korowa opened a new pull request, #11627:
URL: https://github.com/apache/datafusion/pull/11627
## Which issue does this PR close?
<!--
We generally require a GitHub issue to be filed for all bug fixes and
enhancements and this helps us generate change logs for our releases. You can
link an issue to this PR using the GitHub syntax. For example `Closes #123`
indicates that this PR will close issue #123.
-->
Related to #6937.
## Rationale for this change
<!--
Why are you proposing this change? If this is already explained clearly in
the issue then this section is not needed.
Explaining clearly why changes are proposed helps reviewers understand your
changes and offer better suggestions for fixes.
-->
Currently DF plans (almost always) two aggregation operators -- Partial and
Final, executing one after another with Partial output being input for Final.
In case aggregate input is almost/close to unique, Partial aggregation doesn't
group data well (output row count +- same as input rowcount), and DF ends up
with doing the same work twice.
Suggestion is to start skipping partial aggregation after some fixed amount
of input rows, in case at that moment accumulated unique groups / input rows
exceeds some fixed threshold value (which by default is somewhere between 0.5
and 1, but closer to 1), and produce batched "as-is", replacing aggregate
accumulators inputs with corresponding intermediate aggregate states (in order
not to break record batch schema for downstream operators -- specifically, for
`CoalesceBatches`)
## What changes are included in this PR?
- Execution configuration options
`skip_partial_aggregation_probe_rows_threshold` and
`skip_partial_aggregation_probe_ratio_threshold` -- the first is responsible
for input rows to aggregate before checking aggregation ratio, the second --
for rate threshold
- `GroupedHashAggregateStream.skip_aggregation_probe` and related methods
for updating state / obtaining information if further input aggregation may be
skipped
- `GroupsAccumulator.convert_to_state`, and its implementations for
`PrimitiveGroupsAccumulator` (sum / min / max) and `Count` accumulators --
method responsible for converting `RecordBatch` to intermediate aggregate
state, without grouping input data, and
``GroupsAccumulator.convert_to_state_supported`, which indicates that
accumulator is able to perform conversion described above.
<!--
There is no need to duplicate the description in the issue here but it is
sometimes worth providing a summary of the individual changes in this PR.
-->
## Are these changes tested?
<!--
We typically require tests for all PRs in order to:
1. Prevent the code from being accidentally broken by subsequent changes
2. Serve as another way to document the expected behavior of the code
If tests are not included in your PR, please explain why (for example, are
they covered by existing tests)?
-->
Added tests for switching to `SkippingAggregation` state for aggregate
stream, and sqllogictests to validate correctness of accumulators in skipping
aggregation mode.
## Are there any user-facing changes?
<!--
If there are user-facing changes then we may require documentation to be
updated before approving the PR.
-->
No
<!--
If there are any breaking changes to public APIs, please add the `api
change` label.
-->
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For additional commands, e-mail: github-h...@datafusion.apache.org
