - https://github.com/apache/spark-connect-rust - https://github.com/apache/spark-connect-go ### Why are the changes needed? Since Apache Spark 2.0, PySpark has been taking advantage of `spark.rdd.compres` by default stably to compress serialized RDD partitions and has been saving memory and disk at a small CPU cost. Since `PySpark` has long defaulted this to `true`; making it the core default applies the same proven default across all languages. - #1568 For the record, Python is the first user-facing language in the Apache Spark website and its behavior is also the same. - https://spark.apache.org <img width="1071" height="347" alt="Screenshot 2026-06-26 at 13 37 43" src="https://github.com/user-attachments/assets/cc92c785-9763-44cc-9277-c1b40bf2a0c3" /> ### Does this PR introduce _any_ user-facing change? - For PySpark users, no. - For non-PySpark users, yes. Serialized RDD partitions are now compressed by default. To restore the previous behavior, set `spark.rdd.compress` to `false`. ### How was this patch tested? Pass the CIs. ### Was this patch authored or co-authored using generative AI tooling? Generated-by: Claude Opus 4.8 Closes #56824 from dongjoon-hyun/SPARK-57723. Authored-by: Dongjoon Hyun <dongjoon@apache.org> Signed-off-by: Dongjoon Hyun <dongjoon@apache.org> (cherry picked from commit 0ac3189) Signed-off-by: Dongjoon Hyun <dongjoon@apache.org>
Member
Author
|
Merged to master/4.x for Apache Spark 4.3.0. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What changes were proposed in this pull request?
This PR aims to enable
spark.rdd.compressby default for non-Python languages because we are supporting more and more languages viaSpark Connect, we had better match the behavior with PySpark for the feature parity across all languages. Currently, onlyPythonlanguage has this advantage unfairly in the Apache Spark language eco-system.spark/python/pyspark/core/context.py
Line 88 in c3e9ea2
Why are the changes needed?
Since Apache Spark 2.0, PySpark has been taking advantage of
spark.rdd.compresby default stably to compress serialized RDD partitions and has been saving memory and disk at a small CPU cost. SincePySparkhas long defaulted this totrue; making it the core default applies the same proven default across all languages.For the record, Python is the first user-facing language in the Apache Spark website and its behavior is also the same.
Does this PR introduce any user-facing change?
spark.rdd.compresstofalse.How was this patch tested?
Pass the CIs.
Was this patch authored or co-authored using generative AI tooling?
Generated-by: Claude Opus 4.8