Repository navigation
docs: explain spark.comet.shuffle.native.maxBufferBytes in the shuffle tuning guide - #7308
Merged
Merged
Conversation
…e tuning guide The setting was described only in the generated configuration reference and the contributor guide. Add a short section on what it does, when to set it, and what it costs, and mention it in the tuning guide's index.
github-actions
Bot
added
the
documentation
andygrove
approved these changes
Oct 28, 2026
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Which issue does this PR close?
None. This documents an existing setting that the user guide did not explain.
Rationale for this change
spark.comet.shuffle.native.maxBufferBytesis only described in the generated configuration reference and the contributor guide, so users whose tasks end in a native shuffle have no guidance on when to set it or what it costs.By default the writer buffers until the memory pool refuses it. In a wide group-by we ran (87 sort keys, 12 tasks per executor), that was about 2.6 GB per map task, and executors were killed for exceeding their containers. With the limit at
256m, far fewer executors were killed, and a local reproduction of the same job showed no slowdown and a 3% larger shuffle. At128mand below, the stage reading the shuffle got slower.What changes are included in this PR?
How are these changes tested?
Documentation only. prettier, which CI runs on Markdown, makes no changes.