Skip to content

Conversation

@xuanyuanking
Copy link
Member

@xuanyuanking xuanyuanking commented Nov 25, 2019

What changes were proposed in this pull request?

  1. Make maxNumPostShufflePartitions config obey reducePostShfflePartitions config.
  2. Update the description for all the SQLConf affected by spark.sql.adaptive.enabled.

Why are the changes needed?

Make the relation between these confs clearer.

Does this PR introduce any user-facing change?

No

How was this patch tested?

Existing UT.

buildConf("spark.sql.adaptive.shuffle.targetPostShuffleInputSize")
.doc("The target post-shuffle input size in bytes of a task.")
.doc("The target post-shuffle input size in bytes of a task. This configuration only has " +
s"an effect when '${ADAPTIVE_EXECUTION_ENABLED.key}' is enabled.")
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

and also when spark.sql.adaptive.shuffle.reducePostShufflePartitions.enabled is true?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks, fix the doc of SHUFFLE_MIN_NUM_POSTSHUFFLE_PARTITIONS and SHUFFLE_TARGET_POSTSHUFFLE_INPUT_SIZE.

@SparkQA
Copy link

SparkQA commented Nov 25, 2019

Test build #114407 has finished for PR 26664 at commit cd71aac.

  • This patch passes all tests.
  • This patch merges cleanly.
  • This patch adds no public classes.

@SparkQA
Copy link

SparkQA commented Nov 25, 2019

Test build #114423 has finished for PR 26664 at commit 81c536d.

  • This patch passes all tests.
  • This patch merges cleanly.
  • This patch adds no public classes.

@xuanyuanking xuanyuanking changed the title [SPARK-9853][Doc][Follow-up] Update the description for all the SQLConf affected by spark.sql.adaptive.enabled [SPARK-30074][SQL] The maxNumPostShufflePartitions config should obey reducePostShufflePartitions enabled Nov 29, 2019
@SparkQA
Copy link

SparkQA commented Nov 29, 2019

Test build #114602 has finished for PR 26664 at commit 69070d7.

  • This patch fails due to an unknown error code, -9.
  • This patch merges cleanly.
  • This patch adds no public classes.

@xuanyuanking
Copy link
Member Author

retest this please.

@SparkQA
Copy link

SparkQA commented Nov 29, 2019

Test build #114617 has finished for PR 26664 at commit 69070d7.

  • This patch passes all tests.
  • This patch merges cleanly.
  • This patch adds no public classes.

@cloud-fan cloud-fan closed this in d1465a1 Dec 2, 2019
@cloud-fan
Copy link
Contributor

thanks, merging to master!

@xuanyuanking
Copy link
Member Author

Thanks!

@xuanyuanking xuanyuanking deleted the SPARK-9853-follow branch December 2, 2019 04:54
case class EnsureRequirements(conf: SQLConf) extends Rule[SparkPlan] {
private def defaultNumPreShufflePartitions: Int =
if (conf.adaptiveExecutionEnabled) {
if (conf.adaptiveExecutionEnabled && conf.reducePostShufflePartitionsEnabled) {
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@xuanyuanking is there any UT verifying the number of shuffle partitions dependent on values of these two configurations ?

attilapiros pushed a commit to attilapiros/spark that referenced this pull request Dec 6, 2019
… reducePostShufflePartitions enabled

### What changes were proposed in this pull request?
1. Make maxNumPostShufflePartitions config obey reducePostShfflePartitions config.
2. Update the description for all the SQLConf affected by `spark.sql.adaptive.enabled`.

### Why are the changes needed?
Make the relation between these confs clearer.

### Does this PR introduce any user-facing change?
No

### How was this patch tested?
Existing UT.

Closes apache#26664 from xuanyuanking/SPARK-9853-follow.

Authored-by: Yuanjian Li <[email protected]>
Signed-off-by: Wenchen Fan <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants