zhuqi-lucas commented on issue #7363:
URL: https://github.com/apache/arrow-rs/issues/7363#issuecomment-2816591518

   When i increase the data set from 100000 to 1000000, there were regressions 
about the PointLookup:
   
   And i try to fix it with this commit:
   
https://github.com/apache/arrow-rs/commit/0c3aa9b3d4fa913029742f36b650f7b1e1ea2339
   branch polish_better_decoder
   https://github.com/apache/arrow-rs/pull/7428
   
   
   It seems it will fix the regression about PointLookup with big data set:
   
   ```rust
   critcmp better-decoder polish_better_decoder
   group                                                                        
         better-decoder                         polish_better_decoder
   -----                                                                        
         --------------                         ---------------------
   arrow_reader_row_filter/Composite/all_columns/async                          
         1.03     12.9±0.14ms        ? ?/sec    1.00     12.6±0.12ms        ? 
?/sec
   arrow_reader_row_filter/Composite/all_columns/sync                           
         1.00     14.5±0.10ms        ? ?/sec    1.00     14.6±0.31ms        ? 
?/sec
   arrow_reader_row_filter/Composite/exclude_filter_column/async                
         1.03     12.7±0.44ms        ? ?/sec    1.00     12.4±0.20ms        ? 
?/sec
   arrow_reader_row_filter/Composite/exclude_filter_column/sync                 
         1.02     13.3±0.32ms        ? ?/sec    1.00     13.1±0.21ms        ? 
?/sec
   arrow_reader_row_filter/ModeratelySelectiveClustered/all_columns/async       
         1.00     12.3±0.11ms        ? ?/sec    1.00     12.3±0.16ms        ? 
?/sec
   arrow_reader_row_filter/ModeratelySelectiveClustered/all_columns/sync        
         1.00     13.1±0.12ms        ? ?/sec    1.00     13.2±0.14ms        ? 
?/sec
   
arrow_reader_row_filter/ModeratelySelectiveClustered/exclude_filter_column/async
      1.01     12.3±0.44ms        ? ?/sec    1.00     12.1±0.17ms        ? ?/sec
   
arrow_reader_row_filter/ModeratelySelectiveClustered/exclude_filter_column/sync 
      1.00     12.7±0.10ms        ? ?/sec    1.01     12.8±0.17ms        ? ?/sec
   arrow_reader_row_filter/ModeratelySelectiveUnclustered/all_columns/async     
         1.01     41.7±0.59ms        ? ?/sec    1.00     41.4±0.51ms        ? 
?/sec
   arrow_reader_row_filter/ModeratelySelectiveUnclustered/all_columns/sync      
         1.00     41.7±0.34ms        ? ?/sec    1.00     41.6±0.36ms        ? 
?/sec
   
arrow_reader_row_filter/ModeratelySelectiveUnclustered/exclude_filter_column/async
    1.00     33.9±0.28ms        ? ?/sec    1.00     33.8±0.29ms        ? ?/sec
   
arrow_reader_row_filter/ModeratelySelectiveUnclustered/exclude_filter_column/sync
     1.00     34.3±0.28ms        ? ?/sec    1.00     34.2±0.33ms        ? ?/sec
   arrow_reader_row_filter/PointLookup/all_columns/async                        
         3.89     10.2±0.07ms        ? ?/sec    1.00      2.6±0.03ms        ? 
?/sec
   arrow_reader_row_filter/PointLookup/all_columns/sync                         
         1.02      2.9±0.14ms        ? ?/sec    1.00      2.8±0.04ms        ? 
?/sec
   arrow_reader_row_filter/PointLookup/exclude_filter_column/async              
         3.91     10.2±0.08ms        ? ?/sec    1.00      2.6±0.04ms        ? 
?/sec
   arrow_reader_row_filter/PointLookup/exclude_filter_column/sync               
         1.00      2.8±0.03ms        ? ?/sec    1.00      2.8±0.05ms        ? 
?/sec
   arrow_reader_row_filter/SelectiveUnclustered/all_columns/async               
         1.00     15.1±0.56ms        ? ?/sec    1.01     15.2±0.23ms        ? 
?/sec
   arrow_reader_row_filter/SelectiveUnclustered/all_columns/sync                
         1.00     16.3±0.21ms        ? ?/sec    1.01     16.5±0.18ms        ? 
?/sec
   arrow_reader_row_filter/SelectiveUnclustered/exclude_filter_column/async     
         1.00     14.0±0.16ms        ? ?/sec    1.01     14.1±0.15ms        ? 
?/sec
   arrow_reader_row_filter/SelectiveUnclustered/exclude_filter_column/sync      
         1.01     14.8±0.54ms        ? ?/sec    1.00     14.6±0.13ms        ? 
?/sec
   arrow_reader_row_filter/UnselectiveClustered/all_columns/async               
         1.00     20.6±0.14ms        ? ?/sec    1.03     21.3±0.25ms        ? 
?/sec
   arrow_reader_row_filter/UnselectiveClustered/all_columns/sync                
         1.00     21.8±0.19ms        ? ?/sec    1.03     22.4±0.17ms        ? 
?/sec
   arrow_reader_row_filter/UnselectiveClustered/exclude_filter_column/async     
         1.00     19.8±0.24ms        ? ?/sec    1.02     20.2±0.18ms        ? 
?/sec
   arrow_reader_row_filter/UnselectiveClustered/exclude_filter_column/sync      
         1.00     20.5±0.22ms        ? ?/sec    1.03     21.1±0.16ms        ? 
?/sec
   arrow_reader_row_filter/UnselectiveUnclustered/all_columns/async             
         1.00     15.1±0.13ms        ? ?/sec    1.01     15.2±0.17ms        ? 
?/sec
   arrow_reader_row_filter/UnselectiveUnclustered/all_columns/sync              
         1.00     16.5±0.15ms        ? ?/sec    1.00     16.4±0.14ms        ? 
?/sec
   arrow_reader_row_filter/UnselectiveUnclustered/exclude_filter_column/async   
         1.01     14.1±0.13ms        ? ?/sec    1.00     14.0±0.18ms        ? 
?/sec
   arrow_reader_row_filter/UnselectiveUnclustered/exclude_filter_column/sync    
         1.00     14.6±0.10ms        ? ?/sec    1.00     14.6±0.18ms        ? 
?/sec
   arrow_reader_row_filter/Utf8ViewNonEmpty/all_columns/async                   
         1.00     95.9±0.73ms        ? ?/sec    1.00     95.8±0.94ms        ? 
?/sec
   arrow_reader_row_filter/Utf8ViewNonEmpty/all_columns/sync                    
         1.01    104.4±0.78ms        ? ?/sec    1.00    103.1±1.08ms        ? 
?/sec
   arrow_reader_row_filter/Utf8ViewNonEmpty/exclude_filter_column/async         
         1.01     72.7±0.44ms        ? ?/sec    1.00     71.8±0.57ms        ? 
?/sec
   arrow_reader_row_filter/Utf8ViewNonEmpty/exclude_filter_column/sync          
         1.02     73.6±0.94ms        ? ?/sec    1.00     72.1±0.74ms        ? 
?/sec
   ```


-- 
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...@arrow.apache.org

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

Reply via email to