Skip to content

Instantly share code, notes, and snippets.

@asi1024
Created December 4, 2019 09:32
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save asi1024/563e47924533196d130df35ce04388bd to your computer and use it in GitHub Desktop.
Save asi1024/563e47924533196d130df35ce04388bd to your computer and use it in GitHub Desktop.
cupy.sum (shape = ( 1, 16777216), axis=0): 13.978 us +/- 0.560 (min: 13.467 / max: 21.584) us 186.376 us +/- 0.719 (min: 177.312 / max: 193.696) us
cupy.sum (shape = ( 2, 8388608), axis=0): 17.943 us +/-76.250 (min: 13.828 / max: 1720.384) us 181.624 us +/-73.942 (min: 168.544 / max: 1832.480) us
cupy.sum (shape = ( 4, 4194304), axis=0): 14.479 us +/- 0.561 (min: 13.962 / max: 22.706) us 187.462 us +/- 0.741 (min: 177.504 / max: 195.456) us
cupy.sum (shape = ( 8, 2097152), axis=0): 14.375 us +/- 0.522 (min: 13.851 / max: 21.145) us 199.112 us +/- 0.744 (min: 188.768 / max: 205.216) us
cupy.sum (shape = ( 16, 1048576), axis=0): 14.507 us +/- 0.473 (min: 13.974 / max: 19.830) us 213.054 us +/- 0.664 (min: 203.136 / max: 218.048) us
cupy.sum (shape = ( 32, 524288), axis=0): 14.273 us +/- 0.384 (min: 13.735 / max: 18.258) us 142.936 us +/- 0.606 (min: 133.408 / max: 146.912) us
cupy.sum (shape = ( 64, 262144), axis=0): 14.538 us +/- 0.639 (min: 14.056 / max: 26.289) us 119.616 us +/- 0.888 (min: 109.696 / max: 131.392) us
cupy.sum (shape = ( 128, 131072), axis=0): 14.468 us +/- 0.402 (min: 14.029 / max: 18.161) us 108.068 us +/- 0.655 (min: 98.208 / max: 112.000) us
cupy.sum (shape = ( 256, 65536), axis=0): 14.423 us +/- 0.462 (min: 13.854 / max: 21.458) us 103.478 us +/- 0.711 (min: 93.696 / max: 110.048) us
cupy.sum (shape = ( 512, 32768), axis=0): 14.271 us +/- 0.613 (min: 13.821 / max: 23.596) us 106.380 us +/- 0.821 (min: 97.696 / max: 115.456) us
cupy.sum (shape = ( 1024, 16384), axis=0): 14.407 us +/- 0.631 (min: 13.903 / max: 26.125) us 105.622 us +/- 0.855 (min: 96.672 / max: 116.192) us
cupy.sum (shape = ( 2048, 8192), axis=0): 14.382 us +/- 0.523 (min: 13.862 / max: 21.998) us 103.210 us +/- 0.882 (min: 93.248 / max: 108.320) us
cupy.sum (shape = ( 4096, 4096), axis=0): 14.435 us +/- 0.326 (min: 13.933 / max: 16.864) us 146.204 us +/- 0.943 (min: 137.312 / max: 148.896) us
cupy.sum (shape = ( 8192, 2048), axis=0): 14.446 us +/- 0.550 (min: 13.825 / max: 21.506) us 230.483 us +/- 1.560 (min: 220.544 / max: 237.952) us
cupy.sum (shape = ( 16384, 1024), axis=0): 14.491 us +/- 0.551 (min: 13.810 / max: 22.311) us 429.061 us +/- 1.524 (min: 418.816 / max: 436.576) us
cupy.sum (shape = ( 32768, 512), axis=0): 14.534 us +/- 0.447 (min: 13.780 / max: 17.911) us 460.629 us +/- 1.925 (min: 453.344 / max: 466.464) us
cupy.sum (shape = ( 65536, 256), axis=0): 15.136 us +/- 9.876 (min: 14.085 / max: 235.215) us 460.450 us +/- 9.986 (min: 450.880 / max: 680.384) us
cupy.sum (shape = ( 131072, 128), axis=0): 14.544 us +/- 0.508 (min: 13.985 / max: 19.066) us 431.472 us +/- 3.070 (min: 421.824 / max: 445.056) us
cupy.sum (shape = ( 262144, 64), axis=0): 14.663 us +/- 0.577 (min: 13.914 / max: 21.018) us 447.116 us +/- 2.501 (min: 439.104 / max: 457.696) us
cupy.sum (shape = ( 524288, 32), axis=0): 14.543 us +/- 0.486 (min: 13.951 / max: 18.541) us 489.926 us +/- 1.843 (min: 479.456 / max: 498.240) us
cupy.sum (shape = ( 1048576, 16), axis=0): 14.762 us +/- 0.578 (min: 14.033 / max: 19.189) us 946.685 us +/- 1.997 (min: 936.128 / max: 953.600) us
cupy.sum (shape = ( 2097152, 8), axis=0): 15.462 us +/- 6.250 (min: 14.283 / max: 154.249) us 1900.120 us +/-14.003 (min: 1871.360 / max: 2009.824) us
cupy.sum (shape = ( 4194304, 4), axis=0): 15.928 us +/- 9.999 (min: 14.354 / max: 237.904) us 3154.846 us +/-10.603 (min: 3140.640 / max: 3379.744) us
cupy.sum (shape = ( 8388608, 2), axis=0): 16.689 us +/-13.111 (min: 14.536 / max: 225.868) us 6127.232 us +/-14.433 (min: 6110.176 / max: 6339.072) us
cupy.sum (shape = (16777216, 1), axis=0): 17.029 us +/-12.984 (min: 14.719 / max: 228.357) us 12185.731 us +/-19.733 (min:12139.776 / max:12388.960) us
cupy.sum (shape = ( 1, 16777216), axis=1): 17.937 us +/- 9.507 (min: 15.564 / max: 229.891) us 12186.025 us +/-17.802 (min:12121.152 / max:12371.968) us
cupy.sum (shape = ( 2, 8388608), axis=1): 16.557 us +/- 0.516 (min: 15.538 / max: 20.564) us 7414.710 us +/- 8.782 (min: 7395.040 / max: 7445.568) us
cupy.sum (shape = ( 4, 4194304), axis=1): 16.786 us +/- 9.270 (min: 15.140 / max: 223.227) us 3737.826 us +/-10.453 (min: 3724.704 / max: 3948.768) us
cupy.sum (shape = ( 8, 2097152), axis=1): 15.842 us +/- 0.650 (min: 14.942 / max: 20.441) us 1894.013 us +/- 3.188 (min: 1877.344 / max: 1904.416) us
cupy.sum (shape = ( 16, 1048576), axis=1): 15.453 us +/- 0.545 (min: 14.730 / max: 19.252) us 962.182 us +/- 2.073 (min: 951.040 / max: 968.928) us
cupy.sum (shape = ( 32, 524288), axis=1): 15.312 us +/- 0.513 (min: 14.707 / max: 19.323) us 499.570 us +/- 1.506 (min: 486.816 / max: 504.160) us
cupy.sum (shape = ( 64, 262144), axis=1): 15.197 us +/- 0.406 (min: 14.704 / max: 18.649) us 261.719 us +/- 0.999 (min: 251.552 / max: 266.016) us
cupy.sum (shape = ( 128, 131072), axis=1): 15.113 us +/- 0.557 (min: 14.521 / max: 22.043) us 163.623 us +/- 1.207 (min: 153.312 / max: 168.352) us
cupy.sum (shape = ( 256, 65536), axis=1): 15.168 us +/- 0.584 (min: 14.678 / max: 23.104) us 156.811 us +/- 1.082 (min: 145.664 / max: 163.584) us
cupy.sum (shape = ( 512, 32768), axis=1): 15.066 us +/- 0.407 (min: 14.544 / max: 19.159) us 156.532 us +/- 1.413 (min: 145.408 / max: 161.184) us
cupy.sum (shape = ( 1024, 16384), axis=1): 15.259 us +/- 0.399 (min: 14.659 / max: 19.386) us 146.860 us +/- 0.781 (min: 136.000 / max: 151.552) us
cupy.sum (shape = ( 2048, 8192), axis=1): 15.183 us +/- 0.371 (min: 14.570 / max: 17.863) us 149.854 us +/- 0.645 (min: 139.296 / max: 152.640) us
cupy.sum (shape = ( 4096, 4096), axis=1): 15.264 us +/- 0.418 (min: 14.700 / max: 19.956) us 171.823 us +/- 0.691 (min: 160.960 / max: 176.256) us
cupy.sum (shape = ( 8192, 2048), axis=1): 15.192 us +/- 0.482 (min: 14.663 / max: 22.609) us 203.890 us +/- 0.705 (min: 193.120 / max: 211.040) us
cupy.sum (shape = ( 16384, 1024), axis=1): 15.291 us +/- 0.714 (min: 14.689 / max: 24.315) us 283.284 us +/- 0.788 (min: 272.832 / max: 291.776) us
cupy.sum (shape = ( 32768, 512), axis=1): 15.270 us +/- 0.552 (min: 14.607 / max: 19.550) us 449.926 us +/- 0.697 (min: 439.360 / max: 453.920) us
cupy.sum (shape = ( 65536, 256), axis=1): 15.354 us +/- 0.623 (min: 14.719 / max: 23.115) us 424.833 us +/- 0.699 (min: 414.144 / max: 429.152) us
cupy.sum (shape = ( 131072, 128), axis=1): 15.408 us +/- 0.908 (min: 14.726 / max: 26.606) us 404.224 us +/- 0.963 (min: 393.728 / max: 413.568) us
cupy.sum (shape = ( 262144, 64), axis=1): 15.289 us +/- 1.052 (min: 14.547 / max: 34.206) us 390.180 us +/- 1.111 (min: 379.520 / max: 407.648) us
cupy.sum (shape = ( 524288, 32), axis=1): 15.306 us +/- 0.970 (min: 14.581 / max: 29.031) us 387.451 us +/- 1.021 (min: 376.992 / max: 401.056) us
cupy.sum (shape = ( 1048576, 16), axis=1): 15.101 us +/- 0.418 (min: 14.558 / max: 19.304) us 368.273 us +/- 0.660 (min: 357.952 / max: 371.616) us
cupy.sum (shape = ( 2097152, 8), axis=1): 15.549 us +/- 9.895 (min: 14.421 / max: 236.362) us 329.150 us +/-10.018 (min: 318.496 / max: 552.384) us
cupy.sum (shape = ( 4194304, 4), axis=1): 15.281 us +/- 0.713 (min: 14.696 / max: 26.524) us 302.469 us +/- 0.822 (min: 292.192 / max: 311.968) us
cupy.sum (shape = ( 8388608, 2), axis=1): 15.216 us +/- 0.617 (min: 14.655 / max: 24.024) us 288.715 us +/- 0.790 (min: 278.336 / max: 296.128) us
cupy.sum (shape = (16777216, 1), axis=1): 15.028 us +/- 0.910 (min: 14.439 / max: 29.776) us 187.542 us +/- 0.977 (min: 176.544 / max: 200.544) us
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment