Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Ogl 170rev1 2024-01-10 Single node results #5

Open
wants to merge 132 commits into
base: main
Choose a base branch
from

Conversation

greole
Copy link
Contributor

@greole greole commented Jan 10, 2024

Description

This PR adds updated LidDrivenCavity3D results for OGL after updating the ginkgo communicating send-idxs only once. This these results are an update of #4

Copy link
Contributor

github-actions bot commented Jan 10, 2024

Results Overview

Unpreconditioned

nCells nProcs nNodes executor Host TimeStep SolveP MomentumPredictor PISOStep
0 1e+06 4 1 cuda hkn 1269.35 404.205 290.796 472.852
1 1e+06 4 1 dpcpp i20 1113.79 433.898 147.816 473.307
2 1e+06 8 1 CPU nla 1423.91 633.234 104.167 655.429
3 1e+06 8 1 cuda hkn 3594.89 1668.5 158.604 1708.63
4 1e+06 8 1 dpcpp i20 907.562 390.098 72.4849 411.667
5 1e+06 8 1 hip nla 656.757 246.474 109.465 269.448
6 1e+06 12 1 dpcpp i20 8960.13 4433.46 51.2537 4450.36
7 1e+06 16 1 CPU nla 772.45 345.464 52.476 357.822
8 1e+06 16 1 cuda hkn 5049.55 2444.05 93.8783 2471.07
9 1e+06 16 1 hip nla 595.604 256.157 53.1675 268.893
10 1e+06 32 1 CPU nla 635.559 283.43 41.3389 295.056
11 1e+06 32 1 hip nla 765.425 347.096 43.0688 359.036
12 1e+06 76 1 CPU hkn 271.542 112.655 21.0083 122.227
13 1e+06 76 1 cuda hkn 26536.4 13242.5 21.3131 13254.2
14 1e+06 112 1 CPU i20 107.449 45.1393 8.39905 48.6053
41 8e+06 4 1 cuda hkn 9402.53 2521.61 2693.37 3202.8
42 8e+06 4 1 dpcpp i20 11631.5 4517.5 1596.74 4922.65
43 8e+06 8 1 CPU nla 37805.3 18242.3 868.501 18430.8
44 8e+06 8 1 cuda hkn 11551.6 4604.98 1471.47 4959.28
45 8e+06 8 1 dpcpp i20 5402.73 2047.13 836.096 2241.22
46 8e+06 8 1 hip nla 3600.11 1133.99 876.537 1324.11
47 8e+06 12 1 dpcpp i20 28371.3 13706.7 602.292 13853.2
48 8e+06 16 1 CPU nla 17323 8265.18 486.774 8388.9
49 8e+06 16 1 cuda hkn 16221.8 7335.83 940.148 7580.95
50 8e+06 16 1 hip nla 3313.73 1241.92 518.398 1369.24
51 8e+06 32 1 CPU nla 15157.9 7200.33 450.805 7324.15
52 8e+06 32 1 hip nla 4995.5 2116.44 460.524 2238.17
53 8e+06 76 1 CPU hkn 11815.4 5577.07 373.771 5693.28
54 8e+06 76 1 cuda hkn 60449.7 29889.5 372.341 30010.2
55 8e+06 112 1 CPU i20 4872.52 2276.31 189.841 2327.85
82 2.7e+07 4 1 cuda hkn 37299.3 9662.82 11981.9 12126.9
83 2.7e+07 4 1 dpcpp i20 48272.4 19399.2 5831.74 20870.6
84 2.7e+07 8 1 CPU nla 203598 99571.6 2964.8 100195
85 2.7e+07 8 1 cuda hkn 31311.6 11245.2 5380.76 12650.7
86 2.7e+07 8 1 dpcpp i20 30883.1 13084.4 2845.21 13846.7
87 2.7e+07 8 1 hip nla 11233.5 3383.82 2969.25 4009.19
88 2.7e+07 12 1 dpcpp i20 57502.9 27043.9 2042.86 27601.8
89 2.7e+07 16 1 CPU nla 126516 61838.9 1763.65 62275
90 2.7e+07 16 1 cuda hkn 34505.3 14382 3404.39 15324
91 2.7e+07 16 1 hip nla 9229.68 3191.64 1770.65 3627.78
92 2.7e+07 32 1 CPU nla 118888 58085.6 1655.34 58513.3
93 2.7e+07 32 1 hip nla 10706.2 3986.93 1672.34 4414.71
94 2.7e+07 76 1 CPU hkn 100970 49264.1 1424.23 49681.2
95 2.7e+07 76 1 cuda hkn 101331 49440.9 1428.33 49860.9
96 2.7e+07 112 1 CPU i20 47640.3 23154.2 807.186 23367.4
123 6.4e+07 4 1 cuda hkn 91951.8 25020.8 28131 30739
124 6.4e+07 4 1 dpcpp i20 134851 56481 13342.4 59973.2
125 6.4e+07 8 1 CPU nla 625633 307556 7040.07 309012
126 6.4e+07 8 1 cuda hkn 70406 23622.5 14823.4 27038.8
127 6.4e+07 8 1 dpcpp i20 73861.3 31223.5 6830.8 33092.1
128 6.4e+07 8 1 hip nla 27986 8741.07 7030.45 10195.1
129 6.4e+07 12 1 dpcpp i20 117912 54826.1 4863.12 56229.7
130 6.4e+07 16 1 CPU nla 391502 192403 4168.06 193429
131 6.4e+07 16 1 cuda hkn 66505.3 26301.9 8334.17 28556
132 6.4e+07 16 1 hip nla 23989.6 8642.56 4180.39 9668.03
133 6.4e+07 32 1 CPU nla 367910 180741 3929.81 181751
134 6.4e+07 32 1 hip nla 25670.2 9609.78 3955.51 10619.1
135 6.4e+07 76 1 CPU hkn 334223 164236 3425.47 165198
136 6.4e+07 76 1 cuda hkn 152342 73279.8 3427.56 74252.9
137 6.4e+07 112 1 CPU i20 174280 85507.9 1992.32 86027.3
164 1.25e+08 4 1 dpcpp i20 320618 139002 26555.6 145757
165 1.25e+08 8 1 dpcpp i20 168610 73125.4 13733.3 76706.4
166 1.25e+08 12 1 dpcpp i20 227618 105921 9502.99 108536
167 1.25e+08 112 1 CPU i20 444850 219281 3893.83 220290

Preconditioned

nCells nProcs nNodes executor preconditioner Host TimeStep SolveP MomentumPredictor PISOStep
0 1000000 4 1 cuda BJ hkn 1296.43 417.936 290.075 486.896
1 1000000 4 1 cuda GISAI hkn 3403.09 1471.27 291.157 1539.88
2 1000000 4 1 cuda Multigrid hkn 3385.17 1461.7 291.554 1530.65
3 1000000 8 1 CPU DIC nla 1008.95 424.698 105.907 447.185
5 1000000 8 1 CPU GaussSeidel nla 674.104 257.792 105.246 280.08
6 1000000 8 1 cuda BJ hkn 3580.68 1660.38 159.798 1701.18
7 1000000 8 1 cuda GISAI hkn 3818.82 1781.6 157.644 1821.23
8 1000000 8 1 cuda Multigrid hkn 1728.81 730.236 163.124 773.086
9 1000000 8 1 hip BJ nla 655.408 244.19 110.857 268.058
10 1000000 8 1 hip GISAI nla 553.786 194.169 110.308 217.509
11 1000000 8 1 hip Multigrid nla 641.008 237.591 110.354 261.105
12 1000000 16 1 CPU DIC nla 499.978 209.313 52.2421 221.55
14 1000000 16 1 CPU GaussSeidel nla 278.837 98.9277 51.9362 111.212
15 1000000 16 1 cuda BJ hkn 5835.56 2839 92.8896 2864.98
16 1000000 16 1 cuda GISAI hkn 5372.2 2607.8 92.8549 2633.26
17 1000000 16 1 cuda Multigrid hkn 4848.38 2345.74 92.8751 2371.43
18 1000000 16 1 hip BJ nla 594.946 255.06 53.4875 268.235
19 1000000 16 1 hip GISAI nla 460.262 188.112 53.382 201.022
20 1000000 16 1 hip Multigrid nla 539.806 226.862 53.6956 240.579
21 1000000 32 1 CPU DIC nla 425.433 178.084 42.0129 189.698
23 1000000 32 1 CPU GaussSeidel nla 253.854 92.576 42.0206 103.826
24 1000000 32 1 hip BJ nla 766.785 346.221 45.8328 358.384
25 1000000 32 1 hip GISAI nla 586.781 255.327 46.0275 268.283
26 1000000 32 1 hip Multigrid nla 620.568 273.405 45.0096 285.714
27 1000000 76 1 CPU DIC hkn 179.74 66.7822 20.3974 76.3751
28 1000000 76 1 CPU GaussSeidel hkn 468.481 210.95 20.695 220.624
29 1000000 76 1 cuda BJ hkn 25637.5 12792.9 21.1161 12804.8
30 1000000 76 1 cuda GISAI hkn 18484.3 9216.72 21.741 9227.97
31 1000000 76 1 cuda Multigrid hkn 21804.7 10873.9 21.8042 10887.7
122 8000000 4 1 cuda BJ hkn 9406.28 2533.42 2693.68 3197.52
123 8000000 4 1 cuda GISAI hkn 9538.99 2608.96 2684.26 3270.21
124 8000000 4 1 cuda Multigrid hkn 10283.7 2947.45 2718.72 3622.34
125 8000000 8 1 CPU DIC nla 23318.7 10996.6 871.812 11185.3
127 8000000 8 1 CPU GaussSeidel nla 7402.77 3041.09 870.745 3228.6
128 8000000 8 1 cuda BJ hkn 11396.6 4530.4 1466.77 4884.16
129 8000000 8 1 cuda GISAI hkn 8917.64 3291.43 1469.49 3643.25
130 8000000 8 1 cuda Multigrid hkn 9156.7 3417.77 1459.4 3767.46
131 8000000 8 1 hip BJ nla 3634.98 1152.06 873.999 1342.87
132 8000000 8 1 hip GISAI nla 3087.51 873.813 884.067 1064.49
133 8000000 8 1 hip Multigrid nla 3139.99 903.354 876.908 1093.74
134 8000000 16 1 CPU DIC nla 12807.8 6007.32 487.377 6130.88
136 8000000 16 1 CPU GaussSeidel nla 3176.28 1190.93 487.798 1314.85
137 8000000 16 1 cuda BJ hkn 15660.7 7053.38 939.169 7299.4
138 8000000 16 1 cuda GISAI hkn 10801.6 4620.83 946.123 4867.58
139 8000000 16 1 cuda Multigrid hkn 7559.11 2997.97 948.681 3244.42
140 8000000 16 1 hip BJ nla 3334.28 1250.19 521.193 1377.58
141 8000000 16 1 hip GISAI nla 2764.78 965.338 521.95 1092.42
142 8000000 16 1 hip Multigrid nla 2545.68 858.455 515.303 985.913
143 8000000 32 1 CPU DIC nla 11216.9 5234.35 448.952 5354.49
145 8000000 32 1 CPU GaussSeidel nla 2441.19 847.274 448.202 966.989
146 8000000 32 1 hip BJ nla 4971.05 2101.52 463.131 2223.98
147 8000000 32 1 hip GISAI nla 3970.29 1601.56 463.692 1723.97
148 8000000 32 1 hip Multigrid nla 3030.97 1128.91 468.676 1252.19
149 8000000 76 1 CPU DIC hkn 9324.05 4330.48 374.959 4447.5
150 8000000 76 1 CPU GaussSeidel hkn 6641.91 2988.6 375.085 3105.75
151 8000000 76 1 cuda BJ hkn 59274.3 29300.7 374.371 29420.9
152 8000000 76 1 cuda GISAI hkn 37873.9 18602.9 369.414 18723.2
153 8000000 76 1 cuda Multigrid hkn 37004.2 18166.7 372.532 18286.6
244 27000000 4 1 cuda BJ hkn 39666.9 9697.11 14400.2 12162.9
245 27000000 4 1 cuda GISAI hkn 27233.8 8076.23 4821.4 10507.8
247 27000000 8 1 CPU DIC nla 130337 62944.6 2959.58 63567.9
249 27000000 8 1 CPU GaussSeidel nla 43124.4 19344.9 2956.24 19963.3
250 27000000 8 1 cuda BJ hkn 30807.5 11128.1 5219.71 12490.4
251 27000000 8 1 cuda GISAI hkn 19985.2 7023.97 2464.03 8386.28
253 27000000 8 1 hip BJ nla 11316.9 3430.12 2964.81 4054.61
254 27000000 8 1 hip GISAI nla 9345.62 2452.8 2965.16 3079.62
255 27000000 8 1 hip Multigrid nla 10310.1 2921.65 2971.47 3548.28
256 27000000 16 1 CPU DIC nla 73637.7 35399.9 1763.33 35835.6
258 27000000 16 1 CPU GaussSeidel nla 15330.8 6248.66 1764.24 6682.18
259 27000000 16 1 cuda BJ hkn 34562 14428.5 3379.51 15367.1
260 27000000 16 1 cuda Multigrid hkn 21010.2 7646.15 3374.31 8584.72
261 27000000 16 1 hip BJ nla 9277.07 3215.91 1770.01 3652.35
262 27000000 16 1 hip GISAI nla 7588.15 2372.8 1775.92 2808.99
263 27000000 16 1 hip Multigrid nla 7867.73 2508.95 1772.01 2946.45
264 27000000 32 1 CPU DIC nla 71211.6 34248.3 1652.98 34676.2
266 27000000 32 1 CPU GaussSeidel nla 14024.9 5656.83 1654.17 6083.52
267 27000000 32 1 hip BJ nla 10635.9 3951.43 1673.9 4379.84
268 27000000 32 1 hip GISAI nla 10511.2 3896.88 1671.28 4325.36
269 27000000 32 1 hip Multigrid nla 8755.31 3009.31 1673.39 3438.99
270 27000000 76 1 CPU DIC hkn 66177.2 31870.9 1425.65 32286.6
271 27000000 76 1 CPU GaussSeidel hkn 36485.6 17032.8 1423.25 17444
272 27000000 76 1 cuda BJ hkn 102483 50018.7 1426.12 50439.3
273 27000000 76 1 cuda Multigrid hkn 60078.9 28817.2 1424.02 29238.2
364 64000000 4 1 cuda BJ hkn 88406.8 25248.6 23816.6 30988.6
365 64000000 4 1 cuda GISAI hkn 67428.9 19369.9 13744 25205.9
367 64000000 8 1 CPU DIC nla 423882 206685 7032.03 208143
369 64000000 8 1 CPU GaussSeidel nla 83867.9 36693.6 7008.37 38145
370 64000000 8 1 cuda BJ hkn 73402.7 23847.6 17524.1 27254.3
371 64000000 8 1 cuda GISAI hkn 42976.2 13504.3 7412.19 16886.8
373 64000000 8 1 hip BJ nla 28402.2 8948.95 7025.11 10405.2
374 64000000 8 1 hip GISAI nla 21689.1 5633.34 7013.82 7086.08
375 64000000 8 1 hip Multigrid nla 26675.7 8086.4 7024.19 9541.05
376 64000000 16 1 CPU DIC nla 252932 123115 4174.67 124140
378 64000000 16 1 CPU GaussSeidel nla 44971 19142.2 4169.92 20164.4
379 64000000 16 1 cuda BJ hkn 66808.9 26409.6 8392.78 28678.9
380 64000000 16 1 cuda GISAI hkn 33713.1 11998.6 3993.18 14223.9
382 64000000 16 1 hip BJ nla 24453.8 8871.58 4183.2 9897.82
383 64000000 16 1 hip GISAI nla 17186.1 5256.7 4183.27 6284.24
384 64000000 16 1 hip Multigrid nla 20379.6 6834.63 4183.62 7861.22
385 64000000 32 1 CPU DIC nla 232953 113246 3930.14 114271
387 64000000 32 1 CPU GaussSeidel nla 30972.9 12271.1 3936.08 13280.4
388 64000000 32 1 hip BJ nla 25538.2 9546.63 3948.44 10554.9
389 64000000 32 1 hip GISAI nla 20591.7 7099.57 3957.72 8108.61
390 64000000 32 1 hip Multigrid nla 20958.1 7254.01 3950.89 8263.73
391 64000000 76 1 CPU DIC hkn 200136 97193.4 3421.09 98155.6
392 64000000 76 1 CPU GaussSeidel hkn 173054 83659.7 3407.31 84621.6
393 64000000 76 1 cuda BJ hkn 152537 73388.8 3410.7 74358.8
394 64000000 76 1 cuda Multigrid hkn 107816 51024.7 3415.98 51993.8

Plots

Test Image 6

@greole greole force-pushed the ogl_170_rev1/2024-01-10_09_21 branch 2 times, most recently from bf254cf to 58aad99 Compare January 10, 2024 12:05
@greole greole changed the title Add new logs -> /home/greole/data/code/exasim_project/benchmark_data/… ogl_170_rev1/2024-01-10_09_21 Jan 10, 2024
@greole greole changed the title ogl_170_rev1/2024-01-10_09_21 Ogl_170_rev1/2024-01-10_09_21 Jan 10, 2024
@greole greole changed the title Ogl_170_rev1/2024-01-10_09_21 Ogl 170_rev1 2024-01-10_09_21 Jan 10, 2024
@greole greole changed the title Ogl 170_rev1 2024-01-10_09_21 Ogl 170rev1 2024-01-10 Jan 10, 2024
@greole greole force-pushed the ogl_170_rev1/2024-01-10_09_21 branch 17 times, most recently from b20c6a7 to 20c9d35 Compare January 11, 2024 09:12
@github-actions github-actions bot force-pushed the ogl_170_rev1/2024-01-10_09_21 branch from c0d0c8c to 29a6a50 Compare January 11, 2024 09:28
@greole greole force-pushed the ogl_170_rev1/2024-01-10_09_21 branch from 29a6a50 to 399fae3 Compare January 11, 2024 09:31
@greole greole force-pushed the ogl_170_rev1/2024-01-10_09_21 branch from 4142410 to 62e6d8f Compare February 2, 2024 19:59
@greole greole force-pushed the ogl_170_rev1/2024-01-10_09_21 branch from cf6fece to 04ccee5 Compare February 18, 2024 16:17
@greole
Copy link
Contributor Author

greole commented Feb 18, 2024

Some results

fvOpsP results

Speedup results

Cost per time step

@greole greole force-pushed the ogl_170_rev1/2024-01-10_09_21 branch from 3656e14 to 0de2ab6 Compare February 18, 2024 19:36
@greole greole force-pushed the ogl_170_rev1/2024-01-10_09_21 branch from 78ea30f to 173ea30 Compare February 28, 2024 08:11
@greole greole force-pushed the ogl_170_rev1/2024-01-10_09_21 branch from 0091f49 to 9cf8cfd Compare February 28, 2024 09:33
@greole greole force-pushed the ogl_170_rev1/2024-01-10_09_21 branch from 909d3ae to f2ffbe1 Compare February 28, 2024 11:22
@github-actions github-actions bot force-pushed the ogl_170_rev1/2024-01-10_09_21 branch 2 times, most recently from 387bd49 to 60cb45f Compare February 28, 2024 11:29
@greole greole force-pushed the ogl_170_rev1/2024-01-10_09_21 branch from 60cb45f to a0b75ef Compare February 29, 2024 08:43
@greole greole force-pushed the ogl_170_rev1/2024-01-10_09_21 branch from 3bd5668 to a299fa6 Compare February 29, 2024 09:01
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants