GPU:            Tesla P100-PCIE-12GB
Block size:     512
--------------------------------------

Code version:   V0
        Total              Mem-xch            Sorting           
Q20:    6559 [usec]        1147 [usec]        5419 [usec]       
Q21:    12 [msec]          2031 [usec]        10 [msec]         
Q22:    23 [msec]          3801 [usec]        19 [msec]         
Q23:    49 [msec]          8534 [usec]        40 [msec]         
Q24:    101 [msec]         15 [msec]          85 [msec]         
Q25:    210 [msec]         29 [msec]          181 [msec]        
Q26:    442 [msec]         57 [msec]          385 [msec]        
Q27:    939 [msec]         112 [msec]         826 [msec]        
Q28:    1986 [msec]        223 [msec]         1762 [msec]       
Q29:    4211 [msec]        446 [msec]         3765 [msec]       
Q30:    9071 [msec]        1037 [msec]        8034 [msec]       

Code version:   V1
        Total              Mem-xch            Sorting           
Q20:    4559 [usec]        1144 [usec]        3418 [usec]       
Q21:    9014 [usec]        2018 [usec]        6995 [usec]       
Q22:    18 [msec]          3821 [usec]        14 [msec]         
Q23:    39 [msec]          8564 [usec]        30 [msec]         
Q24:    80 [msec]          15 [msec]          64 [msec]         
Q25:    169 [msec]         29 [msec]          139 [msec]        
Q26:    358 [msec]         57 [msec]          300 [msec]        
Q27:    761 [msec]         113 [msec]         648 [msec]        
Q28:    1618 [msec]        224 [msec]         1394 [msec]       
Q29:    3440 [msec]        447 [msec]         2992 [msec]       
Q30:    7326 [msec]        914 [msec]         6411 [msec]       

Code version:   V2
        Total              Mem-xch            Sorting           
Q20:    4407 [usec]        1146 [usec]        3264 [usec]       
Q21:    8751 [usec]        2025 [usec]        6728 [usec]       
Q22:    17 [msec]          3818 [usec]        13 [msec]         
Q23:    38 [msec]          8515 [usec]        29 [msec]         
Q24:    75 [msec]          15 [msec]          60 [msec]         
Q25:    159 [msec]         29 [msec]          129 [msec]        
Q26:    338 [msec]         58 [msec]          280 [msec]        
Q27:    719 [msec]         113 [msec]         606 [msec]        
Q28:    1533 [msec]        225 [msec]         1308 [msec]       
Q29:    3262 [msec]        447 [msec]         2815 [msec]       
Q30:    6932 [msec]        883 [msec]         6048 [msec]       
