[SPARK-47720][CORE] Update spark.speculation.multiplier to 3 and spark.speculation.quantile to 0.9
#45858
Conversation
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
…park.speculation.quantile` to 0.9
Member
Author
|
WDYT, @cloud-fan and @HyukjinKwon ? I received many requests due to the following. |
mridulm
approved these changes
Apr 3, 2024
Contributor
mridulm
left a comment
mridulm
left a comment
There was a problem hiding this comment.
We actually use 4 and 0.9 internally :-)
We also have a minTaskRuntime of 30s ... though this might not work well for streaming apps.
Member
|
cc @Ngone51 and @jiangxb1987 FYI |
Member
Author
|
Thank you, @mridulm and @HyukjinKwon . |
Member
Author
|
Merged to master for Apache Spark 4.0.0. |
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Outdated suggestions cannot be applied.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 adjust the following in order to make Spark speculative execution behavior less aggressive from Apache Spark 4.0.0.
spark.speculation.multiplier: 1.5 -> 3spark.speculation.quantile: 0.75 -> 0.9Why are the changes needed?
Although
spark.speculationis disabled by default, this has been used in many production use cases.Does this PR introduce any user-facing change?
This will make a speculative execution less agressive.
How was this patch tested?
Pass the CIs.
Was this patch authored or co-authored using generative AI tooling?
No.