Dandandan opened a new pull request #8837:
URL: https://github.com/apache/arrow/pull/8837


   This PR speeds up the (non-simd) comparison kernels by ~8 times:
   
   ```
   eq Float32              time:   [105.96 us 106.01 us 106.06 us]              
         
                           change: [-82.463% -82.423% -82.378%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 8 outliers among 100 measurements (8.00%)
     1 (1.00%) low severe
     4 (4.00%) high mild
     3 (3.00%) high severe
   
   eq scalar Float32       time:   [61.439 us 61.530 us 61.662 us]              
                
                           change: [-88.282% -88.255% -88.221%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 10 outliers among 100 measurements (10.00%)
     1 (1.00%) low mild
     4 (4.00%) high mild
     5 (5.00%) high severe
   
   neq Float32             time:   [71.018 us 71.080 us 71.144 us]              
          
                           change: [-86.580% -86.563% -86.546%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 15 outliers among 100 measurements (15.00%)
     2 (2.00%) low severe
     4 (4.00%) low mild
     5 (5.00%) high mild
     4 (4.00%) high severe
   
   neq scalar Float32      time:   [68.706 us 68.773 us 68.838 us]              
                 
                           change: [-86.207% -86.188% -86.171%] (p = 0.00 < 
0.05)
                           Performance has improved.
   
   lt Float32              time:   [70.655 us 70.703 us 70.753 us]              
         
                           change: [-85.629% -85.617% -85.604%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 1 outliers among 100 measurements (1.00%)
     1 (1.00%) high mild
   
   lt scalar Float32       time:   [50.626 us 50.664 us 50.698 us]              
                 
                           change: [-89.802% -89.764% -89.731%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 2 outliers among 100 measurements (2.00%)
     2 (2.00%) high mild
   
   lt_eq Float32           time:   [101.34 us 101.43 us 101.51 us]              
            
                           change: [-82.825% -82.797% -82.767%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 4 outliers among 100 measurements (4.00%)
     2 (2.00%) high mild
     2 (2.00%) high severe
   
   lt_eq scalar Float32    time:   [68.894 us 68.913 us 68.933 us]              
                   
                           change: [-86.575% -86.557% -86.538%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 3 outliers among 100 measurements (3.00%)
     1 (1.00%) low severe
     2 (2.00%) high mild
   
   gt Float32              time:   [71.260 us 71.332 us 71.400 us]              
         
                           change: [-87.481% -87.450% -87.418%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 3 outliers among 100 measurements (3.00%)
     3 (3.00%) high mild
   
   gt scalar Float32       time:   [38.852 us 38.888 us 38.929 us]              
                 
                           change: [-91.745% -91.733% -91.721%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 13 outliers among 100 measurements (13.00%)
     1 (1.00%) low severe
     11 (11.00%) high mild
     1 (1.00%) high severe
   
   gt_eq Float32           time:   [99.404 us 99.451 us 99.503 us]              
            
                           change: [-80.870% -80.848% -80.827%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 1 outliers among 100 measurements (1.00%)
     1 (1.00%) high mild
   
   gt_eq scalar Float32    time:   [55.892 us 55.926 us 55.963 us]              
                   
                           change: [-88.783% -88.751% -88.727%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 6 outliers among 100 measurements (6.00%)
     1 (1.00%) low severe
     1 (1.00%) low mild
     2 (2.00%) high mild
     2 (2.00%) high severe
   ```


----------------------------------------------------------------
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.

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


Reply via email to