Skip to content

Instantly share code, notes, and snippets.

@quasiben

quasiben/benchmark_array-selene_a100_cuda112_condacupy860_cub_cutensor-results.txt

Created April 14, 2021 14:55
Show Gist options
  • Save quasiben/afd9d7c8461142b2162235e71387bc7a to your computer and use it in GitHub Desktop.
Save quasiben/afd9d7c8461142b2162235e71387bc7a to your computer and use it in GitHub Desktop.
Download ZIP
Raw
benchmark_array-selene_a100_cuda112_condacupy860_cub_cutensor-results.txt
--------------------------------------------------------------------------------------------------------------------- benchmark: 48 tests ----------------------------------------------------------------------------------------------------------------------
Name (time in us) Min Max Mean StdDev Median IQR Outliers OPS Rounds Iterations
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
test_Array_Slicing[shape0-cupy] 23.8450 (1.0) 68.3790 (2.03) 34.2448 (1.19) 19.1792 (8.08) 25.4280 (1.0) 14.3270 (5.18) 1;1 29,201.5031 (0.84) 5 1
test_Sum[shape0-cupy] 26.1990 (1.10) 33.7540 (1.0) 28.8906 (1.0) 3.0931 (1.30) 27.4720 (1.08) 4.2782 (1.55) 1;0 34,613.3295 (1.0) 5 1
test_FFT[shape0-cupy] 41.9300 (1.76) 436.1740 (12.92) 128.4188 (4.45) 172.5639 (72.72) 47.2290 (1.86) 122.6818 (44.37) 1;1 7,787.0225 (0.22) 5 1
test_Elementwise[shape0-cupy] 95.1290 (3.99) 129.7750 (3.84) 106.6370 (3.69) 14.2790 (6.02) 100.1290 (3.94) 18.8510 (6.82) 1;0 9,377.6077 (0.27) 5 1
test_Array_Slicing[shape0-numpy] 102.8250 (4.31) 430.8930 (12.77) 169.4020 (5.86) 146.1847 (61.60) 104.1770 (4.10) 84.6023 (30.59) 1;1 5,903.1176 (0.17) 5 1
test_Matrix_Multiplication[shape0-cupy] 145.0640 (6.08) 192.9750 (5.72) 155.6338 (5.39) 20.9038 (8.81) 146.5060 (5.76) 13.7597 (4.98) 1;1 6,425.3383 (0.19) 5 1
test_Sum[shape0-numpy] 230.1450 (9.65) 235.6950 (6.98) 232.8878 (8.06) 2.3731 (1.0) 232.3280 (9.14) 4.1303 (1.49) 2;0 4,293.9129 (0.12) 5 1
test_Array_Slicing[shape1-cupy] 371.9320 (15.60) 382.7930 (11.34) 376.7174 (13.04) 4.2745 (1.80) 377.6930 (14.85) 6.0280 (2.18) 2;0 2,654.5097 (0.08) 5 1
test_Stencil[shape0-cupy] 375.3590 (15.74) 483.3241 (14.32) 402.3442 (13.93) 45.6042 (19.22) 383.0340 (15.06) 35.1138 (12.70) 1;1 2,485.4340 (0.07) 5 1
test_Sum[shape1-cupy] 440.6120 (18.48) 448.5070 (13.29) 444.5292 (15.39) 2.9396 (1.24) 444.9800 (17.50) 3.8300 (1.39) 2;0 2,249.5710 (0.06) 5 1
test_Standard_Deviation[shape0-cupy] 739.4470 (31.01) 868.4100 (25.73) 768.1444 (26.59) 56.1304 (23.65) 744.5050 (29.28) 36.7858 (13.30) 1;1 1,301.8386 (0.04) 5 1
test_Standard_Deviation[shape0-numpy] 1,228.6400 (51.53) 1,375.1770 (40.74) 1,264.5398 (43.77) 62.1232 (26.18) 1,240.2120 (48.77) 43.7803 (15.83) 1;1 790.8015 (0.02) 5 1
test_Array_Slicing[shape2-cupy] 1,331.0430 (55.82) 1,387.0090 (41.09) 1,344.0798 (46.52) 24.0793 (10.15) 1,335.0110 (52.50) 16.8625 (6.10) 1;1 744.0034 (0.02) 5 1
test_Sum[shape2-cupy] 1,682.6770 (70.57) 1,692.1750 (50.13) 1,686.4506 (58.37) 3.4797 (1.47) 1,685.9140 (66.30) 2.7653 (1.0) 2;1 592.9613 (0.02) 5 1
test_Stencil[shape1-cupy] 2,376.6600 (99.67) 2,733.3230 (80.98) 2,452.9720 (84.91) 156.9077 (66.12) 2,381.2880 (93.65) 102.6830 (37.13) 1;1 407.6687 (0.01) 5 1
test_FFT[shape1-cupy] 3,971.0180 (166.53) 5,534.0800 (163.95) 4,286.8684 (148.38) 697.2337 (293.81) 3,973.2420 (156.25) 400.3250 (144.77) 1;1 233.2705 (0.01) 5 1
test_Stencil[shape0-numpy] 5,135.4620 (215.37) 5,162.8730 (152.96) 5,148.6406 (178.21) 13.0889 (5.52) 5,144.9800 (202.34) 24.9847 (9.04) 3;0 194.2260 (0.01) 5 1
test_Elementwise[shape1-cupy] 7,357.8630 (308.57) 7,381.8780 (218.70) 7,364.0722 (254.90) 10.0100 (4.22) 7,360.0760 (289.45) 6.6278 (2.40) 1;1 135.7944 (0.00) 5 1
test_Stencil[shape2-cupy] 8,732.9970 (366.24) 8,962.4710 (265.52) 8,801.4406 (304.65) 99.1884 (41.80) 8,744.1890 (343.88) 131.1280 (47.42) 1;0 113.6178 (0.00) 5 1
test_FFT[shape0-numpy] 8,734.4030 (366.30) 10,348.2190 (306.58) 9,378.1438 (324.61) 589.7605 (248.52) 9,276.2550 (364.80) 420.3080 (152.00) 2;1 106.6309 (0.00) 5 1
test_Matrix_Multiplication[shape0-numpy] 12,897.5020 (540.89) 22,765.3150 (674.45) 17,899.9636 (619.58) 4,394.6983 (>1000.0) 17,978.4350 (707.03) 8,134.2293 (>1000.0) 2;0 55.8660 (0.00) 5 1
test_FFT[shape2-cupy] 16,209.9460 (679.80) 21,305.8080 (631.21) 17,236.0336 (596.60) 2,275.0976 (958.70) 16,217.0990 (637.76) 1,292.9232 (467.56) 1;1 58.0180 (0.00) 5 1
test_Elementwise[shape0-numpy] 27,771.9960 (>1000.0) 28,641.0280 (848.52) 27,974.7296 (968.30) 374.4958 (157.81) 27,810.0680 (>1000.0) 285.5770 (103.27) 1;1 35.7465 (0.00) 5 1
test_Elementwise[shape2-cupy] 29,773.3460 (>1000.0) 29,797.3310 (882.78) 29,781.1506 (>1000.0) 9.4193 (3.97) 29,777.4230 (>1000.0) 8.8965 (3.22) 1;0 33.5783 (0.00) 5 1
test_Array_Slicing[shape1-numpy] 36,647.0250 (>1000.0) 45,033.3310 (>1000.0) 38,477.1356 (>1000.0) 3,667.2096 (>1000.0) 36,947.2030 (>1000.0) 2,214.2717 (800.74) 1;1 25.9895 (0.00) 5 1
test_Sum[shape1-numpy] 39,653.0200 (>1000.0) 42,129.5650 (>1000.0) 40,158.7058 (>1000.0) 1,101.8073 (464.29) 39,663.8090 (>1000.0) 636.6300 (230.22) 1;1 24.9012 (0.00) 5 1
test_Matrix_Multiplication[shape1-cupy] 113,231.5220 (>1000.0) 116,171.9040 (>1000.0) 114,400.8738 (>1000.0) 1,375.2765 (579.53) 113,800.2650 (>1000.0) 2,478.2215 (896.19) 1;0 8.7412 (0.00) 5 1
test_Standard_Deviation[shape1-cupy] 126,602.3380 (>1000.0) 126,742.0920 (>1000.0) 126,679.4118 (>1000.0) 62.2057 (26.21) 126,680.6360 (>1000.0) 113.7478 (41.13) 2;0 7.8939 (0.00) 5 1
test_Array_Slicing[shape2-numpy] 143,948.1390 (>1000.0) 144,565.9560 (>1000.0) 144,285.5806 (>1000.0) 256.1585 (107.94) 144,362.5320 (>1000.0) 422.5147 (152.79) 2;0 6.9307 (0.00) 5 1
test_Sum[shape2-numpy] 160,213.0490 (>1000.0) 160,405.6230 (>1000.0) 160,272.5858 (>1000.0) 80.9402 (34.11) 160,225.8840 (>1000.0) 100.3382 (36.29) 1;0 6.2394 (0.00) 5 1
test_Standard_Deviation[shape1-numpy] 257,765.0780 (>1000.0) 258,219.5160 (>1000.0) 257,917.4476 (>1000.0) 175.2828 (73.86) 257,860.2970 (>1000.0) 145.1763 (52.50) 1;1 3.8772 (0.00) 5 1
test_SVD[shape0-cupy] 349,500.6480 (>1000.0) 351,251.6440 (>1000.0) 350,185.4930 (>1000.0) 648.7546 (273.38) 350,086.5140 (>1000.0) 615.8282 (222.70) 2;0 2.8556 (0.00) 5 1
test_Standard_Deviation[shape2-cupy] 522,902.2950 (>1000.0) 523,047.3690 (>1000.0) 523,004.2354 (>1000.0) 59.1479 (24.92) 523,014.8580 (>1000.0) 59.2103 (21.41) 1;0 1.9120 (0.00) 5 1
test_SVD[shape1-cupy] 570,993.4460 (>1000.0) 573,269.9040 (>1000.0) 572,556.6632 (>1000.0) 959.4923 (404.32) 573,106.8960 (>1000.0) 1,225.0645 (443.02) 1;0 1.7466 (0.00) 5 1
test_SVD[shape0-numpy] 607,759.1050 (>1000.0) 717,584.5350 (>1000.0) 674,094.6362 (>1000.0) 41,854.2931 (>1000.0) 691,067.4360 (>1000.0) 48,933.9372 (>1000.0) 2;0 1.4835 (0.00) 5 1
test_Stencil[shape1-numpy] 675,651.9230 (>1000.0) 677,725.0190 (>1000.0) 676,997.8938 (>1000.0) 826.8699 (348.43) 677,131.9590 (>1000.0) 1,058.0940 (382.64) 1;0 1.4771 (0.00) 5 1
test_SVD[shape2-cupy] 687,790.0340 (>1000.0) 688,756.3290 (>1000.0) 688,205.1056 (>1000.0) 415.9957 (175.30) 688,041.8990 (>1000.0) 702.7787 (254.14) 1;0 1.4531 (0.00) 5 1
test_Standard_Deviation[shape2-numpy] 1,025,110.6170 (>1000.0) 1,025,409.2099 (>1000.0) 1,025,257.6924 (>1000.0) 107.5934 (45.34) 1,025,256.3220 (>1000.0) 118.5089 (42.86) 2;0 0.9754 (0.00) 5 1
test_Matrix_Multiplication[shape2-cupy] 1,072,740.3410 (>1000.0) 1,107,216.0880 (>1000.0) 1,095,655.4510 (>1000.0) 13,743.0598 (>1000.0) 1,099,032.7260 (>1000.0) 16,377.8923 (>1000.0) 1;0 0.9127 (0.00) 5 1
test_FFT[shape1-numpy] 1,133,962.7020 (>1000.0) 1,139,672.6750 (>1000.0) 1,136,271.0174 (>1000.0) 2,345.8061 (988.50) 1,135,998.1760 (>1000.0) 3,727.0908 (>1000.0) 1;0 0.8801 (0.00) 5 1
test_Stencil[shape2-numpy] 2,669,522.5710 (>1000.0) 2,700,417.0040 (>1000.0) 2,677,199.4276 (>1000.0) 13,108.1619 (>1000.0) 2,673,073.3190 (>1000.0) 10,556.6253 (>1000.0) 1;1 0.3735 (0.00) 5 1
test_Elementwise[shape1-numpy] 2,956,565.6590 (>1000.0) 2,958,248.8870 (>1000.0) 2,957,679.1736 (>1000.0) 698.6931 (294.42) 2,957,886.9130 (>1000.0) 990.5858 (358.22) 1;0 0.3381 (0.00) 5 1
test_Matrix_Multiplication[shape1-numpy] 3,408,981.7290 (>1000.0) 4,732,441.4140 (>1000.0) 4,219,186.5692 (>1000.0) 554,879.9142 (>1000.0) 4,388,464.6020 (>1000.0) 886,978.0014 (>1000.0) 1;0 0.2370 (0.00) 5 1
test_FFT[shape2-numpy] 5,082,444.5630 (>1000.0) 5,257,166.7550 (>1000.0) 5,119,828.0538 (>1000.0) 76,913.1020 (>1000.0) 5,083,306.6770 (>1000.0) 51,710.8175 (>1000.0) 1;1 0.1953 (0.00) 5 1
test_SVD[shape1-numpy] 9,877,809.0570 (>1000.0) 11,593,440.5390 (>1000.0) 10,808,625.1488 (>1000.0) 769,316.8001 (>1000.0) 11,133,236.5440 (>1000.0) 1,355,827.8395 (>1000.0) 2;0 0.0925 (0.00) 5 1
test_Elementwise[shape2-numpy] 11,824,573.2430 (>1000.0) 11,827,224.7000 (>1000.0) 11,826,016.4238 (>1000.0) 1,047.4587 (441.39) 11,826,100.8770 (>1000.0) 1,627.7310 (588.63) 2;0 0.0846 (0.00) 5 1
test_Matrix_Multiplication[shape2-numpy] 16,953,924.7560 (>1000.0) 27,210,432.3900 (>1000.0) 22,697,820.4338 (>1000.0) 4,922,681.0377 (>1000.0) 25,609,917.4810 (>1000.0) 8,719,820.0880 (>1000.0) 2;0 0.0441 (0.00) 5 1
test_SVD[shape2-numpy] 44,696,545.8730 (>1000.0) 55,508,688.9100 (>1000.0) 48,012,223.7540 (>1000.0) 4,514,649.7305 (>1000.0) 45,736,539.1640 (>1000.0) 5,577,422.9925 (>1000.0) 1;0 0.0208 (0.00) 5 1
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment