cory-master/BabelStream
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add v2.0 results

Browse Source
This commit is contained in:
Tom Deakin 2016-08-26 15:54:22 +01:00
parent 2352b6cc42
commit 21f0bcfde4
73 changed files with 1061 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
results/v2.0/acc-cray/K20X.txt Normal file
View File

@ -0,0 +1,13 @@
GPU-STREAM
Version: 2.0
Implementation: OpenACC
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 116702.138 0.00460 0.00462 0.00461
Mul 174325.378 0.00308 0.00310 0.00309
Add 261591.864 0.00308 0.00310 0.00309
Triad 175206.996 0.00460 0.00462 0.00461
Application 1396462 resources: utime ~2s, stime ~1s, Rss ~876780, inblocks ~601, outblocks ~323

12
results/v2.0/acc-cray/K40.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: OpenACC
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 120256.706 0.00446 0.00458 0.00448
Mul 185251.125 0.00290 0.00292 0.00291
Add 277727.285 0.00290 0.00388 0.00292
Triad 181094.123 0.00445 0.00448 0.00446

12
results/v2.0/acc-cray/K80.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: OpenACC
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 113606.114 0.00473 0.00560 0.00476
Mul 175729.637 0.00306 0.00367 0.00309
Add 263518.810 0.00306 0.00369 0.00309
Triad 170709.791 0.00472 0.00573 0.00477

13
results/v2.0/acc-pgi-kernel/broadwell.txt Normal file
View File

@ -0,0 +1,13 @@
GPU-STREAM
Version: 2.0
Implementation: OpenACC
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 37575.728 0.01429 0.01440 0.01434
Mul 37848.869 0.01418 0.01433 0.01425
Add 45589.244 0.01766 0.01781 0.01772
Triad 46657.287 0.01726 0.01736 0.01730
Application 1454136 resources: utime ~287s, stime ~1s, Rss ~789384, inblocks ~374, outblocks ~387

13
results/v2.0/acc-pgi-kernel/haswell.txt Normal file
View File

@ -0,0 +1,13 @@
GPU-STREAM
Version: 2.0
Implementation: OpenACC
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 50148.428 0.01071 0.01077 0.01073
Mul 51051.215 0.01052 0.01059 0.01055
Add 55360.902 0.01455 0.01462 0.01458
Triad 54556.116 0.01476 0.01486 0.01481
Application 1454139 resources: utime ~167s, stime ~1s, Rss ~789120, inblocks ~316, outblocks ~387

12
results/v2.0/acc-pgi-kernel/ivybridge.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: OpenACC
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 82520.672 0.00651 0.01660 0.00835
Mul 80283.532 0.00669 0.01680 0.00862
Add 84622.881 0.00952 0.02310 0.01205
Triad 83561.609 0.00964 0.02359 0.01253

12
results/v2.0/acc-pgi-kernel/sandybridge.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: OpenACC
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 28718.686 0.01869 0.03758 0.01919
Mul 27994.499 0.01918 0.02873 0.01984
Add 28014.895 0.02875 0.03684 0.02941
Triad 28070.552 0.02869 0.04073 0.02952

12
results/v2.0/acc-pgi-loops/980Ti.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: OpenACC
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 258666.277 0.00208 0.00208 0.00208
Mul 258730.975 0.00208 0.00208 0.00208
Add 265497.286 0.00303 0.00304 0.00304
Triad 266836.306 0.00302 0.00303 0.00302

12
results/v2.0/acc-pgi-loops/IvyBridge.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: OpenACC
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 77879.649 0.00689 0.02005 0.00893
Mul 59127.697 0.00908 0.01900 0.01069
Add 63608.223 0.01266 0.02617 0.01511
Triad 64017.868 0.01258 0.02615 0.01513

12
results/v2.0/acc-pgi-loops/S9150.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: OpenACC
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 256996.540 0.00209 0.00217 0.00212
Mul 258109.030 0.00208 0.00215 0.00212
Add 265716.643 0.00303 0.00315 0.00306
Triad 265499.387 0.00303 0.00307 0.00306

12
results/v2.0/acc-pgi-loops/SandyBridge.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: OpenACC
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 28017.895 0.01916 0.01995 0.01942
Mul 22199.944 0.02418 0.02536 0.02469
Add 23458.271 0.03433 0.03588 0.03497
Triad 23644.549 0.03406 0.03532 0.03465

13
results/v2.0/acc-pgi-loops/broadwell.txt Normal file
View File

@ -0,0 +1,13 @@
GPU-STREAM
Version: 2.0
Implementation: OpenACC
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 34338.639 0.01563 0.01574 0.01569
Mul 38249.125 0.01404 0.01413 0.01408
Add 40155.137 0.02005 0.02015 0.02009
Triad 40995.666 0.01964 0.01971 0.01967
Application 1396691 resources: utime ~357s, stime ~1s, Rss ~789348, inblocks ~365, outblocks ~350

13
results/v2.0/acc-pgi-loops/haswell.txt Normal file
View File

@ -0,0 +1,13 @@
GPU-STREAM
Version: 2.0
Implementation: OpenACC
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 46401.864 0.01157 0.01163 0.01160
Mul 40767.924 0.01317 0.01326 0.01321
Add 42062.027 0.01915 0.01925 0.01920
Triad 42666.377 0.01887 0.01900 0.01893
Application 1396695 resources: utime ~240s, stime ~1s, Rss ~789084, inblocks ~307, outblocks ~350

14
results/v2.0/cuda-cray/K40.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: CUDA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device Tesla K40m
Driver: 7050
Function MBytes/sec Min (sec) Max Average
Copy 194135.310 0.00277 0.00278 0.00277
Mul 194049.073 0.00277 0.00280 0.00278
Add 190956.372 0.00422 0.00423 0.00422
Triad 190822.844 0.00422 0.00423 0.00422

14
results/v2.0/cuda-cray/K80.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: CUDA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device Tesla K80
Driver: 7050
Function MBytes/sec Min (sec) Max Average
Copy 179851.070 0.00299 0.00357 0.00303
Mul 179720.423 0.00299 0.00358 0.00303
Add 176265.359 0.00457 0.00539 0.00461
Triad 176116.986 0.00457 0.00540 0.00461

14
results/v2.0/cuda-gnu/980ti.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: CUDA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device GeForce GTX 980 Ti
Driver: 7050
Function MBytes/sec Min (sec) Max Average
Copy 263497.383 0.00204 0.00204 0.00204
Mul 263283.395 0.00204 0.00204 0.00204
Add 269113.982 0.00299 0.00300 0.00300
Triad 269153.828 0.00299 0.00300 0.00300

15
results/v2.0/cuda-gnu/K20X.txt Normal file
View File

@ -0,0 +1,15 @@
GPU-STREAM
Version: 2.0
Implementation: CUDA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device Tesla K20X
Driver: 7050
Function MBytes/sec Min (sec) Max Average
Copy 181833.763 0.00295 0.00298 0.00297
Mul 181354.354 0.00296 0.00305 0.00297
Add 179955.484 0.00448 0.00449 0.00448
Triad 179798.066 0.00448 0.00450 0.00449
Application 1396457 resources: utime ~3s, stime ~1s, Rss ~871996, inblocks ~690, outblocks ~1373

14
results/v2.0/cuda-x86/Haswell.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: CUDA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device DEVICE EMULATION MODE
Driver: PGI
Function MBytes/sec Min (sec) Max Average
Copy 38778.163 0.01384 0.01391 0.01388
Mul 38124.361 0.01408 0.01412 0.01410
Add 41817.646 0.01926 0.01934 0.01930
Triad 42446.352 0.01897 0.01906 0.01901

14
results/v2.0/cuda-x86/IvyBridge.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: CUDA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device DEVICE EMULATION MODE
Driver: PGI
Function MBytes/sec Min (sec) Max Average
Copy 57308.251 0.00937 0.02134 0.01109
Mul 55999.151 0.00959 0.02233 0.01134
Add 63534.754 0.01268 0.02962 0.01492
Triad 64546.130 0.01248 0.02873 0.01492

14
results/v2.0/cuda-x86/SandyBridge Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: CUDA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device DEVICE EMULATION MODE
Driver: PGI
Function MBytes/sec Min (sec) Max Average
Copy 21626.429 0.02482 0.03784 0.02526
Mul 21321.415 0.02518 0.02603 0.02551
Add 23394.375 0.03442 0.03588 0.03506
Triad 23527.878 0.03423 0.03550 0.03486

12
results/v2.0/gcc-power8/raja.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: RAJA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 227063.854 0.00236 0.01007 0.00449
Mul 218404.455 0.00246 0.01080 0.00449
Add 257927.160 0.00312 0.01815 0.00647
Triad 253019.962 0.00318 0.01535 0.00635

13
results/v2.0/knl/kokkos-128.txt Normal file
View File

@ -0,0 +1,13 @@
OMP_NUM_THREADS=128 numactl -m 1 ./gpu-stream-kokkos
GPU-STREAM
Version: 2.0
Implementation: KOKKOS
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 284255.707 0.00189 0.00209 0.00199
Mul 259925.621 0.00207 0.00483 0.00426
Add 301882.418 0.00267 0.00295 0.00279
Triad 293037.412 0.00275 0.00314 0.00293

33
results/v2.0/knl/mccalpin.txt Normal file
View File

@ -0,0 +1,33 @@
-------------------------------------------------------------
STREAM version $Revision: 5.10 $
-------------------------------------------------------------
This system uses 8 bytes per array element.
-------------------------------------------------------------
Array size = 33554432 (elements), Offset = 0 (elements)
Memory per array = 256.0 MiB (= 0.2 GiB).
Total memory required = 768.0 MiB (= 0.8 GiB).
Each kernel will be executed 100 times.
The *best* time for each kernel (excluding the first iteration)
will be used to compute the reported bandwidth.
-------------------------------------------------------------
Number of Threads requested = 64
Number of Threads counted = 64
-------------------------------------------------------------
Your clock granularity/precision appears to be 1 microseconds.
Each test below will take on the order of 1507 microseconds.
(= 1507 clock ticks)
Increase the size of the arrays if this shows that
you are not getting at least 20 clock ticks per test.
-------------------------------------------------------------
WARNING -- The above is only a rough guideline.
For best results, please be sure you know the
precision of your system timer.
-------------------------------------------------------------
Function Best Rate MB/s Avg time Min time Max time
Copy: 387306.5 0.001418 0.001386 0.001456
Scale: 414238.4 0.001321 0.001296 0.001368
Add: 444668.2 0.001849 0.001811 0.001875
Triad: 447436.7 0.001855 0.001800 0.001949
-------------------------------------------------------------
Solution Validates: avg error less than 1.000000e-13 on all three arrays
-------------------------------------------------------------

12
results/v2.0/knl/omp3.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: Reference OpenMP
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 355432.548 0.00151 0.00159 0.00154
Mul 303409.465 0.00177 0.00420 0.00214
Add 317176.372 0.00254 0.00266 0.00259
Triad 296841.725 0.00271 0.00307 0.00288

12
results/v2.0/knl/openacc-pgi.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: OpenACC
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 323365.370 0.00166 0.00176 0.00172
Mul 162148.544 0.00331 0.00590 0.00432
Add 425349.852 0.00189 0.00203 0.00193
Triad 239556.020 0.00336 0.00664 0.00549

12
results/v2.0/knl/raja.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: RAJA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 358479.622 0.00150 0.00166 0.00157
Mul 301367.207 0.00178 0.00369 0.00201
Add 317005.071 0.00254 0.00271 0.00261
Triad 298105.168 0.00270 0.00303 0.00286

12
results/v2.0/kokkos/980ti.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: KOKKOS
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 263006.605 0.00204 0.00205 0.00204
Mul 262996.298 0.00204 0.00205 0.00204
Add 268536.157 0.00300 0.00301 0.00300
Triad 268594.912 0.00300 0.00301 0.00300

12
results/v2.0/kokkos/K20X.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: KOKKOS
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 182239.282 0.00295 0.00299 0.00296
Mul 182179.668 0.00295 0.00298 0.00296
Add 182333.793 0.00442 0.00451 0.00443
Triad 182213.531 0.00442 0.00444 0.00443

12
results/v2.0/kokkos/K40.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: KOKKOS
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 192464.515 0.00279 0.00283 0.00281
Mul 192505.853 0.00279 0.00283 0.00281
Add 193303.390 0.00417 0.00419 0.00417
Triad 193249.349 0.00417 0.00419 0.00417

12
results/v2.0/kokkos/K80.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: KOKKOS
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 181880.210 0.00295 0.00352 0.00301
Mul 181906.524 0.00295 0.00353 0.00301
Add 179304.662 0.00449 0.00532 0.00456
Triad 179172.535 0.00449 0.00531 0.00456

12
results/v2.0/kokkos/broadwell.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: KOKKOS
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 86500.072 0.00621 0.00627 0.00622
Mul 84168.782 0.00638 0.00643 0.00640
Add 94162.571 0.00855 0.00867 0.00858
Triad 96282.261 0.00836 0.00843 0.00838

12
results/v2.0/kokkos/haswell.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: KOKKOS
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 79740.484 0.00673 0.00688 0.00678
Mul 77939.661 0.00689 0.00701 0.00694
Add 87154.341 0.00924 0.00954 0.00935
Triad 88503.861 0.00910 0.00945 0.00919

12
results/v2.0/kokkos/ivybridge.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: KOKKOS
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 56720.246 0.00947 0.01173 0.00999
Mul 55256.554 0.00972 0.03029 0.01052
Add 62562.358 0.01287 0.03304 0.01384
Triad 62965.518 0.01279 0.02534 0.01364

12
results/v2.0/kokkos/sandybridge.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: KOKKOS
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 51337.144 0.01046 0.01126 0.01063
Mul 51613.662 0.01040 0.01464 0.01064
Add 54473.120 0.01478 0.01544 0.01506
Triad 54461.048 0.01479 0.01568 0.01523

14
results/v2.0/ocl-gnu/980ti.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: OpenCL
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device GeForce GTX 980 Ti
Driver: 352.21
Function MBytes/sec Min (sec) Max Average
Copy 262864.310 0.00204 0.00206 0.00205
Mul 262886.578 0.00204 0.00205 0.00205
Add 268781.736 0.00300 0.00300 0.00300
Triad 268727.204 0.00300 0.00301 0.00300

14
results/v2.0/ocl-gnu/IvyBridge.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: OpenCL
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz
Driver: 1.2.0.92
Function MBytes/sec Min (sec) Max Average
Copy 47971.490 0.01119 0.02317 0.01306
Mul 46385.194 0.01157 0.02247 0.01341
Add 53319.761 0.01510 0.02831 0.01769
Triad 53374.243 0.01509 0.02794 0.01707

15
results/v2.0/ocl-gnu/K20X.txt Normal file
View File

@ -0,0 +1,15 @@
GPU-STREAM
Version: 2.0
Implementation: OpenCL
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device Tesla K20X
Driver: 352.68
Function MBytes/sec Min (sec) Max Average
Copy 182194.259 0.00295 0.00298 0.00296
Mul 182081.736 0.00295 0.00296 0.00296
Add 182723.055 0.00441 0.00443 0.00442
Triad 182719.573 0.00441 0.00443 0.00442
Application 1396458 resources: utime ~3s, stime ~1s, Rss ~1670780, inblocks ~2549, outblocks ~464

14
results/v2.0/ocl-gnu/K40.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: OpenCL
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device Tesla K40m
Driver: 352.79
Function MBytes/sec Min (sec) Max Average
Copy 190876.305 0.00281 0.00285 0.00283
Mul 190558.963 0.00282 0.00284 0.00283
Add 191437.004 0.00421 0.00422 0.00421
Triad 191420.077 0.00421 0.00423 0.00421

14
results/v2.0/ocl-gnu/K80.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: OpenCL
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device Tesla K80
Driver: 352.79
Function MBytes/sec Min (sec) Max Average
Copy 181392.835 0.00296 0.00357 0.00303
Mul 181350.127 0.00296 0.00356 0.00303
Add 181786.662 0.00443 0.00532 0.00453
Triad 181670.318 0.00443 0.00533 0.00454

14
results/v2.0/ocl-gnu/S9150.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: OpenCL
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device Hawaii
Driver: 1912.5 (VM)
Function MBytes/sec Min (sec) Max Average
Copy 259805.874 0.00207 0.00214 0.00211
Mul 260956.698 0.00206 0.00214 0.00211
Add 268425.077 0.00300 0.00306 0.00304
Triad 267506.939 0.00301 0.00307 0.00305

14
results/v2.0/ocl-gnu/SandyBridge.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: OpenCL
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz
Driver: 1.2.0.8
Function MBytes/sec Min (sec) Max Average
Copy 27240.466 0.01971 0.02092 0.02015
Mul 26733.421 0.02008 0.02266 0.02043
Add 29405.722 0.02739 0.03030 0.02808
Triad 29734.601 0.02708 0.02990 0.02783

15
results/v2.0/ocl-gnu/broadwell.txt Normal file
View File

@ -0,0 +1,15 @@
GPU-STREAM
Version: 2.0
Implementation: OpenCL
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz
Driver: 1.2.0.57
Function MBytes/sec Min (sec) Max Average
Copy 43567.760 0.01232 0.01257 0.01247
Mul 42995.296 0.01249 0.01257 0.01253
Add 48537.031 0.01659 0.01672 0.01664
Triad 49238.925 0.01636 0.01652 0.01645
Application 1407386 resources: utime ~130s, stime ~1s, Rss ~1647432, inblocks ~817, outblocks ~464

14
results/v2.0/ocl-gnu/fury.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: OpenCL
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device Fiji
Driver: 1912.5 (VM)
Function MBytes/sec Min (sec) Max Average
Copy 429587.115 0.00125 0.00141 0.00128
Mul 429295.476 0.00125 0.00135 0.00128
Add 442443.451 0.00182 0.00192 0.00186
Triad 442069.177 0.00182 0.00194 0.00186

15
results/v2.0/ocl-gnu/haswell.txt Normal file
View File

@ -0,0 +1,15 @@
GPU-STREAM
Version: 2.0
Implementation: OpenCL
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device Intel(R) Xeon(R) CPU E5-2698 v3 @ 2.30GHz
Driver: 1.2.0.57
Function MBytes/sec Min (sec) Max Average
Copy 39316.996 0.01365 0.01413 0.01388
Mul 39127.564 0.01372 0.01418 0.01392
Add 43768.720 0.01840 0.01955 0.01871
Triad 44121.647 0.01825 0.01892 0.01847
Application 1407392 resources: utime ~106s, stime ~1s, Rss ~1642860, inblocks ~459, outblocks ~464

14
results/v2.0/ocl-gnu/knl.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: OpenCL
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using OpenCL device Intel(R) Xeon Phi(TM) CPU 7210 @ 1.30GHz
Driver: 1.2.0.10002
Function MBytes/sec Min (sec) Max Average
Copy 218490.851 0.00246 0.00431 0.00257
Mul 216827.480 0.00248 0.00276 0.00258
Add 233472.011 0.00345 0.00365 0.00354
Triad 236852.515 0.00340 0.00365 0.00351

13
results/v2.0/omp3-cray/broadwell.txt Normal file
View File

@ -0,0 +1,13 @@
GPU-STREAM
Version: 2.0
Implementation: Reference OpenMP
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 88171.606 0.00609 0.00614 0.00611
Mul 85751.031 0.00626 0.00632 0.00628
Add 96048.112 0.00838 0.00845 0.00842
Triad 98169.628 0.00820 0.00827 0.00824
Application 1396470 resources: utime ~133s, stime ~3s, Rss ~1576044, inblocks ~6345, outblocks ~16023

13
results/v2.0/omp3-cray/haswell.txt Normal file
View File

@ -0,0 +1,13 @@
GPU-STREAM
Version: 2.0
Implementation: Reference OpenMP
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 81012.503 0.00663 0.00668 0.00665
Mul 79344.699 0.00677 0.00683 0.00678
Add 89615.218 0.00899 0.00907 0.00901
Triad 90999.378 0.00885 0.00893 0.00887
Application 1396725 resources: utime ~104s, stime ~2s, Rss ~1578772, inblocks ~544, outblocks ~213

12
results/v2.0/omp3-intel/IvyBridge.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: Reference OpenMP
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 56676.385 0.00947 0.01567 0.01040
Mul 55505.151 0.00967 0.01513 0.01075
Add 61874.931 0.01302 0.01930 0.01435
Triad 62073.488 0.01297 0.01899 0.01420

12
results/v2.0/omp3-intel/SandyBridge.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: Reference OpenMP
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 54991.394 0.00976 0.01134 0.01002
Mul 48003.655 0.01118 0.01986 0.01146
Add 52111.180 0.01545 0.01621 0.01575
Triad 52985.444 0.01520 0.01608 0.01557

12
results/v2.0/omp40-cray/K40.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: OpenMP 4.0
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 183090.545 0.00293 0.00295 0.00294
Mul 182133.500 0.00295 0.00298 0.00295
Add 180897.478 0.00445 0.00447 0.00446
Triad 180637.056 0.00446 0.00447 0.00446

12
results/v2.0/omp40-cray/K80.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: OpenMP 4.0
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 169214.022 0.00317 0.00437 0.00323
Mul 168803.444 0.00318 0.00435 0.00323
Add 167171.006 0.00482 0.00571 0.00486
Triad 166943.598 0.00482 0.00710 0.00489

12
results/v2.0/omp45-clang/980ti.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: OpenMP 4.5
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 232637.036 0.00231 0.00715 0.00238
Mul 227777.705 0.00236 0.00259 0.00247
Add 246187.342 0.00327 0.00343 0.00332
Triad 239670.377 0.00336 0.00362 0.00347

13
results/v2.0/omp45-cray/K20X.txt Normal file
View File

@ -0,0 +1,13 @@
GPU-STREAM
Version: 2.0
Implementation: OpenMP 4.5
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 174380.925 0.00308 0.00310 0.00309
Mul 174416.162 0.00308 0.00310 0.00309
Add 175158.103 0.00460 0.00462 0.00461
Triad 175104.249 0.00460 0.00462 0.00461
Application 1396463 resources: utime ~2s, stime ~1s, Rss ~876708, inblocks ~600, outblocks ~327

34
results/v2.0/original-cray/broadwell.txt Normal file
View File

@ -0,0 +1,34 @@
-------------------------------------------------------------
STREAM version $Revision: 5.10 $
-------------------------------------------------------------
This system uses 8 bytes per array element.
-------------------------------------------------------------
Array size = 33554432 (elements), Offset = 0 (elements)
Memory per array = 256.0 MiB (= 0.2 GiB).
Total memory required = 768.0 MiB (= 0.8 GiB).
Each kernel will be executed 100 times.
The *best* time for each kernel (excluding the first iteration)
will be used to compute the reported bandwidth.
-------------------------------------------------------------
Number of Threads requested = 44
Number of Threads counted = 44
-------------------------------------------------------------
Your clock granularity/precision appears to be 1 microseconds.
Each test below will take on the order of 4125 microseconds.
(= 4125 clock ticks)
Increase the size of the arrays if this shows that
you are not getting at least 20 clock ticks per test.
-------------------------------------------------------------
WARNING -- The above is only a rough guideline.
For best results, please be sure you know the
precision of your system timer.
-------------------------------------------------------------
Function Best Rate MB/s Avg time Min time Max time
Copy: 92980.4 0.005803 0.005774 0.005843
Scale: 97951.2 0.005527 0.005481 0.005586
Add: 123058.1 0.006592 0.006544 0.006655
Triad: 124799.5 0.006492 0.006453 0.006544
-------------------------------------------------------------
Solution Validates: avg error less than 1.000000e-13 on all three arrays
-------------------------------------------------------------
Application 1396471 resources: utime ~111s, stime ~3s, Rss ~788804, inblocks ~5778, outblocks ~14259

34
results/v2.0/original-cray/haswell.txt Normal file
View File

@ -0,0 +1,34 @@
-------------------------------------------------------------
STREAM version $Revision: 5.10 $
-------------------------------------------------------------
This system uses 8 bytes per array element.
-------------------------------------------------------------
Array size = 33554432 (elements), Offset = 0 (elements)
Memory per array = 256.0 MiB (= 0.2 GiB).
Total memory required = 768.0 MiB (= 0.8 GiB).
Each kernel will be executed 100 times.
The *best* time for each kernel (excluding the first iteration)
will be used to compute the reported bandwidth.
-------------------------------------------------------------
Number of Threads requested = 32
Number of Threads counted = 32
-------------------------------------------------------------
Your clock granularity/precision appears to be 1 microseconds.
Each test below will take on the order of 4685 microseconds.
(= 4685 clock ticks)
Increase the size of the arrays if this shows that
you are not getting at least 20 clock ticks per test.
-------------------------------------------------------------
WARNING -- The above is only a rough guideline.
For best results, please be sure you know the
precision of your system timer.
-------------------------------------------------------------
Function Best Rate MB/s Avg time Min time Max time
Copy: 84400.3 0.006386 0.006361 0.006415
Scale: 99272.6 0.005457 0.005408 0.005546
Add: 118080.7 0.006854 0.006820 0.006892
Triad: 116271.9 0.006969 0.006926 0.007042
-------------------------------------------------------------
Solution Validates: avg error less than 1.000000e-13 on all three arrays
-------------------------------------------------------------
Application 1396734 resources: utime ~84s, stime ~2s, Rss ~791400, inblocks ~490, outblocks ~54

33
results/v2.0/original-icc/ivybridge.txt Normal file
View File

@ -0,0 +1,33 @@
-------------------------------------------------------------
STREAM version $Revision: 5.10 $
-------------------------------------------------------------
This system uses 8 bytes per array element.
-------------------------------------------------------------
Array size = 33554432 (elements), Offset = 0 (elements)
Memory per array = 256.0 MiB (= 0.2 GiB).
Total memory required = 768.0 MiB (= 0.8 GiB).
Each kernel will be executed 100 times.
The *best* time for each kernel (excluding the first iteration)
will be used to compute the reported bandwidth.
-------------------------------------------------------------
Number of Threads requested = 24
Number of Threads counted = 24
-------------------------------------------------------------
Your clock granularity/precision appears to be 2 microseconds.
Each test below will take on the order of 7369 microseconds.
(= 3684 clock ticks)
Increase the size of the arrays if this shows that
you are not getting at least 20 clock ticks per test.
-------------------------------------------------------------
WARNING -- The above is only a rough guideline.
For best results, please be sure you know the
precision of your system timer.
-------------------------------------------------------------
Function Best Rate MB/s Avg time Min time Max time
Copy: 57739.9 0.009647 0.009298 0.010455
Scale: 74390.5 0.008626 0.007217 0.010923
Add: 83859.7 0.010991 0.009603 0.013830
Triad: 82738.1 0.011216 0.009733 0.015579
-------------------------------------------------------------
Solution Validates: avg error less than 1.000000e-13 on all three arrays
-------------------------------------------------------------

33
results/v2.0/original-icc/sandybridge.txt Normal file
View File

@ -0,0 +1,33 @@
-------------------------------------------------------------
STREAM version $Revision: 5.10 $
-------------------------------------------------------------
This system uses 8 bytes per array element.
-------------------------------------------------------------
Array size = 33554432 (elements), Offset = 0 (elements)
Memory per array = 256.0 MiB (= 0.2 GiB).
Total memory required = 768.0 MiB (= 0.8 GiB).
Each kernel will be executed 100 times.
The *best* time for each kernel (excluding the first iteration)
will be used to compute the reported bandwidth.
-------------------------------------------------------------
Number of Threads requested = 16
Number of Threads counted = 16
-------------------------------------------------------------
Your clock granularity/precision appears to be 1 microseconds.
Each test below will take on the order of 6841 microseconds.
(= 6841 clock ticks)
Increase the size of the arrays if this shows that
you are not getting at least 20 clock ticks per test.
-------------------------------------------------------------
WARNING -- The above is only a rough guideline.
For best results, please be sure you know the
precision of your system timer.
-------------------------------------------------------------
Function Best Rate MB/s Avg time Min time Max time
Copy: 57290.4 0.009506 0.009371 0.009757
Scale: 66890.4 0.008211 0.008026 0.008573
Add: 65196.5 0.012791 0.012352 0.023453
Triad: 64351.8 0.012753 0.012514 0.013659
-------------------------------------------------------------
Solution Validates: avg error less than 1.000000e-13 on all three arrays
-------------------------------------------------------------

12
results/v2.0/raja/980ti.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: RAJA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 262849.638 0.00204 0.00205 0.00204
Mul 262842.175 0.00204 0.00205 0.00205
Add 268802.910 0.00300 0.00300 0.00300
Triad 268830.368 0.00300 0.00300 0.00300

12
results/v2.0/raja/K20X.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: RAJA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 182375.663 0.00294 0.00296 0.00295
Mul 182515.909 0.00294 0.00296 0.00295
Add 181695.649 0.00443 0.00445 0.00444
Triad 181436.686 0.00444 0.00445 0.00445

12
results/v2.0/raja/K40.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: RAJA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 191811.130 0.00280 0.00282 0.00281
Mul 191720.029 0.00280 0.00282 0.00281
Add 192768.490 0.00418 0.00419 0.00418
Triad 192718.253 0.00418 0.00419 0.00418

12
results/v2.0/raja/K80.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: RAJA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 182711.634 0.00294 0.00354 0.00302
Mul 182628.907 0.00294 0.00354 0.00302
Add 178679.382 0.00451 0.00533 0.00462
Triad 178467.177 0.00451 0.00534 0.00462

12
results/v2.0/raja/broadwell.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: RAJA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 88094.593 0.00609 0.00619 0.00615
Mul 85183.815 0.00630 0.00643 0.00633
Add 95834.438 0.00840 0.00850 0.00846
Triad 97943.551 0.00822 0.00836 0.00827

12
results/v2.0/raja/haswell.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: RAJA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 80888.406 0.00664 0.00671 0.00667
Mul 79024.295 0.00679 0.00686 0.00682
Add 89360.767 0.00901 0.00910 0.00905
Triad 90744.543 0.00887 0.00893 0.00890

12
results/v2.0/raja/ivybridge.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: RAJA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 55500.859 0.00967 0.01992 0.01139
Mul 55232.718 0.00972 0.01978 0.01151
Add 62568.702 0.01287 0.02616 0.01523
Triad 64105.913 0.01256 0.02570 0.01497

12
results/v2.0/raja/sandybridge.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: RAJA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 51576.355 0.01041 0.01123 0.01064
Mul 50943.953 0.01054 0.01359 0.01074
Add 53535.927 0.01504 0.01598 0.01535
Triad 53928.576 0.01493 0.01579 0.01534

14
results/v2.0/sycl/IvyBridge.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: SYCL
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using SYCL device Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz
Driver: 1.2.0.92
Function MBytes/sec Min (sec) Max Average
Copy 39756.124 0.01350 0.02534 0.01567
Mul 38899.994 0.01380 0.02237 0.01554
Add 46878.810 0.01718 0.02802 0.01919
Triad 51324.819 0.01569 0.02555 0.01748

14
results/v2.0/sycl/S9150.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: SYCL
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using SYCL device Hawaii
Driver: 1912.5 (VM)
Function MBytes/sec Min (sec) Max Average
Copy 222466.991 0.00241 0.00251 0.00245
Mul 224827.470 0.00239 0.00247 0.00244
Add 271092.068 0.00297 0.00305 0.00301
Triad 269725.824 0.00299 0.00304 0.00302

14
results/v2.0/sycl/fury.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: SYCL
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using SYCL device Fiji
Driver: 1912.5 (VM)
Function MBytes/sec Min (sec) Max Average
Copy 419830.223 0.00128 0.00141 0.00130
Mul 419811.510 0.00128 0.00136 0.00130
Add 432957.387 0.00186 0.00193 0.00188
Triad 430761.906 0.00187 0.00194 0.00189

14
results/v2.0/sycl/knl.txt Normal file
View File

@ -0,0 +1,14 @@
GPU-STREAM
Version: 2.0
Implementation: SYCL
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Using SYCL device Intel(R) Xeon Phi(TM) CPU 7210 @ 1.30GHz
Driver: 1.2.0.10002
Function MBytes/sec Min (sec) Max Average
Copy 188284.193 0.00285 0.00447 0.00298
Mul 185567.824 0.00289 0.00762 0.00307
Add 207104.230 0.00389 0.00614 0.00404
Triad 207078.189 0.00389 0.01483 0.00415

12
results/v2.0/xl-power8/kokkos.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: KOKKOS
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 260801.154 0.00206 0.00955 0.00234
Mul 231963.990 0.00231 0.01149 0.00264
Add 292167.544 0.00276 0.01150 0.00309
Triad 298266.810 0.00270 0.01533 0.00316

33
results/v2.0/xl-power8/mccalpin.txt Normal file
View File

@ -0,0 +1,33 @@
-------------------------------------------------------------
STREAM version $Revision: 5.10 $
-------------------------------------------------------------
This system uses 8 bytes per array element.
-------------------------------------------------------------
Array size = 33554432 (elements), Offset = 0 (elements)
Memory per array = 256.0 MiB (= 0.2 GiB).
Total memory required = 768.0 MiB (= 0.8 GiB).
Each kernel will be executed 100 times.
The *best* time for each kernel (excluding the first iteration)
will be used to compute the reported bandwidth.
-------------------------------------------------------------
Number of Threads requested = 20
Number of Threads counted = 20
-------------------------------------------------------------
Your clock granularity/precision appears to be 1 microseconds.
Each test below will take on the order of 2250 microseconds.
(= 2250 clock ticks)
Increase the size of the arrays if this shows that
you are not getting at least 20 clock ticks per test.
-------------------------------------------------------------
WARNING -- The above is only a rough guideline.
For best results, please be sure you know the
precision of your system timer.
-------------------------------------------------------------
Function Best Rate MB/s Avg time Min time Max time
Copy: 261745.9 0.002101 0.002051 0.002517
Scale: 253352.8 0.002188 0.002119 0.003140
Add: 239468.3 0.003499 0.003363 0.004400
Triad: 245151.7 0.003468 0.003285 0.004771
-------------------------------------------------------------
Solution Validates: avg error less than 1.000000e-13 on all three arrays
-------------------------------------------------------------

12
results/v2.0/xl-power8/omp3.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: Reference OpenMP
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 223266.147 0.00240 0.00271 0.00248
Mul 196756.047 0.00273 0.00372 0.00297
Add 210090.244 0.00383 0.00441 0.00396
Triad 212958.097 0.00378 0.00500 0.00409

12
results/v2.0/xl-power8/raja.txt Normal file
View File

@ -0,0 +1,12 @@
GPU-STREAM
Version: 2.0
Implementation: RAJA
Running kernels 100 times
Precision: double
Array size: 268.4 MB (=0.3 GB)
Total size: 805.3 MB (=0.8 GB)
Function MBytes/sec Min (sec) Max Average
Copy 257796.702 0.00208 0.01017 0.00423
Mul 218399.746 0.00246 0.01163 0.00440
Add 269553.023 0.00299 0.01575 0.00620
Triad 279022.596 0.00289 0.01569 0.00614