Skip to content

Instantly share code, notes, and snippets.

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 treo/bfb7ff408fc9bdd4191a44e72fc57a0e to your computer and use it in GitHub Desktop.
Save treo/bfb7ff408fc9bdd4191a44e72fc57a0e to your computer and use it in GitHub Desktop.
ND4J Benchmark CSVs
Benchmark Mode Threads Samples Score Score Error (99,9%) Unit
com.example.BlasWrapper.asum sample 1 160928 162,498384 1,053965 ns/op
com.example.BlasWrapper.axpy sample 1 130358 199,636348 1,038751 ns/op
com.example.BlasWrapper.copy sample 1 131610 194,338682 1,378393 ns/op
com.example.BlasWrapper.dot sample 1 128147 197,497062 1,028926 ns/op
com.example.BlasWrapper.gemv sample 1 179699 3965,476647 33,478549 ns/op
com.example.BlasWrapper.ger sample 1 172296 3269,134072 6,846023 ns/op
com.example.BlasWrapper.iamax sample 1 191449 260,563529 0,439925 ns/op
com.example.BlasWrapper.nrm2 sample 1 130469 195,956258 1,022481 ns/op
com.example.BlasWrapper.swap sample 1 122086 207,398809 0,956560 ns/op
com.example.Large_NDArray.add sample 1 40 361,542451 117,254975 ms/op
com.example.Large_NDArray.addi sample 1 139 90,440623 0,470210 ms/op
com.example.Large_NDArray.assign sample 1 132 92,066164 0,620488 ms/op
com.example.Large_NDArray.dup sample 1 145426 1,266161 0,023340 us/op
com.example.Large_NDArray.mul sample 1 39 374,684436 122,545308 ms/op
com.example.Large_NDArray.muli sample 1 132 92,700672 1,497638 ms/op
com.example.Large_NDArray.sub sample 1 40 368,784179 120,844024 ms/op
com.example.Large_NDArray.subi sample 1 132 91,811964 0,689069 ms/op
com.example.Large_NDArray.sumNumber sample 1 419 29,032057 0,201504 ms/op
com.example.Medium_NDArray.add sample 1 110305 109,748275 14,261133 us/op
com.example.Medium_NDArray.addi sample 1 143660 10,429420 0,025919 us/op
com.example.Medium_NDArray.assign sample 1 163268 9,174165 0,021814 us/op
com.example.Medium_NDArray.cumsum sample 1 1045 11553,306947 94,745029 us/op
com.example.Medium_NDArray.cumsumi sample 1 1094 11011,890077 72,739696 us/op
com.example.Medium_NDArray.dup sample 1 144440 1,211340 0,023901 us/op
com.example.Medium_NDArray.mul sample 1 101458 118,182276 15,387407 us/op
com.example.Medium_NDArray.muli sample 1 149200 10,043866 0,007222 us/op
com.example.Medium_NDArray.sub sample 1 73480 169,747071 43,186397 us/op
com.example.Medium_NDArray.subi sample 1 148878 10,089799 0,080796 us/op
com.example.Medium_NDArray.sumNumber sample 1 148987 40,282622 0,517039 us/op
com.example.Nd4j.toFlattened_CC_Small sample 1 9017 1374,907890 300,378115 us/op
com.example.Nd4j.toFlattened_CF_Small sample 1 5876 2060,208185 362,347621 us/op
com.example.Nd4j.toFlattened_FC_Small sample 1 5866 2061,124517 357,809272 us/op
com.example.Nd4j.toFlattened_FF_Small sample 1 8893 1375,187208 284,604296 us/op
com.example.Small_NDArray.add sample 1 175366 1753,479534 86,832835 ns/op
com.example.Small_NDArray.addi sample 1 232342 220,314115 0,725371 ns/op
com.example.Small_NDArray.assign sample 1 231455 223,240444 9,024004 ns/op
com.example.Small_NDArray.cumsum sample 1 132393 45258,993210 139,182230 ns/op
com.example.Small_NDArray.cumsumi sample 1 138453 43279,501578 104,445244 ns/op
com.example.Small_NDArray.dup sample 1 145086 1258,614987 32,231858 ns/op
com.example.Small_NDArray.mul sample 1 112925 1655,368306 25,962230 ns/op
com.example.Small_NDArray.muli sample 1 119408 212,707298 1,476346 ns/op
com.example.Small_NDArray.sub sample 1 105279 1591,240551 25,502304 ns/op
com.example.Small_NDArray.subi sample 1 234276 216,308282 1,126093 ns/op
com.example.Small_NDArray.sumNumber sample 1 122289 3648,885288 7294,179534 ns/op
Benchmark Mode Threads Samples Score Score Error (99,9%) Unit
com.example.BlasWrapper.asum sample 1 217587 232,581395 10,233567 ns/op
com.example.BlasWrapper.axpy sample 1 121184 213,622962 6,356601 ns/op
com.example.BlasWrapper.copy sample 1 152982 321,974030 1,341532 ns/op
com.example.BlasWrapper.dot sample 1 120598 212,302783 1,846405 ns/op
com.example.BlasWrapper.gemv sample 1 162101 8466,519614 82,751298 ns/op
com.example.BlasWrapper.ger sample 1 191795 3971,305404 70,698543 ns/op
com.example.BlasWrapper.iamax sample 1 182725 271,384091 1,211914 ns/op
com.example.BlasWrapper.nrm2 sample 1 184294 273,911207 9,135310 ns/op
com.example.BlasWrapper.swap sample 1 214823 235,736597 1,311325 ns/op
com.example.Large_NDArray.add sample 1 20 783,443558 143,902871 ms/op
com.example.Large_NDArray.addi sample 1 48 258,321067 3,766151 ms/op
com.example.Large_NDArray.assign sample 1 60 237,349547 3,030399 ms/op
com.example.Large_NDArray.dup sample 1 109431 2,700351 0,030738 us/op
com.example.Large_NDArray.mul sample 1 20 759,903027 142,025094 ms/op
com.example.Large_NDArray.muli sample 1 48 255,628629 2,432502 ms/op
com.example.Large_NDArray.sub sample 1 20 756,862157 143,383813 ms/op
com.example.Large_NDArray.subi sample 1 48 256,507904 2,941932 ms/op
com.example.Large_NDArray.sumNumber sample 1 48 316,724565 3,398481 ms/op
com.example.Medium_NDArray.add sample 1 50094 269,383515 105,996272 us/op
com.example.Medium_NDArray.addi sample 1 141282 84,805779 0,234272 us/op
com.example.Medium_NDArray.assign sample 1 151955 78,830547 0,216339 us/op
com.example.Medium_NDArray.cumsum sample 1 589 20534,186812 221,862783 us/op
com.example.Medium_NDArray.cumsumi sample 1 603 20108,846700 82,178000 us/op
com.example.Medium_NDArray.dup sample 1 108299 2,735178 0,032686 us/op
com.example.Medium_NDArray.mul sample 1 47442 266,258179 87,246165 us/op
com.example.Medium_NDArray.muli sample 1 136125 88,021264 0,255594 us/op
com.example.Medium_NDArray.sub sample 1 51125 265,866553 86,110237 us/op
com.example.Medium_NDArray.subi sample 1 137515 87,129712 0,275177 us/op
com.example.Medium_NDArray.sumNumber sample 1 94664 126,634186 2,995408 us/op
com.example.Nd4j.toFlattened_CC_Small sample 1 10884 1315,981501 390,455360 us/op
com.example.Nd4j.toFlattened_CF_Small sample 1 5207 2438,602411 462,315822 us/op
com.example.Nd4j.toFlattened_FC_Small sample 1 4986 2517,593749 507,003775 us/op
com.example.Nd4j.toFlattened_FF_Small sample 1 10884 1324,839609 388,274817 us/op
com.example.Small_NDArray.add sample 1 155319 4136,807062 69,039895 ns/op
com.example.Small_NDArray.addi sample 1 156035 1216,510674 3,236772 ns/op
com.example.Small_NDArray.assign sample 1 172705 1104,819137 23,034357 ns/op
com.example.Small_NDArray.cumsum sample 1 139504 85885,392362 290,711869 ns/op
com.example.Small_NDArray.cumsumi sample 1 145831 82175,367158 223,898511 ns/op
com.example.Small_NDArray.dup sample 1 111993 2745,967087 43,797758 ns/op
com.example.Small_NDArray.mul sample 1 146886 7567,340087 11322,395100 ns/op
com.example.Small_NDArray.muli sample 1 157844 1214,606548 26,194230 ns/op
com.example.Small_NDArray.sub sample 1 155034 4128,706709 28,047979 ns/op
com.example.Small_NDArray.subi sample 1 153995 1231,019631 3,286074 ns/op
com.example.Small_NDArray.sumNumber sample 1 138895 4470,102538 24,315348 ns/op
Benchmark Mode Threads Samples Score Score Error (99,9%) Unit
com.example.BlasWrapper.asum sample 1 232502 218,377588 6,139234 ns/op
com.example.BlasWrapper.axpy sample 1 129550 204,306746 2,024827 ns/op
com.example.BlasWrapper.copy sample 1 159657 313,995941 7,663477 ns/op
com.example.BlasWrapper.dot sample 1 132947 203,799446 21,926324 ns/op
com.example.BlasWrapper.gemv sample 1 215782 27374,621720 992,720884 ns/op
com.example.BlasWrapper.ger sample 1 201915 3743,025332 39,917710 ns/op
com.example.BlasWrapper.iamax sample 1 191734 261,318384 0,544533 ns/op
com.example.BlasWrapper.nrm2 sample 1 195517 258,041071 0,935421 ns/op
com.example.BlasWrapper.swap sample 1 124361 207,782753 1,247185 ns/op
com.example.Large_NDArray.add sample 1 20 772,118938 137,776701 ms/op
com.example.Large_NDArray.addi sample 1 48 260,745899 4,165958 ms/op
com.example.Large_NDArray.assign sample 1 55 249,542023 5,710353 ms/op
com.example.Large_NDArray.dup sample 1 141395 20,983002 0,138128 us/op
com.example.Large_NDArray.mul sample 1 20 788,004864 180,848344 ms/op
com.example.Large_NDArray.muli sample 1 48 264,050005 6,243650 ms/op
com.example.Large_NDArray.sub sample 1 20 768,396493 135,155132 ms/op
com.example.Large_NDArray.subi sample 1 48 262,438912 4,499352 ms/op
com.example.Large_NDArray.sumNumber sample 1 48 321,650688 2,397699 ms/op
com.example.Medium_NDArray.add sample 1 50976 249,141223 67,752592 us/op
com.example.Medium_NDArray.addi sample 1 140675 85,187899 0,272839 us/op
com.example.Medium_NDArray.assign sample 1 138991 86,202088 0,284701 us/op
com.example.Medium_NDArray.cumsum sample 1 591 20390,289976 158,436660 us/op
com.example.Medium_NDArray.cumsumi sample 1 603 19993,153380 129,080091 us/op
com.example.Medium_NDArray.dup sample 1 141653 20,873028 0,124323 us/op
com.example.Medium_NDArray.mul sample 1 49619 259,066779 71,594965 us/op
com.example.Medium_NDArray.muli sample 1 138660 86,421019 0,260526 us/op
com.example.Medium_NDArray.sub sample 1 52570 261,767475 86,143312 us/op
com.example.Medium_NDArray.subi sample 1 136213 87,967708 0,259696 us/op
com.example.Medium_NDArray.sumNumber sample 1 94369 127,032653 3,042772 us/op
com.example.Nd4j.toFlattened_CC_Small sample 1 10421 1273,258560 374,545164 us/op
com.example.Nd4j.toFlattened_CF_Small sample 1 5187 2438,099186 420,889921 us/op
com.example.Nd4j.toFlattened_FC_Small sample 1 4780 2558,832040 465,293268 us/op
com.example.Nd4j.toFlattened_FF_Small sample 1 10284 1288,915392 390,401182 us/op
com.example.Small_NDArray.add sample 1 143271 41594,008111 1067,428967 ns/op
com.example.Small_NDArray.addi sample 1 164243 18331,076174 132,499816 ns/op
com.example.Small_NDArray.assign sample 1 163204 18357,749565 117,510155 ns/op
com.example.Small_NDArray.cumsum sample 1 116975 102475,288839 373,005944 ns/op
com.example.Small_NDArray.cumsumi sample 1 148970 80419,369484 243,827345 ns/op
com.example.Small_NDArray.dup sample 1 144042 20972,930673 1106,358753 ns/op
com.example.Small_NDArray.mul sample 1 147435 40749,048733 1757,558245 ns/op
com.example.Small_NDArray.muli sample 1 165005 18147,888949 119,862087 ns/op
com.example.Small_NDArray.sub sample 1 148891 40368,829923 1700,573407 ns/op
com.example.Small_NDArray.subi sample 1 164169 18263,217148 116,694569 ns/op
com.example.Small_NDArray.sumNumber sample 1 148033 40280,193741 338,428146 ns/op
Benchmark Mode Threads Samples Score Score Error (99,9%) Unit
com.example.BlasWrapper.asum sample 1 98915 121313,437355 2493,210434 ns/op
com.example.BlasWrapper.axpy sample 1 138222 24089,708324 4916,988070 ns/op
com.example.BlasWrapper.copy sample 1 140594 23180,181231 4158,377030 ns/op
com.example.BlasWrapper.dot sample 1 98771 121387,812698 3094,123637 ns/op
com.example.BlasWrapper.gemv sample 1 153465 58763,964265 3504,188107 ns/op
com.example.BlasWrapper.iamax sample 1 113142 105965,393152 2437,743303 ns/op
com.example.BlasWrapper.nrm2 sample 1 88601 135338,979199 2626,750035 ns/op
com.example.BlasWrapper.swap sample 1 140816 23088,253494 4379,591477 ns/op
com.example.Large_NDArray.addi sample 1 36 374,545522 4,073024 ms/op
com.example.Large_NDArray.assign sample 1 36 374,632903 4,010492 ms/op
com.example.Large_NDArray.dup sample 1 100126 119,762816 5,153372 us/op
com.example.Large_NDArray.muli sample 1 36 376,686364 6,737549 ms/op
com.example.Large_NDArray.subi sample 1 36 376,744619 6,823393 ms/op
com.example.Large_NDArray.sumNumber sample 1 84 147,349894 0,095344 ms/op
com.example.Medium_NDArray.add sample 1 17298 692,418701 12,930659 us/op
com.example.Medium_NDArray.addi sample 1 38678 309,605752 4,979669 us/op
com.example.Medium_NDArray.assign sample 1 35077 341,298631 6,551230 us/op
com.example.Medium_NDArray.cumsum sample 1 316 38768,069671 498,763351 us/op
com.example.Medium_NDArray.cumsumi sample 1 299 40952,766930 324,555605 us/op
com.example.Medium_NDArray.dup sample 1 99506 120,518894 5,200327 us/op
com.example.Medium_NDArray.mul sample 1 17294 692,864844 14,470768 us/op
com.example.Medium_NDArray.muli sample 1 33235 359,977639 8,382935 us/op
com.example.Medium_NDArray.sub sample 1 17292 692,714504 11,049934 us/op
com.example.Medium_NDArray.subi sample 1 34075 351,280165 8,844424 us/op
com.example.Medium_NDArray.sumNumber sample 1 21725 551,646711 6,959124 us/op
com.example.Nd4j.toFlattened_CC_Small sample 1 14763 4734,373239 5465,162081 us/op
com.example.Nd4j.toFlattened_CF_Small sample 1 101 160891,924277 107481,252778 us/op
com.example.Nd4j.toFlattened_FC_Small sample 1 95 186249,900463 131018,528380 us/op
com.example.Nd4j.toFlattened_FF_Small sample 1 15036 4656,217896 5365,933481 us/op
com.example.Small_NDArray.add sample 1 47509 252456,467280 5968,486507 ns/op
com.example.Small_NDArray.addi sample 1 98171 122146,160149 2537,581830 ns/op
com.example.Small_NDArray.assign sample 1 97886 122479,045747 2511,945246 ns/op
com.example.Small_NDArray.cumsum sample 1 31616 378852,761134 5889,774218 ns/op
com.example.Small_NDArray.cumsumi sample 1 73622 162780,177664 634,282012 ns/op
com.example.Small_NDArray.dup sample 1 99954 120349,568161 4913,751243 ns/op
com.example.Small_NDArray.mul sample 1 50397 238452,156755 5844,980080 ns/op
com.example.Small_NDArray.muli sample 1 100505 119281,706741 2544,184127 ns/op
com.example.Small_NDArray.sub sample 1 50191 240085,441294 5804,064463 ns/op
com.example.Small_NDArray.subi sample 1 100794 118952,449392 2481,011922 ns/op
com.example.Small_NDArray.sumNumber sample 1 51796 231464,691328 4915,567967 ns/op
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment