tustvold commented on PR #2231:
URL: https://github.com/apache/arrow-rs/pull/2231#issuecomment-1200125233

   Benchmarks
   
   ```
   write_batch primitive/4096 values primitive                                  
                                           
                           time:   [1.4416 ms 1.4419 ms 1.4422 ms]
                           thrpt:  [122.33 MiB/s 122.35 MiB/s 122.37 MiB/s]
                    change:
                           time:   [-22.390% -22.341% -22.291%] (p = 0.00 < 
0.05)
                           thrpt:  [+28.684% +28.767% +28.849%]
                           Performance has improved.
   Found 2 outliers among 100 measurements (2.00%)
     1 (1.00%) high mild
     1 (1.00%) high severe
   Benchmarking write_batch primitive/4096 values primitive non-null: Warming 
up for 3.0000 s
   Warning: Unable to complete 100 samples in 5.0s. You may wish to increase 
target time to 7.1s, enable flat sampling, or reduce sample count to 50.
   write_batch primitive/4096 values primitive non-null                         
                                                    
                           time:   [1.4112 ms 1.4119 ms 1.4127 ms]
                           thrpt:  [122.46 MiB/s 122.53 MiB/s 122.59 MiB/s]
                    change:
                           time:   [-9.7045% -9.6526% -9.6050%] (p = 0.00 < 
0.05)
                           thrpt:  [+10.626% +10.684% +10.748%]
                           Performance has improved.
   Found 15 outliers among 100 measurements (15.00%)
     7 (7.00%) low mild
     6 (6.00%) high mild
     2 (2.00%) high severe
   write_batch primitive/4096 values bool                                       
                                     
                           time:   [101.05 us 101.09 us 101.14 us]
                           thrpt:  [11.240 MiB/s 11.245 MiB/s 11.250 MiB/s]
                    change:
                           time:   [-3.9386% -3.8687% -3.7961%] (p = 0.00 < 
0.05)
                           thrpt:  [+3.9459% +4.0244% +4.1001%]
                           Performance has improved.
   Found 7 outliers among 100 measurements (7.00%)
     1 (1.00%) low severe
     2 (2.00%) high mild
     4 (4.00%) high severe
   write_batch primitive/4096 values bool non-null                              
                                               
                           time:   [43.685 us 43.707 us 43.728 us]
                           thrpt:  [14.830 MiB/s 14.837 MiB/s 14.845 MiB/s]
                    change:
                           time:   [+3.4561% +3.5779% +3.7335%] (p = 0.00 < 
0.05)
                           thrpt:  [-3.5992% -3.4543% -3.3406%]
                           Performance has regressed.
   Found 6 outliers among 100 measurements (6.00%)
     2 (2.00%) low mild
     1 (1.00%) high mild
     3 (3.00%) high severe
   write_batch primitive/4096 values string                                     
                                       
                           time:   [844.97 us 845.63 us 846.37 us]
                           thrpt:  [94.037 MiB/s 94.119 MiB/s 94.192 MiB/s]
                    change:
                           time:   [-14.692% -14.560% -14.353%] (p = 0.00 < 
0.05)
                           thrpt:  [+16.758% +17.041% +17.222%]
                           Performance has improved.
   Found 6 outliers among 100 measurements (6.00%)
     2 (2.00%) high mild
     4 (4.00%) high severe
   write_batch primitive/4096 values string non-null                            
                                                
                           time:   [961.22 us 961.63 us 962.07 us]
                           thrpt:  [81.713 MiB/s 81.750 MiB/s 81.785 MiB/s]
                    change:
                           time:   [-6.1525% -6.0546% -5.9641%] (p = 0.00 < 
0.05)
                           thrpt:  [+6.3424% +6.4448% +6.5558%]
                           Performance has improved.
   Found 1 outliers among 100 measurements (1.00%)
     1 (1.00%) high severe
   
   Benchmarking write_batch nested/4096 values primitive list: Warming up for 
3.0000 s
   Warning: Unable to complete 100 samples in 5.0s. You may wish to increase 
target time to 9.1s, enable flat sampling, or reduce sample count to 50.
   write_batch nested/4096 values primitive list                                
                                             
                           time:   [1.7915 ms 1.7917 ms 1.7919 ms]
                           thrpt:  [91.399 MiB/s 91.411 MiB/s 91.423 MiB/s]
                    change:
                           time:   [-6.7029% -6.6786% -6.6536%] (p = 0.00 < 
0.05)
                           thrpt:  [+7.1279% +7.1566% +7.1845%]
                           Performance has improved.
   Found 13 outliers among 100 measurements (13.00%)
     2 (2.00%) low severe
     6 (6.00%) high mild
     5 (5.00%) high severe
   write_batch nested/4096 values primitive list non-null                       
                                                      
                           time:   [2.1664 ms 2.1672 ms 2.1680 ms]
                           thrpt:  [87.847 MiB/s 87.878 MiB/s 87.910 MiB/s]
                    change:
                           time:   [-9.0613% -9.0201% -8.9817%] (p = 0.00 < 
0.05)
                           thrpt:  [+9.8680% +9.9144% +9.9642%]
                           Performance has improved.
   Found 15 outliers among 100 measurements (15.00%)
     15 (15.00%) low mild
   ```
   
   So a roughly 10% performance improvement, likely from not needing to 
zero-allocate arrays


-- 
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]

Reply via email to