You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@datafusion.apache.org by "2010YOUY01 (via GitHub)" <gi...@apache.org> on 2025/10/31 11:59:13 UTC

[I] Support `reduction_fator` metric (for `EXPLAIN ANALYZE`) in `AggregateExec` [datafusion]

2010YOUY01 opened a new issue, #18410:
URL: https://github.com/apache/datafusion/issues/18410

   ### Is your feature request related to a problem or challenge?
   
   In the pre-aggregation stage (In datafusion's implementation, `AggregateExec` with `Partial` mode), a `reduction_factor` can be calculated as `output_rows (after partial aggregation) / input_rows` in `AggregateExec`
   
   Background for 'partial aggregation': https://github.com/apache/datafusion/blob/e65dafe7330f0324b28d9037f4e5a73cf12e99ce/datafusion/physical-plan/src/aggregates/row_hash.rs#L303
   
   ### Example with `datafusion-cli`
   ```
   > set datafusion.explain.analyze_level = summary;
   
   0 row(s) fetched.
   Elapsed 0.000 seconds.
   
   > create table t1(a int, b int);
   0 row(s) fetched.
   Elapsed 0.007 seconds.
   
   > insert into t1 values (1,10), (1, 20), (2,10), (2,30);
   +-------+
   | count |
   +-------+
   | 4     |
   +-------+
   1 row(s) fetched.
   Elapsed 0.011 seconds.
   
   > explain analyze select a, sum(b)
   from t1
   group by a;
   +-------------------+---------------------------------------------------------------------------------------------------------------------------------------------------+
   | plan_type         | plan                                                                                                                                              |
   +-------------------+---------------------------------------------------------------------------------------------------------------------------------------------------+
   | Plan with Metrics | AggregateExec: mode=FinalPartitioned, gby=[a@0 as a], aggr=[sum(t1.b)], metrics=[output_rows=2, elapsed_compute=4.67121ms, output_bytes=1088.0 B] |
   |                   |   CoalesceBatchesExec: target_batch_size=8192, metrics=[output_rows=2, elapsed_compute=17.875µs, output_bytes=192.0 KB]                           |
   |                   |     RepartitionExec: partitioning=Hash([a@0], 14), input_partitions=14, metrics=[]                                                                |
   |                   |       RepartitionExec: partitioning=RoundRobinBatch(14), input_partitions=1, metrics=[]                                                           |
   |                   |         AggregateExec: mode=Partial, gby=[a@0 as a], aggr=[sum(t1.b)], metrics=[output_rows=2, elapsed_compute=1.931542ms, output_bytes=544.0 B]  |
   |                   |           DataSourceExec: partitions=1, partition_sizes=[1], metrics=[]                                                                           |
   |                   |                                                                                                                                                   |
   +-------------------+---------------------------------------------------------------------------------------------------------------------------------------------------+
   1 row(s) fetched.
   Elapsed 0.018 seconds.
   ```
   
   You can see for the bottom `AggregateExec` with `Partial` mode, it has 4 input rows, and after the pre-aggregation, the output size has been reduced to 2 rows, so a `reduction_factor` metric can be calculated as `2/4 => 50%`
   
   ### Describe the solution you'd like
   
   Add `reduction_factor` metric for `AggregateExec` with `Partial` mode.
   
   Reference PR: https://github.com/apache/datafusion/pull/18406
   
   ### Describe alternatives you've considered
   
   _No response_
   
   ### Additional context
   
   _No response_


-- 
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-unsubscribe@datafusion.apache.org.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org


Re: [I] Support `reduction_fator` metric (for `EXPLAIN ANALYZE`) in `AggregateExec` [datafusion]

Posted by "petern48 (via GitHub)" <gi...@apache.org>.
petern48 commented on issue #18410:
URL: https://github.com/apache/datafusion/issues/18410#issuecomment-3478593734

   take


-- 
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-unsubscribe@datafusion.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org


Re: [I] Support `reduction_fator` metric (for `EXPLAIN ANALYZE`) in `AggregateExec` [datafusion]

Posted by "alamb (via GitHub)" <gi...@apache.org>.
alamb closed issue #18410: Support `reduction_fator` metric (for `EXPLAIN ANALYZE`) in `AggregateExec`
URL: https://github.com/apache/datafusion/issues/18410


-- 
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-unsubscribe@datafusion.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org