Update K40 an K20 ECC on results with v0.9 of code

This commit is contained in:
Tom Deakin 2015-08-01 16:43:14 +01:00
parent ed656f5d8e
commit e9cfdfb1cb
4 changed files with 36 additions and 24 deletions

View File

@ -1,12 +1,14 @@
GPU-STREAM GPU-STREAM
Version: 0.0 Version: 0.9
Implementation: CUDA Implementation: CUDA
Precision: double
Warning: array size must divide 1024 Running kernels 10 times
Resizing array from 50000000 to 49999872 Array size: 400.0 MB (=0.4 GB)
Total size: 1200.0 MB (=1.2 GB)
Using CUDA device Tesla K20c Using CUDA device Tesla K20c
Function MBytes/sec Min (sec) Max Average Function MBytes/sec Min (sec) Max Average
Copy 152227 0.0052553 0.00527829 0.0052649 Copy 152296.519 0.00551 0.00553 0.00552
Mul 152206 0.00525604 0.00527526 0.0052646 Mul 152284.216 0.00551 0.00553 0.00552
Add 150350 0.00798137 0.00800578 0.00799512 Add 150549.336 0.00836 0.00838 0.00837
Triad 150349 0.00798142 0.00801054 0.00799636 Triad 150597.842 0.00836 0.00838 0.00837

View File

@ -1,12 +1,14 @@
GPU-STREAM GPU-STREAM
Version: 0.0 Version: 0.9
Implementation: CUDA Implementation: CUDA
Precision: double
Warning: array size must divide 1024 Running kernels 10 times
Resizing array from 50000000 to 49999872 Array size: 400.0 MB (=0.4 GB)
Total size: 1200.0 MB (=1.2 GB)
Using CUDA device Tesla K40c Using CUDA device Tesla K40c
Function MBytes/sec Min (sec) Max Average Function MBytes/sec Min (sec) Max Average
Copy 194203 0.00411939 0.00412812 0.00412312 Copy 194335.669 0.00432 0.00433 0.00432
Mul 194111 0.00412135 0.00412981 0.00412629 Mul 194171.527 0.00432 0.00433 0.00433
Add 191683 0.00626033 0.00626831 0.0062641 Add 191294.438 0.00658 0.00659 0.00658
Triad 191560 0.00626436 0.00628582 0.00626866 Triad 191240.187 0.00658 0.00659 0.00658

View File

@ -1,10 +1,14 @@
GPU-STREAM GPU-STREAM
Version: 0.0 Version: 0.9
Implementation: OpenCL Implementation: OpenCL
Precision: double
Running kernels 10 times
Array size: 400.0 MB (=0.4 GB)
Total size: 1200.0 MB (=1.2 GB)
Using OpenCL device Tesla K20c Using OpenCL device Tesla K20c
Function MBytes/sec Min (sec) Max Average Function MBytes/sec Min (sec) Max Average
Copy 149700 0.00534404 0.00537675 0.00535703 Copy 150397.867 0.00558 0.00561 0.00559
Mul 149734 0.0053428 0.00537013 0.00535519 Mul 150241.232 0.00558 0.00560 0.00559
Add 151108 0.00794132 0.00797839 0.00795825 Add 151673.787 0.00830 0.00833 0.00831
Triad 151008 0.0079466 0.00797772 0.0079599 Triad 151699.186 0.00829 0.00833 0.00831

View File

@ -1,10 +1,14 @@
GPU-STREAM GPU-STREAM
Version: 0.0 Version: 0.9
Implementation: OpenCL Implementation: OpenCL
Precision: double
Running kernels 10 times
Array size: 400.0 MB (=0.4 GB)
Total size: 1200.0 MB (=1.2 GB)
Using OpenCL device Tesla K40c Using OpenCL device Tesla K40c
Function MBytes/sec Min (sec) Max Average Function MBytes/sec Min (sec) Max Average
Copy 190343 0.00420295 0.00422588 0.00421376 Copy 190354.872 0.00441 0.00443 0.00442
Mul 190266 0.00420465 0.00422835 0.00421347 Mul 190199.107 0.00441 0.00443 0.00442
Add 190144 0.00631101 0.00632369 0.00631618 Add 190946.380 0.00659 0.00660 0.00660
Triad 190105 0.00631229 0.00632309 0.00631735 Triad 190991.101 0.00659 0.00661 0.00660