refactor: refine experimental final aggregate spill #18907
Merged
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
github-actions
bot
added
the
pr-refactor
3 tasks
There was a problem hiding this comment.
Pull Request Overview
This PR refactors the aggregate spill implementation by renaming settings and introducing a new spiller architecture with multi-level spill support. The main changes consolidate spill logic, add a maximum recursion depth setting for aggregate spills, and improve memory management during aggregation.
- Renamed
enable_experiment_aggregate_finaltoenable_experiment_aggregatewith updated default from 0 to 1 - Added
max_aggregate_spill_levelsetting to limit recursion depth during aggregate spilling - Replaced
FinalAggregateSpillerwithNewAggregateSpillerusing parquet-based spilling
Reviewed Changes
Copilot reviewed 17 out of 17 changed files in this pull request and generated 4 comments.
Show a summary per file
| File | Description |
|---|---|
| settings_getter_setter.rs | Renamed getter method to match new setting name and added getter for max spill level |
| settings_default.rs | Renamed setting, changed default to 1, and added max_aggregate_spill_level setting |
| new_transform_final_aggregate.rs | Major refactor to extract state management, improve spilling logic, and add multi-level spill support |
| new_final_aggregate_state.rs | New file containing extracted state management structs (RoundPhase, LocalRoundState, FinalAggregateSharedState) |
| new_aggregate_spiller.rs | New file implementing parquet-based spiller replacing legacy serialization approach |
| final_aggregate_spiller.rs | Deleted file - legacy spiller implementation removed |
| aggregate_meta.rs | Added NewSpilledPayload struct and NewBucketSpilled enum variant |
| transform_partition_bucket*.rs | Added unreachable handlers for NewBucketSpilled variant |
| payload/partitioned_payload/aggregate_hashtable | Added reset_for_reuse methods to support hashtable reuse |
| build_partition_bucket.rs | Updated to use new spiller and pass max_aggregate_spill_level parameter |
| physical_aggregate_final.rs | Updated to use renamed setting |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
...elines/processors/transforms/aggregator/new_final_aggregate/new_transform_final_aggregate.rs
Outdated
Show resolved
Hide resolved
.../src/pipelines/processors/transforms/aggregator/new_final_aggregate/new_aggregate_spiller.rs
Outdated
Show resolved
Hide resolved
.../src/pipelines/processors/transforms/aggregator/new_final_aggregate/new_aggregate_spiller.rs
Outdated
Show resolved
Hide resolved
dqhl76
added
ci-benchmark
Docker Image for PR
|
fix: re-enable check spill by setting max spill depth refactor: split local/shared state
dqhl76
force-pushed
the
performance-final-aggregate-31
branch
from
7ceabe9 to
4587399
Compare
dqhl76
force-pushed
the
performance-final-aggregate-31
branch
from
4587399 to
1ba503a
Compare
dqhl76
added
ci-benchmark
Docker Image for PR
|
dqhl76
added
ci-benchmark
Docker Image for PR
|
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
...rc/pipelines/processors/transforms/aggregator/new_aggregate/new_transform_final_aggregate.rs
Show resolved
Hide resolved
zhang2014
approved these changes
Nov 5, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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.
I hereby agree to the terms of the CLA available at: https://docs.databend.com/dev/policies/cla/
Summary
SpillsBufferPoolmax_aggregate_spill_levelTests
Type of change
This change is