t1 ARMv8 Cortex-A57 testing with a NVIDIA Jetson Nano Developer Kit and NVIDIA TEGRA on Ubuntu 18.04 via the Phoronix Test Suite.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 2107072-IB-T1161066526 NVIDIA TEGRA - ARMv8 Cortex-A57 Processor: ARMv8 Cortex-A57 @ 1.48GHz (4 Cores), Motherboard: NVIDIA Jetson Nano Developer Kit, Memory: 4096MB, Disk: 64GB SC64G, Graphics: NVIDIA TEGRA, Network: Realtek RTL8111/8168/8411
OS: Ubuntu 18.04, Kernel: 4.9.140-tegra (aarch64), Desktop: GNOME Shell 3.28.4, Display Server: X Server 1.19.6, Compiler: GCC 7.5.0 + Clang 6.0.0-1ubuntu2 + CUDA 10.2, File-System: ext4, Screen Resolution: 3840x2400
Kernel Notes: Transparent Huge Pages: alwaysCompiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only -vProcessor Notes: Scaling Governor: tegra-cpufreq schedutilPython Notes: Python 2.7.17 + Python 3.6.9
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: Vulkan GPU - Model: regnety_400m NVIDIA TEGRA - ARMv8 Cortex-A57 12 24 36 48 60 SE +/- 0.08, N = 3 53.70 MIN: 53.06 / MAX: 79.49 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: Vulkan GPU - Model: squeezenet_ssd NVIDIA TEGRA - ARMv8 Cortex-A57 20 40 60 80 100 SE +/- 0.11, N = 3 83.87 MIN: 82.79 / MAX: 93.9 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: Vulkan GPU - Model: yolov4-tiny NVIDIA TEGRA - ARMv8 Cortex-A57 30 60 90 120 150 SE +/- 0.04, N = 3 154.03 MIN: 152.84 / MAX: 173.17 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: Vulkan GPU - Model: resnet50 NVIDIA TEGRA - ARMv8 Cortex-A57 40 80 120 160 200 SE +/- 0.36, N = 3 198.84 MIN: 197.13 / MAX: 273.39 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: Vulkan GPU - Model: alexnet NVIDIA TEGRA - ARMv8 Cortex-A57 20 40 60 80 100 SE +/- 0.67, N = 3 102.97 MIN: 101.01 / MAX: 123.69 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: Vulkan GPU - Model: resnet18 NVIDIA TEGRA - ARMv8 Cortex-A57 20 40 60 80 100 SE +/- 0.04, N = 3 81.57 MIN: 80.9 / MAX: 88.75 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: Vulkan GPU - Model: vgg16 NVIDIA TEGRA - ARMv8 Cortex-A57 100 200 300 400 500 SE +/- 0.17, N = 3 444.93 MIN: 440.85 / MAX: 483.42 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: Vulkan GPU - Model: googlenet NVIDIA TEGRA - ARMv8 Cortex-A57 20 40 60 80 100 SE +/- 0.04, N = 3 91.27 MIN: 90.57 / MAX: 103.84 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: Vulkan GPU - Model: blazeface NVIDIA TEGRA - ARMv8 Cortex-A57 2 4 6 8 10 SE +/- 0.01, N = 3 7.83 MIN: 7.64 / MAX: 9.29 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: Vulkan GPU - Model: efficientnet-b0 NVIDIA TEGRA - ARMv8 Cortex-A57 15 30 45 60 75 SE +/- 0.07, N = 3 65.67 MIN: 65.03 / MAX: 72.38 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: Vulkan GPU - Model: mnasnet NVIDIA TEGRA - ARMv8 Cortex-A57 7 14 21 28 35 SE +/- 0.00, N = 3 30.70 MIN: 30.27 / MAX: 34.94 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: Vulkan GPU - Model: shufflenet-v2 NVIDIA TEGRA - ARMv8 Cortex-A57 5 10 15 20 25 SE +/- 0.03, N = 3 21.44 MIN: 21.06 / MAX: 27.71 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 NVIDIA TEGRA - ARMv8 Cortex-A57 7 14 21 28 35 SE +/- 0.04, N = 3 29.22 MIN: 28.88 / MAX: 55.01 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 NVIDIA TEGRA - ARMv8 Cortex-A57 7 14 21 28 35 SE +/- 0.04, N = 3 32.15 MIN: 31.71 / MAX: 55.76 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20210525 Target: Vulkan GPU - Model: mobilenet NVIDIA TEGRA - ARMv8 Cortex-A57 20 40 60 80 100 SE +/- 0.10, N = 3 111.09 MIN: 110.15 / MAX: 133.95 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread
ViennaCL ViennaCL is an open-source linear algebra library written in C++ and with support for OpenCL and OpenMP. This test profile makes use of ViennaCL's built-in benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-TT NVIDIA TEGRA - ARMv8 Cortex-A57 0.4815 0.963 1.4445 1.926 2.4075 SE +/- 0.00, N = 3 2.14 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-TN NVIDIA TEGRA - ARMv8 Cortex-A57 0.5558 1.1116 1.6674 2.2232 2.779 SE +/- 0.00, N = 3 2.47 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-NT NVIDIA TEGRA - ARMv8 Cortex-A57 0.4208 0.8416 1.2624 1.6832 2.104 SE +/- 0.01, N = 3 1.87 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.org GFLOPs/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-NN NVIDIA TEGRA - ARMv8 Cortex-A57 0.4838 0.9676 1.4514 1.9352 2.419 SE +/- 0.04, N = 3 2.15 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMV-T NVIDIA TEGRA - ARMv8 Cortex-A57 1.2983 2.5966 3.8949 5.1932 6.4915 SE +/- 0.01, N = 3 5.77 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dGEMV-N NVIDIA TEGRA - ARMv8 Cortex-A57 1.215 2.43 3.645 4.86 6.075 SE +/- 0.02, N = 3 5.40 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dDOT NVIDIA TEGRA - ARMv8 Cortex-A57 2 4 6 8 10 SE +/- 0.01, N = 3 6.89 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dAXPY NVIDIA TEGRA - ARMv8 Cortex-A57 2 4 6 8 10 SE +/- 0.07, N = 3 7.97 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - dCOPY NVIDIA TEGRA - ARMv8 Cortex-A57 3 6 9 12 15 SE +/- 0.00, N = 3 10.6 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - sDOT NVIDIA TEGRA - ARMv8 Cortex-A57 2 4 6 8 10 SE +/- 0.05, N = 3 6.61 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - sAXPY NVIDIA TEGRA - ARMv8 Cortex-A57 2 4 6 8 10 SE +/- 0.04, N = 3 7.89 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
OpenBenchmarking.org GB/s, More Is Better ViennaCL 1.7.1 Test: CPU BLAS - sCOPY NVIDIA TEGRA - ARMv8 Cortex-A57 3 6 9 12 15 SE +/- 0.04, N = 3 9.84 1. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL
NVIDIA TEGRA - ARMv8 Cortex-A57 Processor: ARMv8 Cortex-A57 @ 1.48GHz (4 Cores), Motherboard: NVIDIA Jetson Nano Developer Kit, Memory: 4096MB, Disk: 64GB SC64G, Graphics: NVIDIA TEGRA, Network: Realtek RTL8111/8168/8411
OS: Ubuntu 18.04, Kernel: 4.9.140-tegra (aarch64), Desktop: GNOME Shell 3.28.4, Display Server: X Server 1.19.6, Compiler: GCC 7.5.0 + Clang 6.0.0-1ubuntu2 + CUDA 10.2, File-System: ext4, Screen Resolution: 3840x2400
Kernel Notes: Transparent Huge Pages: alwaysCompiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only -vProcessor Notes: Scaling Governor: tegra-cpufreq schedutilPython Notes: Python 2.7.17 + Python 3.6.9
Testing initiated at 6 July 2021 21:35 by user atm26.