experimental: Implement initial framework for cost-based optimizations to avoid moving to Comet in some cases #618
Conversation
This file contains 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
andygrove
changed the title
andygrove
changed the title
|
Moving to draft until I have unit tests |
parthchandra
reviewed
Jul 2, 2024
| // fall back for sort amd exchange operators | ||
| val fallbackReason = "avoid move to Comet just for sort" | ||
| plan.setTagValue(CometExplainInfo.CBO_FALLBACK, fallbackReason) | ||
| plan.children.head.setTagValue(CometExplainInfo.CBO_FALLBACK, fallbackReason) |
There was a problem hiding this comment.
Setting the fallbackReason for the plan node should be enough I think. No need to set it for the child node.
andygrove
changed the title
|
There is no immediate benefit in merging this so I will close this issue for now. I wrote up my findings in #833 and may revisit this in the future |
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Outdated suggestions cannot be applied.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?
Part of #571
Rationale for this change
This is a first step towards a cost-based optimizer that can avoid moving from Spark to Comet in some cases where the benefit of moving to Comet could be outweighed by the overhead of the surrounding R2C and C2R transitions.
In this case we use a very specific rule to avoid moving to Comet for the final query stage if that query stage is just a SortExec (as we currently see with TPC-H q16).
Note that the actual performance difference of the optimization in this case is negligible because the final query stage is only processing a relatively small number of rows, but the value of this PR is getting the basic mechanism in place for future optimizations.
Before
After
What changes are included in this PR?
How are these changes tested?