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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to