nvidia-gpu-compute-oring-agx-32g

ARMv8 Cortex-A78E testing with a NVIDIA Jetson AGX Orin Developer Kit (36.3.0-gcid-36191598 BIOS) and Orin on Ubuntu 22.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2408272-NE-NVIDIAGPU37.

nvidia-gpu-compute-oring-agx-32gProcessorMotherboardMemoryDiskGraphicsNetworkOSKernelDesktopDisplay ServerDisplay DriverVulkanCompilerFile-SystemScreen ResolutionbaselineARMv8 Cortex-A78E @ 2.20GHz (12 Cores)NVIDIA Jetson AGX Orin Developer Kit (36.3.0-gcid-36191598 BIOS)30GB1000GB Samsung SSD 960 EVO 1TB + 64GB G1M15MOrinRealtek RTL8822CE 802.11ac PCIeUbuntu 22.045.15.136-tegra (aarch64)GNOME Shell 42.9X Server 1.21.1.4NVIDIA1.3.251GCC 11.4.0 + CUDA 12.2ext46582x1234OpenBenchmarking.org- Transparent Huge Pages: always- --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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v - Scaling Governor: tegra194 performance- BAR1 / Visible vRAM Size: N/A- Python 3.10.12- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 but not BHB + srbds: Not affected + tsx_async_abort: Not affected

nvidia-gpu-compute-oring-agx-32gvkfft: FFT + iFFT R2C / C2Rvkfft: FFT + iFFT C2C 1D batched in half precisionvkfft: FFT + iFFT C2C Bluestein in single precisionvkfft: FFT + iFFT C2C 1D batched in double precisionvkfft: FFT + iFFT C2C 1D batched in single precisionvkfft: FFT + iFFT C2C multidimensional in single precisionvkfft: FFT + iFFT C2C Bluestein benchmark in double precisionvkfft: FFT + iFFT C2C 1D batched in single precision, no reshufflingvkresample: 2x - Doublevkresample: 2x - Singleviennacl: CPU BLAS - sCOPYviennacl: CPU BLAS - sAXPYviennacl: CPU BLAS - sDOTviennacl: CPU BLAS - dCOPYviennacl: CPU BLAS - dAXPYviennacl: CPU BLAS - dDOTviennacl: CPU BLAS - dGEMV-Nviennacl: CPU BLAS - dGEMV-Tviennacl: CPU BLAS - dGEMM-NNviennacl: CPU BLAS - dGEMM-NTviennacl: CPU BLAS - dGEMM-TNviennacl: CPU BLAS - dGEMM-TTncnn: Vulkan GPU - mobilenetncnn: Vulkan GPU-v2-v2 - mobilenet-v2ncnn: Vulkan GPU-v3-v3 - mobilenet-v3ncnn: Vulkan GPU - shufflenet-v2ncnn: Vulkan GPU - mnasnetncnn: Vulkan GPU - efficientnet-b0ncnn: Vulkan GPU - blazefacencnn: Vulkan GPU - googlenetncnn: Vulkan GPU - vgg16ncnn: Vulkan GPU - resnet18ncnn: Vulkan GPU - alexnetncnn: Vulkan GPU - resnet50ncnn: Vulkan GPUv2-yolov3v2-yolov3 - mobilenetv2-yolov3ncnn: Vulkan GPU - yolov4-tinyncnn: Vulkan GPU - squeezenet_ssdncnn: Vulkan GPU - regnety_400mncnn: Vulkan GPU - vision_transformerncnn: Vulkan GPU - FastestDetbaseline118183191438673569264481084473428743512.01856.73864.479.463.964.176.061.847.462.127.323.526.326.616.103.233.102.752.995.001.438.7134.686.146.4216.1316.1025.0610.829.84267.883.56OpenBenchmarking.org

VkFFT

Test: FFT + iFFT R2C / C2R

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.3.4Test: FFT + iFFT R2C / C2Rbaseline3K6K9K12K15KSE +/- 34.23, N = 3118181. (CXX) g++ options: -O3

VkFFT

Test: FFT + iFFT C2C 1D batched in half precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.3.4Test: FFT + iFFT C2C 1D batched in half precisionbaseline7K14K21K28K35KSE +/- 122.59, N = 3319141. (CXX) g++ options: -O3

VkFFT

Test: FFT + iFFT C2C Bluestein in single precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.3.4Test: FFT + iFFT C2C Bluestein in single precisionbaseline8001600240032004000SE +/- 2.91, N = 338671. (CXX) g++ options: -O3

VkFFT

Test: FFT + iFFT C2C 1D batched in double precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.3.4Test: FFT + iFFT C2C 1D batched in double precisionbaseline8001600240032004000SE +/- 17.23, N = 335691. (CXX) g++ options: -O3

VkFFT

Test: FFT + iFFT C2C 1D batched in single precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.3.4Test: FFT + iFFT C2C 1D batched in single precisionbaseline6K12K18K24K30KSE +/- 47.76, N = 3264481. (CXX) g++ options: -O3

VkFFT

Test: FFT + iFFT C2C multidimensional in single precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.3.4Test: FFT + iFFT C2C multidimensional in single precisionbaseline2K4K6K8K10KSE +/- 26.42, N = 3108441. (CXX) g++ options: -O3

VkFFT

Test: FFT + iFFT C2C Bluestein benchmark in double precision

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.3.4Test: FFT + iFFT C2C Bluestein benchmark in double precisionbaseline160320480640800SE +/- 0.58, N = 37341. (CXX) g++ options: -O3

VkFFT

Test: FFT + iFFT C2C 1D batched in single precision, no reshuffling

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.3.4Test: FFT + iFFT C2C 1D batched in single precision, no reshufflingbaseline6K12K18K24K30KSE +/- 23.30, N = 3287431. (CXX) g++ options: -O3

VkResample

Upscale: 2x - Precision: Double

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Doublebaseline110220330440550SE +/- 0.02, N = 3512.021. (CXX) g++ options: -O3

VkResample

Upscale: 2x - Precision: Single

OpenBenchmarking.orgms, Fewer Is BetterVkResample 1.0Upscale: 2x - Precision: Singlebaseline1326395265SE +/- 0.04, N = 356.741. (CXX) g++ options: -O3

ViennaCL

Test: CPU BLAS - sCOPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sCOPYbaseline1428425670SE +/- 0.19, N = 364.41. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - sAXPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sAXPYbaseline20406080100SE +/- 0.74, N = 379.41. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - sDOT

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - sDOTbaseline1428425670SE +/- 0.43, N = 363.91. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dCOPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dCOPYbaseline1428425670SE +/- 0.73, N = 364.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dAXPY

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dAXPYbaseline20406080100SE +/- 0.26, N = 376.01. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dDOT

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dDOTbaseline1428425670SE +/- 0.36, N = 361.81. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dGEMV-N

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-Nbaseline1122334455SE +/- 2.85, N = 347.41. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dGEMV-T

OpenBenchmarking.orgGB/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMV-Tbaseline1428425670SE +/- 0.98, N = 362.11. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dGEMM-NN

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NNbaseline612182430SE +/- 0.03, N = 327.31. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dGEMM-NT

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-NTbaseline612182430SE +/- 0.00, N = 323.51. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dGEMM-TN

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TNbaseline612182430SE +/- 0.43, N = 326.31. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

ViennaCL

Test: CPU BLAS - dGEMM-TT

OpenBenchmarking.orgGFLOPs/s, More Is BetterViennaCL 1.7.1Test: CPU BLAS - dGEMM-TTbaseline612182430SE +/- 0.15, N = 326.61. (CXX) g++ options: -fopenmp -O3 -rdynamic -lOpenCL

NCNN

Target: Vulkan GPU - Model: mobilenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: mobilenetbaseline48121620SE +/- 0.54, N = 1216.10MIN: 10.31 / MAX: 86.731. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2baseline0.72681.45362.18042.90723.634SE +/- 0.03, N = 123.23MIN: 3.07 / MAX: 49.981. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3baseline0.69751.3952.09252.793.4875SE +/- 0.02, N = 123.10MIN: 2.93 / MAX: 27.151. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: shufflenet-v2

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: shufflenet-v2baseline0.61881.23761.85642.47523.094SE +/- 0.03, N = 122.75MIN: 2.58 / MAX: 54.91. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: mnasnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: mnasnetbaseline0.67281.34562.01842.69123.364SE +/- 0.03, N = 122.99MIN: 2.82 / MAX: 36.771. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: efficientnet-b0

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: efficientnet-b0baseline1.1252.253.3754.55.625SE +/- 0.02, N = 125.00MIN: 4.8 / MAX: 53.161. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: blazeface

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: blazefacebaseline0.32180.64360.96541.28721.609SE +/- 0.05, N = 121.43MIN: 1.25 / MAX: 38.681. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: googlenet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: googlenetbaseline246810SE +/- 0.10, N = 128.71MIN: 8.13 / MAX: 94.741. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: vgg16

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: vgg16baseline816243240SE +/- 0.35, N = 1234.68MIN: 27.78 / MAX: 73.921. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: resnet18

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: resnet18baseline246810SE +/- 0.05, N = 126.14MIN: 5.81 / MAX: 40.111. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: alexnet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: alexnetbaseline246810SE +/- 0.16, N = 126.42MIN: 5.86 / MAX: 24.851. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: resnet50

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: resnet50baseline48121620SE +/- 0.51, N = 1216.13MIN: 13.81 / MAX: 89.71. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPUv2-yolov3v2-yolov3 - Model: mobilenetv2-yolov3

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPUv2-yolov3v2-yolov3 - Model: mobilenetv2-yolov3baseline48121620SE +/- 0.54, N = 1216.10MIN: 10.31 / MAX: 86.731. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: yolov4-tiny

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: yolov4-tinybaseline612182430SE +/- 0.44, N = 1225.06MIN: 17.44 / MAX: 68.091. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: squeezenet_ssd

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: squeezenet_ssdbaseline3691215SE +/- 0.29, N = 1210.82MIN: 9.19 / MAX: 92.971. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: regnety_400m

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: regnety_400mbaseline3691215SE +/- 0.07, N = 129.84MIN: 9.23 / MAX: 218.631. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: vision_transformer

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: vision_transformerbaseline60120180240300SE +/- 0.75, N = 12267.88MIN: 228.72 / MAX: 358.851. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

NCNN

Target: Vulkan GPU - Model: FastestDet

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: FastestDetbaseline0.8011.6022.4033.2044.005SE +/- 0.06, N = 123.56MIN: 3.22 / MAX: 68.521. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread


Phoronix Test Suite v10.8.5