From e9cfdfb1cb7720e6f5063b151344daf874285014 Mon Sep 17 00:00:00 2001 From: Tom Deakin Date: Sat, 1 Aug 2015 16:43:14 +0100 Subject: [PATCH] Update K40 an K20 ECC on results with v0.9 of code --- results/cuda/nvidia-k20c-ecc-on.txt | 16 +++++++++------- results/cuda/nvidia-k40c-ecc-on.txt | 16 +++++++++------- results/opencl/nvidia-k20c-ecc-on.txt | 14 +++++++++----- results/opencl/nvidia-k40c-ecc-on.txt | 14 +++++++++----- 4 files changed, 36 insertions(+), 24 deletions(-) diff --git a/results/cuda/nvidia-k20c-ecc-on.txt b/results/cuda/nvidia-k20c-ecc-on.txt index 818101f..ce6adaa 100644 --- a/results/cuda/nvidia-k20c-ecc-on.txt +++ b/results/cuda/nvidia-k20c-ecc-on.txt @@ -1,12 +1,14 @@ GPU-STREAM -Version: 0.0 +Version: 0.9 Implementation: CUDA +Precision: double -Warning: array size must divide 1024 -Resizing array from 50000000 to 49999872 +Running kernels 10 times +Array size: 400.0 MB (=0.4 GB) +Total size: 1200.0 MB (=1.2 GB) Using CUDA device Tesla K20c Function MBytes/sec Min (sec) Max Average -Copy 152227 0.0052553 0.00527829 0.0052649 -Mul 152206 0.00525604 0.00527526 0.0052646 -Add 150350 0.00798137 0.00800578 0.00799512 -Triad 150349 0.00798142 0.00801054 0.00799636 +Copy 152296.519 0.00551 0.00553 0.00552 +Mul 152284.216 0.00551 0.00553 0.00552 +Add 150549.336 0.00836 0.00838 0.00837 +Triad 150597.842 0.00836 0.00838 0.00837 diff --git a/results/cuda/nvidia-k40c-ecc-on.txt b/results/cuda/nvidia-k40c-ecc-on.txt index e3cf693..210edf4 100644 --- a/results/cuda/nvidia-k40c-ecc-on.txt +++ b/results/cuda/nvidia-k40c-ecc-on.txt @@ -1,12 +1,14 @@ GPU-STREAM -Version: 0.0 +Version: 0.9 Implementation: CUDA +Precision: double -Warning: array size must divide 1024 -Resizing array from 50000000 to 49999872 +Running kernels 10 times +Array size: 400.0 MB (=0.4 GB) +Total size: 1200.0 MB (=1.2 GB) Using CUDA device Tesla K40c Function MBytes/sec Min (sec) Max Average -Copy 194203 0.00411939 0.00412812 0.00412312 -Mul 194111 0.00412135 0.00412981 0.00412629 -Add 191683 0.00626033 0.00626831 0.0062641 -Triad 191560 0.00626436 0.00628582 0.00626866 +Copy 194335.669 0.00432 0.00433 0.00432 +Mul 194171.527 0.00432 0.00433 0.00433 +Add 191294.438 0.00658 0.00659 0.00658 +Triad 191240.187 0.00658 0.00659 0.00658 diff --git a/results/opencl/nvidia-k20c-ecc-on.txt b/results/opencl/nvidia-k20c-ecc-on.txt index 5c65af6..dbdea64 100644 --- a/results/opencl/nvidia-k20c-ecc-on.txt +++ b/results/opencl/nvidia-k20c-ecc-on.txt @@ -1,10 +1,14 @@ GPU-STREAM -Version: 0.0 +Version: 0.9 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 Function MBytes/sec Min (sec) Max Average -Copy 149700 0.00534404 0.00537675 0.00535703 -Mul 149734 0.0053428 0.00537013 0.00535519 -Add 151108 0.00794132 0.00797839 0.00795825 -Triad 151008 0.0079466 0.00797772 0.0079599 +Copy 150397.867 0.00558 0.00561 0.00559 +Mul 150241.232 0.00558 0.00560 0.00559 +Add 151673.787 0.00830 0.00833 0.00831 +Triad 151699.186 0.00829 0.00833 0.00831 diff --git a/results/opencl/nvidia-k40c-ecc-on.txt b/results/opencl/nvidia-k40c-ecc-on.txt index a980d9b..ca78f1b 100644 --- a/results/opencl/nvidia-k40c-ecc-on.txt +++ b/results/opencl/nvidia-k40c-ecc-on.txt @@ -1,10 +1,14 @@ GPU-STREAM -Version: 0.0 +Version: 0.9 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 Function MBytes/sec Min (sec) Max Average -Copy 190343 0.00420295 0.00422588 0.00421376 -Mul 190266 0.00420465 0.00422835 0.00421347 -Add 190144 0.00631101 0.00632369 0.00631618 -Triad 190105 0.00631229 0.00632309 0.00631735 +Copy 190354.872 0.00441 0.00443 0.00442 +Mul 190199.107 0.00441 0.00443 0.00442 +Add 190946.380 0.00659 0.00660 0.00660 +Triad 190991.101 0.00659 0.00661 0.00660