EPYC Tyan Server

AMD EPYC 7543 32-Core testing with a TYAN S8036GM2NE-LE (V2.00.B21 BIOS) and ASPEED on Debian 11 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 2106220-IB-2106212IB87
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Limit displaying results to tests within:

AV1 2 Tests
BLAS (Basic Linear Algebra Sub-Routine) Tests 2 Tests
Timed Code Compilation 6 Tests
C/C++ Compiler Tests 5 Tests
CPU Massive 13 Tests
Creator Workloads 15 Tests
Encoding 5 Tests
Fortran Tests 4 Tests
Game Development 5 Tests
HPC - High Performance Computing 17 Tests
Machine Learning 7 Tests
Molecular Dynamics 6 Tests
MPI Benchmarks 6 Tests
Multi-Core 22 Tests
NVIDIA GPU Compute 4 Tests
Intel oneAPI 5 Tests
OpenMPI Tests 8 Tests
Programmer / Developer System Benchmarks 6 Tests
Python Tests 9 Tests
Renderers 3 Tests
Scientific Computing 8 Tests
Server CPU Tests 9 Tests
Video Encoding 5 Tests
Common Workstation Benchmarks 2 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
Ubuntu 20.04.2 LTS
June 20 2021
  10 Hours, 37 Minutes
Fedora Server 34
June 21 2021
  9 Hours, 6 Minutes
Debian 10.10
June 21 2021
  8 Hours, 37 Minutes
Debian Bullseye
June 22 2021
  9 Hours, 51 Minutes
Invert Hiding All Results Option
  9 Hours, 33 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


EPYC Tyan ServerProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerOpenGLCompilerFile-SystemScreen ResolutionVulkanUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian BullseyeAMD EPYC 7543 32-Core @ 2.80GHz (32 Cores / 64 Threads)TYAN S8036GM2NE-LE (V2.00.B21 BIOS)AMD Starship/Matisse64GB1000GB Western Digital WD_BLACK SN850 1TBASPEEDVE2282 x Broadcom NetXtreme BCM5720 2-port PCIeUbuntu 20.045.8.0-55-generic (x86_64)GNOME Shell 3.36.4X Server 1.20.94.5 Mesa 20.2.6 (LLVM 11.0.0 256 bits)GCC 9.3.0ext41920x10802 x Broadcom NetXtreme BCM5720 PCIeFedora 345.12.11-300.fc34.x86_64 (x86_64)GCC 11.1.1 20210531xfsDebian 104.19.0-17-amd64 (x86_64)GNOME Shell 3.30.2X ServerGCC 8.3.0ext42 x Broadcom NetXtreme BCM5720 2-port PCIeDebian 115.10.0-7-amd64 (x86_64)GNOME Shell 3.38.41.0.2GCC 10.2.1 20210110OpenBenchmarking.orgKernel Details- Ubuntu 20.04.2 LTS: Transparent Huge Pages: madvise- Fedora Server 34: Transparent Huge Pages: madvise- Debian 10.10: Transparent Huge Pages: always- Debian Bullseye: Transparent Huge Pages: alwaysCompiler Details- Ubuntu 20.04.2 LTS: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - Fedora Server 34: --build=x86_64-redhat-linux --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-initfini-array --enable-languages=c,c++,fortran,objc,obj-c++,ada,go,d,lto --enable-multilib --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=i686 --with-gcc-major-version-only --with-linker-hash-style=gnu --with-tune=generic --without-cuda-driver - Debian 10.10: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib --with-tune=generic --without-cuda-driver -v - Debian Bullseye: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-mutex --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-Km9U7s/gcc-10-10.2.1/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-Km9U7s/gcc-10-10.2.1/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details- Ubuntu 20.04.2 LTS: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0xa001119- Fedora Server 34: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa001119- Debian 10.10: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0xa001119- Debian Bullseye: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa001119Python Details- Ubuntu 20.04.2 LTS: Python 3.8.5- Fedora Server 34: Python 3.9.5- Debian 10.10: Python 2.7.16 + Python 3.7.3- Debian Bullseye: Python 3.9.2Security Details- Ubuntu 20.04.2 LTS: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected - Fedora Server 34: SELinux + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected - Debian 10.10: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected - Debian Bullseye: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian BullseyeResult OverviewPhoronix Test Suite100%170%240%310%NAS Parallel BenchmarksVP9 libvpx EncodingminiFESVT-VP9Timed LLVM CompilationSVT-AV1Timed Node.js Compilationlibavif avifencPlaidMLTimed Linux Kernel CompilationTimed Godot Game Engine CompilationMobile Neural NetworkSVT-HEVCIntel Open Image DenoiseEmbreeONNX RuntimeOpenVKLTNNBRL-CADOSPrayBlenderLuxCoreRenderKripkeNAMDTensorFlow Lite

EPYC Tyan Servernumenta-nab: EXPoSEnwchem: C240 Buckyballhpcg: openvkl: vklBenchmarkonnx: bertsquad-10 - OpenMP CPUonnx: fcn-resnet101-11 - OpenMP CPUonnx: yolov4 - OpenMP CPUmnn: inception-v3mnn: mobilenet-v1-1.0mnn: MobileNetV2_224mnn: SqueezeNetV1.0mnn: resnet-v2-50mnn: squeezenetv1.1mnn: mobilenetV3openfoam: Motorbike 60Mbuild-llvm: Unix Makefilestensorflow-lite: NASNet Mobileplaidml: No - Inference - ResNet 50 - CPUbrl-cad: VGR Performance Metricnpb: EP.Donnx: super-resolution-10 - OpenMP CPUtnn: CPU - DenseNetbuild-llvm: Ninjakripke: blender: Barbershop - CPU-Onlyblender: Classroom - CPU-Onlybuild-nodejs: Time To Compilenpb: BT.Cminife: Smallonnx: shufflenet-v2-10 - OpenMP CPUgpaw: Carbon Nanotubeplaidml: No - Inference - VGG19 - CPUvpxenc: Speed 0 - Bosphorus 4Ksvt-av1: Preset 4 - Bosphorus 4Kplaidml: No - Inference - VGG16 - CPUospray: San Miguel - Path Tracerincompact3d: input.i3d 193 Cells Per Directionnumenta-nab: Earthgecko Skylineopenvino: Person Detection 0106 FP16 - CPUopenvino: Person Detection 0106 FP16 - CPUopenvino: Person Detection 0106 FP32 - CPUopenvino: Person Detection 0106 FP32 - CPUnpb: SP.Copenvino: Face Detection 0106 FP16 - CPUopenvino: Face Detection 0106 FP16 - CPUopenvino: Face Detection 0106 FP32 - CPUopenvino: Face Detection 0106 FP32 - CPUoidn: RTLightmap.hdr.4096x4096luxcorerender: Orange Juice - CPUluxcorerender: Danish Mood - CPUluxcorerender: LuxCore Benchmark - CPUtensorflow-lite: Inception V4luxcorerender: DLSC - CPUtensorflow-lite: Inception ResNet V2tensorflow-lite: SqueezeNettensorflow-lite: Mobilenet Quanttensorflow-lite: Mobilenet Floatopenvino: Age Gender Recognition Retail 0013 FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP32 - CPUopenvino: Age Gender Recognition Retail 0013 FP32 - CPUbuild-linux-kernel: Time To Compilebuild-godot: Time To Compilegromacs: MPI CPU - water_GMX50_barebuild-wasmer: Time To Compileblender: BMW27 - CPU-Onlyospray: XFrog Forest - Path Tracervpxenc: Speed 5 - Bosphorus 4Knamd: ATPase Simulation - 327,506 Atomsxmrig: Monero - 1Mnpb: LU.Cluxcorerender: Rainbow Colors and Prism - CPUxmrig: Wownero - 1Moidn: RT.ldr_alb_nrm.3840x2160embree: Pathtracer ISPC - Asian Dragonsvt-av1: Preset 8 - Bosphorus 4Koidn: RT.hdr_alb_nrm.3840x2160npb: IS.Dnumenta-nab: Bayesian Changepointopenfoam: Motorbike 30Mavifenc: 6, Losslessastcenc: Exhaustivesvt-hevc: 1 - Bosphorus 1080ptnn: CPU - MobileNet v2build-mesa: Time To Compileembree: Pathtracer ISPC - Crownospray: NASA Streamlines - Path Tracertnn: CPU - SqueezeNet v1.1npb: MG.Cnpb: FT.Cnumenta-nab: Relative Entropyavifenc: 10, Losslessavifenc: 10svt-hevc: 10 - Bosphorus 1080pnpb: CG.Csvt-vp9: VMAF Optimized - Bosphorus 1080pnumenta-nab: Windowed Gaussiansvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080ptnn: CPU - SqueezeNet v2ospray: Magnetic Reconnection - Path Tracersvt-vp9: Visual Quality Optimized - Bosphorus 1080psvt-hevc: 7 - Bosphorus 1080pUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye1397.7763116.912.87782974459329124.3612.8493.9476.09421.9544.1442.572237.38286.37588673.19.044224812934.2647752907.266232.75878658979195.79146.51151.64787954.9811203.0410866114.89622.496.161.93126.733.7227.876948076.4143486.444.573448.714.5833234.282548.126.252551.276.250.458.294.324.739366945.4583553969192.646472.943688.40.7620273.780.7620458.8735.13757.9073.63961.24152.854.4713.630.6725221914.9102552.8416.3823123.10.9037.186520.8510.901786.8226.84222.2430.22223.677924.65290.33722.74733.318712.35269.18249265.3647059.6813.0476.2873.817458.1829891.71349.166.805353.9862.597500264.89282.3811.71302964408427722.6032.8943.9755.76718.9334.0422.535257.168111782.48.594177811601.2248843042.861209.69078470519196.64146.65137.58820507.5510017.791091122.136.891.91326.963.7029.036647810058.280.488.294.394.689418925.4683706570453.946557.743885.734.01954.9483.64041.49853.134.4616.230.6735521406.4516.110.9637.383818.3450.961092.1827.88123.516724.71296.54733.389312.35271.39123154.5417996.096.1853.620434.0713587.61291.76305.1464.811500272.55281.371120.7302824327523326.4713.4824.2426.32623.7364.7202.906223.26475742.39.6040912550182822.380185.20178433118192.55146.09123.13987888.3810918.401106427.956.051.85233.543.7375.36833843.850.448.334.364.749424085.4783472469846.246650.144222.929.49263.29352.444.4813.030.6719322334.517.1423003.30.8732.441020.7160.8735.81227.23725.19286.25432.587212.5268.00312.3215.3232.785537.01344.936.671353.7362.217444.44297.41296.573101.211.79412874247328227.6513.7964.0346.20422.5204.6762.616236.58250.57384817.39.614201562965.2145973098.333202.01378112133189.90145.38134.40789192.7713355.41128327.676.521.70733.873.7827.701865033101.400.468.394.424.799322515.4583470369107.346349.843711.532.03153.4263.75660.64152.554.509.310.6713123576.3105456.9516.7524478.30.9437.315314.8750.891731.3421.7727.39523.460524.96307.53420.27333.520412.66269.35850128.6049076.245.2982.873429.7130352.21256.35259.5662.165422.22236.32273.70OpenBenchmarking.org

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial timeseries data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: EXPoSEUbuntu 20.04.2 LTSDebian 10.1030060090012001500SE +/- 15.40, N = 3SE +/- 6.56, N = 31397.781120.73
OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: EXPoSEUbuntu 20.04.2 LTSDebian 10.102004006008001000Min: 1377.84 / Avg: 1397.78 / Max: 1428.08Min: 1108.26 / Avg: 1120.73 / Max: 1130.47

NWChem

NWChem is an open-source high performance computational chemistry package. Per NWChem's documentation, "NWChem aims to provide its users with computational chemistry tools that are scalable both in their ability to treat large scientific computational chemistry problems efficiently, and in their use of available parallel computing resources from high-performance parallel supercomputers to conventional workstation clusters." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNWChem 7.0.2Input: C240 BuckyballUbuntu 20.04.2 LTSDebian Bullseye70014002100280035003116.93101.2-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lz1. (F9X) gfortran options: -lnwctask -lccsd -lmcscf -lselci -lmp2 -lmoints -lstepper -ldriver -loptim -lnwdft -lgradients -lcphf -lesp -lddscf -ldangchang -lguess -lhessian -lvib -lnwcutil -lrimp2 -lproperty -lsolvation -lnwints -lprepar -lnwmd -lnwpw -lofpw -lpaw -lpspw -lband -lnwpwlib -lcafe -lspace -lanalyze -lqhop -lpfft -ldplot -ldrdy -lvscf -lqmmm -lqmd -letrans -ltce -lbq -lmm -lcons -lperfm -ldntmc -lccca -ldimqm -lga -larmci -lpeigs -l64to32 -lopenblas -lpthread -lrt -llapack -lnwcblas -lmpi_usempif08 -lmpi_mpifh -lmpi -lcomex -lm -m64 -ffast-math -std=legacy -fdefault-integer-8 -finline-functions -O2

High Performance Conjugate Gradient

HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1Ubuntu 20.04.2 LTSFedora Server 34Debian Bullseye3691215SE +/- 0.01, N = 3SE +/- 0.60, N = 12SE +/- 0.64, N = 912.8811.7111.791. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1Ubuntu 20.04.2 LTSFedora Server 34Debian Bullseye48121620Min: 12.87 / Avg: 12.88 / Max: 12.89Min: 11.02 / Avg: 11.71 / Max: 18.31Min: 11.12 / Avg: 11.79 / Max: 16.91. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi

OpenVKL

OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye60120180240300SE +/- 4.05, N = 9SE +/- 1.45, N = 3SE +/- 2.31, N = 3SE +/- 3.51, N = 3297296282287MIN: 1 / MAX: 1469MIN: 1 / MAX: 896MIN: 1 / MAX: 852MIN: 1 / MAX: 1036
OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye50100150200250Min: 290 / Avg: 297 / Max: 329Min: 294 / Avg: 296.33 / Max: 299Min: 278 / Avg: 282 / Max: 286Min: 283 / Avg: 287 / Max: 294

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: bertsquad-10 - Device: OpenMP CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye100200300400500SE +/- 3.78, N = 12SE +/- 4.15, N = 12SE +/- 6.36, N = 12SE +/- 4.24, N = 124454404324241. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: bertsquad-10 - Device: OpenMP CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye80160240320400Min: 428.5 / Avg: 445.33 / Max: 474.5Min: 421.5 / Avg: 440.46 / Max: 465Min: 390 / Avg: 431.5 / Max: 460.5Min: 404 / Avg: 424.04 / Max: 4501. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: fcn-resnet101-11 - Device: OpenMP CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye20406080100SE +/- 4.09, N = 9SE +/- 3.41, N = 12SE +/- 1.93, N = 12SE +/- 3.37, N = 9938475731. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: fcn-resnet101-11 - Device: OpenMP CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye20406080100Min: 64 / Avg: 92.56 / Max: 102.5Min: 67 / Avg: 84.29 / Max: 104Min: 66.5 / Avg: 75.13 / Max: 86.5Min: 59.5 / Avg: 72.89 / Max: 931. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: yolov4 - Device: OpenMP CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye60120180240300SE +/- 4.38, N = 12SE +/- 1.96, N = 3SE +/- 4.62, N = 12SE +/- 4.80, N = 92912772332821. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: yolov4 - Device: OpenMP CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye50100150200250Min: 267 / Avg: 290.67 / Max: 320.5Min: 274.5 / Avg: 277.17 / Max: 281Min: 211.5 / Avg: 233.25 / Max: 273Min: 262.5 / Avg: 281.61 / Max: 3031. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: inception-v3Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye714212835SE +/- 0.22, N = 5SE +/- 0.14, N = 15SE +/- 0.23, N = 12SE +/- 0.10, N = 1524.3622.6026.4727.65MIN: 23.32 / MAX: 78.89MIN: 21.09 / MAX: 293.57MIN: 23.92 / MAX: 128.7MIN: 25.83 / MAX: 133.21. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: inception-v3Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye612182430Min: 23.95 / Avg: 24.36 / Max: 25.17Min: 21.87 / Avg: 22.6 / Max: 24.21Min: 25.17 / Avg: 26.47 / Max: 27.81Min: 27.09 / Avg: 27.65 / Max: 28.231. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: mobilenet-v1-1.0Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye0.85411.70822.56233.41644.2705SE +/- 0.077, N = 5SE +/- 0.057, N = 15SE +/- 0.086, N = 12SE +/- 0.083, N = 152.8492.8943.4823.796MIN: 2.57 / MAX: 7.42MIN: 2.55 / MAX: 7.14MIN: 2.98 / MAX: 5.1MIN: 2.99 / MAX: 6.121. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: mobilenet-v1-1.0Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810Min: 2.6 / Avg: 2.85 / Max: 3.06Min: 2.62 / Avg: 2.89 / Max: 3.45Min: 3.02 / Avg: 3.48 / Max: 3.91Min: 3.05 / Avg: 3.8 / Max: 4.321. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: MobileNetV2_224Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye0.95451.9092.86353.8184.7725SE +/- 0.095, N = 5SE +/- 0.046, N = 14SE +/- 0.072, N = 12SE +/- 0.107, N = 153.9473.9754.2424.034MIN: 3.62 / MAX: 5.86MIN: 3.37 / MAX: 43.2MIN: 3.8 / MAX: 6.21MIN: 3.48 / MAX: 7.151. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: MobileNetV2_224Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810Min: 3.66 / Avg: 3.95 / Max: 4.19Min: 3.7 / Avg: 3.97 / Max: 4.24Min: 3.84 / Avg: 4.24 / Max: 4.59Min: 3.55 / Avg: 4.03 / Max: 5.031. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: SqueezeNetV1.0Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810SE +/- 0.096, N = 5SE +/- 0.049, N = 15SE +/- 0.099, N = 12SE +/- 0.094, N = 156.0945.7676.3266.204MIN: 5.73 / MAX: 12.28MIN: 5.34 / MAX: 18.22MIN: 5.64 / MAX: 13.73MIN: 5.43 / MAX: 24.51. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: SqueezeNetV1.0Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye3691215Min: 5.87 / Avg: 6.09 / Max: 6.43Min: 5.56 / Avg: 5.77 / Max: 6.36Min: 5.85 / Avg: 6.33 / Max: 7.28Min: 5.68 / Avg: 6.2 / Max: 6.941. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: resnet-v2-50Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye612182430SE +/- 0.08, N = 5SE +/- 0.28, N = 15SE +/- 0.28, N = 12SE +/- 0.38, N = 1521.9518.9323.7422.52MIN: 20.25 / MAX: 89.25MIN: 17.69 / MAX: 159.34MIN: 22.27 / MAX: 64.14MIN: 20.6 / MAX: 109.281. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: resnet-v2-50Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye612182430Min: 21.65 / Avg: 21.95 / Max: 22.16Min: 17.92 / Avg: 18.93 / Max: 22Min: 22.73 / Avg: 23.74 / Max: 25.53Min: 20.89 / Avg: 22.52 / Max: 25.31. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: squeezenetv1.1Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye1.0622.1243.1864.2485.31SE +/- 0.037, N = 5SE +/- 0.035, N = 15SE +/- 0.046, N = 12SE +/- 0.029, N = 154.1444.0424.7204.676MIN: 4.02 / MAX: 6.05MIN: 3.75 / MAX: 13.49MIN: 4.17 / MAX: 23.32MIN: 4.15 / MAX: 20.31. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: squeezenetv1.1Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810Min: 4.06 / Avg: 4.14 / Max: 4.26Min: 3.8 / Avg: 4.04 / Max: 4.31Min: 4.42 / Avg: 4.72 / Max: 4.92Min: 4.43 / Avg: 4.68 / Max: 4.841. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: mobilenetV3Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye0.65391.30781.96172.61563.2695SE +/- 0.027, N = 5SE +/- 0.029, N = 15SE +/- 0.055, N = 12SE +/- 0.033, N = 152.5722.5352.9062.616MIN: 2.44 / MAX: 4.26MIN: 2.28 / MAX: 3.47MIN: 2.52 / MAX: 6.63MIN: 2.33 / MAX: 5.481. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.2Model: mobilenetV3Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810Min: 2.47 / Avg: 2.57 / Max: 2.63Min: 2.33 / Avg: 2.54 / Max: 2.65Min: 2.6 / Avg: 2.91 / Max: 3.29Min: 2.43 / Avg: 2.62 / Max: 2.851. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenFOAM

OpenFOAM is the leading free, open source software for computational fluid dynamics (CFD). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 60MUbuntu 20.04.2 LTSDebian Bullseye50100150200250SE +/- 0.07, N = 3SE +/- 0.08, N = 3237.38236.581. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 60MUbuntu 20.04.2 LTSDebian Bullseye4080120160200Min: 237.24 / Avg: 237.38 / Max: 237.46Min: 236.42 / Avg: 236.58 / Max: 236.691. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

Timed LLVM Compilation

This test times how long it takes to build the LLVM compiler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 12.0Build System: Unix MakefilesUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye60120180240300SE +/- 1.53, N = 3SE +/- 1.83, N = 3SE +/- 3.17, N = 3SE +/- 2.32, N = 3286.38257.17223.26250.57
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 12.0Build System: Unix MakefilesUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye50100150200250Min: 283.97 / Avg: 286.38 / Max: 289.22Min: 254.2 / Avg: 257.17 / Max: 260.5Min: 216.96 / Avg: 223.26 / Max: 227.03Min: 247.47 / Avg: 250.57 / Max: 255.12

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye20K40K60K80K100KSE +/- 1510.00, N = 15SE +/- 3807.02, N = 15SE +/- 362.65, N = 3SE +/- 2109.59, N = 1588673.1111782.475742.384817.3
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye20K40K60K80K100KMin: 77725 / Avg: 88673.13 / Max: 98496.3Min: 92171.2 / Avg: 111782.45 / Max: 146041Min: 75052.6 / Avg: 75742.33 / Max: 76281.5Min: 75430 / Avg: 84817.29 / Max: 102259

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye3691215SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.10, N = 3SE +/- 0.02, N = 39.048.599.609.61
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye3691215Min: 9.01 / Avg: 9.04 / Max: 9.09Min: 8.54 / Avg: 8.59 / Max: 8.62Min: 9.42 / Avg: 9.6 / Max: 9.78Min: 9.57 / Avg: 9.61 / Max: 9.64

BRL-CAD

BRL-CAD is a cross-platform, open-source solid modeling system with built-in benchmark mode. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.32.2VGR Performance MetricUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye90K180K270K360K450K4224814177814091254201561. (CXX) g++ options: -std=c++11 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -pthread -ldl -lm

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.DUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye6001200180024003000SE +/- 21.45, N = 13SE +/- 28.47, N = 14SE +/- 8.37, N = 32934.261601.222965.21-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian Bullseye: Open MPI 4.1.0
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.DUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye5001000150020002500Min: 2683.67 / Avg: 2934.26 / Max: 2976.57Min: 1278.9 / Avg: 1601.22 / Max: 1642.12Min: 2948.6 / Avg: 2965.21 / Max: 2975.281. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian Bullseye: Open MPI 4.1.0

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: super-resolution-10 - Device: OpenMP CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye11002200330044005500SE +/- 34.58, N = 3SE +/- 26.74, N = 3SE +/- 96.24, N = 12SE +/- 22.65, N = 347754884501845971. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: super-resolution-10 - Device: OpenMP CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye9001800270036004500Min: 4706 / Avg: 4775 / Max: 4813.5Min: 4832 / Avg: 4884.33 / Max: 4920Min: 4542 / Avg: 5018.38 / Max: 5523Min: 4570 / Avg: 4597 / Max: 46421. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: DenseNetUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye7001400210028003500SE +/- 0.81, N = 3SE +/- 20.85, N = 3SE +/- 3.53, N = 3SE +/- 9.62, N = 32907.273042.862822.383098.33MIN: 2834.03 / MAX: 3318.39MIN: 2872.4 / MAX: 3417.28MIN: 2717.84 / MAX: 3264.89MIN: 2905.58 / MAX: 3528.681. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: DenseNetUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye5001000150020002500Min: 2906.41 / Avg: 2907.27 / Max: 2908.89Min: 3006.04 / Avg: 3042.86 / Max: 3078.21Min: 2817.58 / Avg: 2822.38 / Max: 2829.26Min: 3086.43 / Avg: 3098.33 / Max: 3117.371. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

Timed LLVM Compilation

This test times how long it takes to build the LLVM compiler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 12.0Build System: NinjaUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye50100150200250SE +/- 0.86, N = 3SE +/- 0.77, N = 3SE +/- 0.13, N = 3SE +/- 0.11, N = 3232.76209.69185.20202.01
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 12.0Build System: NinjaUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye4080120160200Min: 231.03 / Avg: 232.76 / Max: 233.66Min: 208.32 / Avg: 209.69 / Max: 210.98Min: 184.98 / Avg: 185.2 / Max: 185.44Min: 201.81 / Avg: 202.01 / Max: 202.2

Kripke

Kripke is a simple, scalable, 3D Sn deterministic particle transport code. Its primary purpose is to research how data layout, programming paradigms and architectures effect the implementation and performance of Sn transport. Kripke is developed by LLNL. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgThroughput FoM, More Is BetterKripke 1.2.4Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye20M40M60M80M100MSE +/- 1927474.75, N = 15SE +/- 2249941.31, N = 15SE +/- 2823751.60, N = 12SE +/- 1339128.86, N = 12786589797847051978433118781121331. (CXX) g++ options: -O3 -fopenmp
OpenBenchmarking.orgThroughput FoM, More Is BetterKripke 1.2.4Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye14M28M42M56M70MMin: 64469010 / Avg: 78658978.67 / Max: 88693090Min: 63559540 / Avg: 78470519.33 / Max: 91733410Min: 65937930 / Avg: 78433118.33 / Max: 94946430Min: 69893600 / Avg: 78112133.33 / Max: 860080901. (CXX) g++ options: -O3 -fopenmp

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL, NVIDIA OptiX, and NVIDIA CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Barbershop - Compute: CPU-OnlyUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye4080120160200SE +/- 0.05, N = 3SE +/- 0.29, N = 3SE +/- 0.98, N = 3SE +/- 0.28, N = 3195.79196.64192.55189.90
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Barbershop - Compute: CPU-OnlyUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye4080120160200Min: 195.72 / Avg: 195.79 / Max: 195.88Min: 196.15 / Avg: 196.64 / Max: 197.15Min: 190.89 / Avg: 192.55 / Max: 194.28Min: 189.36 / Avg: 189.9 / Max: 190.28

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: CPU-OnlyUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye306090120150SE +/- 0.13, N = 3SE +/- 0.10, N = 3SE +/- 0.27, N = 3SE +/- 0.13, N = 3146.51146.65146.09145.38
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: CPU-OnlyUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye306090120150Min: 146.24 / Avg: 146.51 / Max: 146.67Min: 146.45 / Avg: 146.65 / Max: 146.8Min: 145.56 / Avg: 146.09 / Max: 146.48Min: 145.15 / Avg: 145.38 / Max: 145.59

Timed Node.js Compilation

This test profile times how long it takes to build/compile Node.js itself from source. Node.js is a JavaScript run-time built from the Chrome V8 JavaScript engine while itself is written in C/C++. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 15.11Time To CompileUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye306090120150SE +/- 0.10, N = 3SE +/- 0.21, N = 3SE +/- 0.17, N = 3SE +/- 0.16, N = 3151.65137.59123.14134.41
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 15.11Time To CompileUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye306090120150Min: 151.52 / Avg: 151.65 / Max: 151.85Min: 137.24 / Avg: 137.59 / Max: 137.96Min: 122.81 / Avg: 123.14 / Max: 123.32Min: 134.1 / Avg: 134.41 / Max: 134.63

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: BT.CUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye20K40K60K80K100KSE +/- 203.96, N = 3SE +/- 482.34, N = 9SE +/- 319.12, N = 3SE +/- 44.18, N = 387954.9820507.5587888.3889192.77-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian 10.10: Open MPI 3.1.34. Debian Bullseye: Open MPI 4.1.0
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: BT.CUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye15K30K45K60K75KMin: 87645.48 / Avg: 87954.98 / Max: 88339.84Min: 16918.37 / Avg: 20507.55 / Max: 21270.48Min: 87259.16 / Avg: 87888.38 / Max: 88295.64Min: 89104.44 / Avg: 89192.77 / Max: 89239.061. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian 10.10: Open MPI 3.1.34. Debian Bullseye: Open MPI 4.1.0

miniFE

MiniFE Finite Element is an application for unstructured implicit finite element codes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgCG Mflops, More Is BetterminiFE 2.2Problem Size: SmallUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye3K6K9K12K15KSE +/- 260.75, N = 15SE +/- 203.18, N = 12SE +/- 241.94, N = 15SE +/- 374.12, N = 1211203.0410017.7910918.4013355.401. (CXX) g++ options: -O3 -fopenmp -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgCG Mflops, More Is BetterminiFE 2.2Problem Size: SmallUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye2K4K6K8K10KMin: 9938.39 / Avg: 11203.04 / Max: 13203.5Min: 8854.68 / Avg: 10017.79 / Max: 11284.3Min: 8624.73 / Avg: 10918.4 / Max: 12216.4Min: 11293.7 / Avg: 13355.43 / Max: 16651.51. (CXX) g++ options: -O3 -fopenmp -pthread -lmpi_cxx -lmpi

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: shufflenet-v2-10 - Device: OpenMP CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye2K4K6K8K10KSE +/- 51.37, N = 3SE +/- 78.91, N = 3SE +/- 108.89, N = 3SE +/- 51.39, N = 3108661091111064112831. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: shufflenet-v2-10 - Device: OpenMP CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye2K4K6K8K10KMin: 10783 / Avg: 10866.17 / Max: 10960Min: 10785.5 / Avg: 10910.67 / Max: 11056.5Min: 10850.5 / Avg: 11064.33 / Max: 11207Min: 11189 / Avg: 11283 / Max: 113661. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

GPAW

GPAW is a density-functional theory (DFT) Python code based on the projector-augmented wave (PAW) method and the atomic simulation environment (ASE). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGPAW 20.1Input: Carbon NanotubeUbuntu 20.04.2 LTS306090120150SE +/- 0.64, N = 3114.901. (CC) gcc options: -pthread -shared -fwrapv -O2 -lxc -lblas -lmpi

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG19 - Device: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye714212835SE +/- 0.19, N = 3SE +/- 0.23, N = 5SE +/- 0.32, N = 3SE +/- 0.24, N = 322.4922.1327.9527.67
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG19 - Device: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye612182430Min: 22.14 / Avg: 22.49 / Max: 22.78Min: 21.46 / Avg: 22.13 / Max: 22.76Min: 27.41 / Avg: 27.95 / Max: 28.52Min: 27.19 / Avg: 27.67 / Max: 27.98

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 0 - Input: Bosphorus 4KUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 36.166.896.056.521. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 0 - Input: Bosphorus 4KUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye3691215Min: 6.15 / Avg: 6.16 / Max: 6.17Min: 6.85 / Avg: 6.89 / Max: 6.96Min: 6.04 / Avg: 6.05 / Max: 6.07Min: 6.46 / Avg: 6.52 / Max: 6.631. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 4 - Input: Bosphorus 4KUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye0.43450.8691.30351.7382.1725SE +/- 0.001, N = 3SE +/- 0.004, N = 3SE +/- 0.003, N = 3SE +/- 0.005, N = 31.9311.9131.8521.7071. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 4 - Input: Bosphorus 4KUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810Min: 1.93 / Avg: 1.93 / Max: 1.93Min: 1.91 / Avg: 1.91 / Max: 1.92Min: 1.85 / Avg: 1.85 / Max: 1.86Min: 1.7 / Avg: 1.71 / Max: 1.711. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye816243240SE +/- 0.23, N = 3SE +/- 0.19, N = 3SE +/- 0.25, N = 3SE +/- 0.46, N = 326.7326.9633.5433.87
OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye714212835Min: 26.27 / Avg: 26.73 / Max: 27.01Min: 26.58 / Avg: 26.96 / Max: 27.2Min: 33.09 / Avg: 33.54 / Max: 33.94Min: 33.2 / Avg: 33.87 / Max: 34.75

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: Path TracerUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye0.85051.7012.55153.4024.2525SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 33.723.703.733.78MIN: 2.87 / MAX: 3.75MIN: 3.01 / MAX: 3.73MIN: 2.99 / MAX: 3.75MIN: 2.73 / MAX: 3.8
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: Path TracerUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810Min: 3.72 / Avg: 3.72 / Max: 3.72Min: 3.68 / Avg: 3.7 / Max: 3.72Min: 3.73 / Avg: 3.73 / Max: 3.73Min: 3.77 / Avg: 3.78 / Max: 3.79

Xcompact3d Incompact3d

Xcompact3d Incompact3d is a Fortran-MPI based, finite difference high-performance code for solving the incompressible Navier-Stokes equation and as many as you need scalar transport equations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per DirectionUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye714212835SE +/- 0.28, N = 6SE +/- 0.09, N = 3SE +/- 0.20, N = 1527.8829.0427.70-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per DirectionUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye612182430Min: 27.25 / Avg: 27.88 / Max: 28.96Min: 28.9 / Avg: 29.04 / Max: 29.21Min: 26.39 / Avg: 27.7 / Max: 28.391. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial timeseries data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Earthgecko SkylineUbuntu 20.04.2 LTSDebian 10.1020406080100SE +/- 0.34, N = 3SE +/- 0.44, N = 376.4175.37
OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Earthgecko SkylineUbuntu 20.04.2 LTSDebian 10.101530456075Min: 75.76 / Avg: 76.41 / Max: 76.9Min: 74.84 / Avg: 75.37 / Max: 76.24

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2021.1Model: Person Detection 0106 FP16 - Device: CPUUbuntu 20.04.2 LTS7001400210028003500SE +/- 1.93, N = 33486.44

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2021.1Model: Person Detection 0106 FP16 - Device: CPUUbuntu 20.04.2 LTS1.02832.05663.08494.11325.1415SE +/- 0.00, N = 34.57

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2021.1Model: Person Detection 0106 FP32 - Device: CPUUbuntu 20.04.2 LTS7001400210028003500SE +/- 16.98, N = 33448.71

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2021.1Model: Person Detection 0106 FP32 - Device: CPUUbuntu 20.04.2 LTS1.03052.0613.09154.1225.1525SE +/- 0.01, N = 34.58

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.CUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye7K14K21K28K35KSE +/- 59.03, N = 3SE +/- 44.59, N = 3SE +/- 79.00, N = 3SE +/- 72.73, N = 333234.2810058.2833843.8533101.40-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian 10.10: Open MPI 3.1.34. Debian Bullseye: Open MPI 4.1.0
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.CUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye6K12K18K24K30KMin: 33117.39 / Avg: 33234.28 / Max: 33307.04Min: 9977.77 / Avg: 10058.28 / Max: 10131.76Min: 33749.33 / Avg: 33843.85 / Max: 34000.76Min: 32956.09 / Avg: 33101.4 / Max: 33179.861. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian 10.10: Open MPI 3.1.34. Debian Bullseye: Open MPI 4.1.0

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2021.1Model: Face Detection 0106 FP16 - Device: CPUUbuntu 20.04.2 LTS5001000150020002500SE +/- 0.08, N = 32548.12

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2021.1Model: Face Detection 0106 FP16 - Device: CPUUbuntu 20.04.2 LTS246810SE +/- 0.01, N = 36.25

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2021.1Model: Face Detection 0106 FP32 - Device: CPUUbuntu 20.04.2 LTS5001000150020002500SE +/- 4.77, N = 32551.27

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2021.1Model: Face Detection 0106 FP32 - Device: CPUUbuntu 20.04.2 LTS246810SE +/- 0.02, N = 36.25

Intel Open Image Denoise

Open Image Denoise is a denoising library for ray-tracing and part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RTLightmap.hdr.4096x4096Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye0.1080.2160.3240.4320.54SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.450.480.440.46
OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RTLightmap.hdr.4096x4096Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye12345Min: 0.45 / Avg: 0.45 / Max: 0.46Min: 0.48 / Avg: 0.48 / Max: 0.48Min: 0.43 / Avg: 0.44 / Max: 0.44Min: 0.46 / Avg: 0.46 / Max: 0.46

LuxCoreRender

LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Orange Juice - Acceleration: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 38.298.298.338.39MIN: 7.33 / MAX: 8.75MIN: 7.33 / MAX: 8.74MIN: 7.33 / MAX: 8.83MIN: 7.42 / MAX: 8.88
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Orange Juice - Acceleration: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye3691215Min: 8.28 / Avg: 8.29 / Max: 8.3Min: 8.26 / Avg: 8.29 / Max: 8.31Min: 8.32 / Avg: 8.33 / Max: 8.35Min: 8.35 / Avg: 8.39 / Max: 8.44

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Danish Mood - Acceleration: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye0.99451.9892.98353.9784.9725SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 34.324.394.364.42MIN: 1.72 / MAX: 5.09MIN: 1.77 / MAX: 5.1MIN: 1.73 / MAX: 5.08MIN: 1.72 / MAX: 5.11
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Danish Mood - Acceleration: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810Min: 4.3 / Avg: 4.32 / Max: 4.35Min: 4.32 / Avg: 4.39 / Max: 4.46Min: 4.31 / Avg: 4.36 / Max: 4.39Min: 4.41 / Avg: 4.42 / Max: 4.44

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: LuxCore Benchmark - Acceleration: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye1.07782.15563.23344.31125.389SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 34.734.684.744.79MIN: 1.81 / MAX: 5.53MIN: 1.77 / MAX: 5.48MIN: 1.76 / MAX: 5.58MIN: 1.84 / MAX: 5.61
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: LuxCore Benchmark - Acceleration: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810Min: 4.72 / Avg: 4.73 / Max: 4.74Min: 4.67 / Avg: 4.68 / Max: 4.7Min: 4.71 / Avg: 4.74 / Max: 4.78Min: 4.76 / Avg: 4.79 / Max: 4.82

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V4Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye200K400K600K800K1000KSE +/- 2093.15, N = 3SE +/- 6145.17, N = 3SE +/- 4599.44, N = 3SE +/- 1376.88, N = 3936694941892942408932251
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V4Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye160K320K480K640K800KMin: 933660 / Avg: 936694 / Max: 940709Min: 935667 / Avg: 941892 / Max: 954182Min: 933269 / Avg: 942408 / Max: 947885Min: 930015 / Avg: 932251 / Max: 934761

LuxCoreRender

LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: DLSC - Acceleration: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye1.23082.46163.69244.92326.154SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 35.455.465.475.45MIN: 5.34 / MAX: 5.8MIN: 5.31 / MAX: 5.8MIN: 5.25 / MAX: 5.85MIN: 5.34 / MAX: 5.91
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: DLSC - Acceleration: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810Min: 5.44 / Avg: 5.45 / Max: 5.47Min: 5.41 / Avg: 5.46 / Max: 5.54Min: 5.43 / Avg: 5.47 / Max: 5.49Min: 5.43 / Avg: 5.45 / Max: 5.48

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V2Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye200K400K600K800K1000KSE +/- 1339.58, N = 3SE +/- 1783.60, N = 3SE +/- 156.42, N = 3SE +/- 1598.54, N = 3835539837065834724834703
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V2Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye150K300K450K600K750KMin: 832863 / Avg: 835538.67 / Max: 836995Min: 834000 / Avg: 837064.67 / Max: 840178Min: 834481 / Avg: 834723.67 / Max: 835016Min: 831574 / Avg: 834703.33 / Max: 836835

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye15K30K45K60K75KSE +/- 188.81, N = 3SE +/- 521.49, N = 3SE +/- 372.87, N = 3SE +/- 292.33, N = 369192.670453.969846.269107.3
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye12K24K36K48K60KMin: 68919.2 / Avg: 69192.63 / Max: 69554.9Min: 69431.8 / Avg: 70453.93 / Max: 71144.7Min: 69390.7 / Avg: 69846.2 / Max: 70585.3Min: 68718.2 / Avg: 69107.33 / Max: 69679.8

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye10K20K30K40K50KSE +/- 88.51, N = 3SE +/- 170.17, N = 3SE +/- 94.70, N = 3SE +/- 25.57, N = 346472.946557.746650.146349.8
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye8K16K24K32K40KMin: 46368.7 / Avg: 46472.87 / Max: 46648.9Min: 46368.5 / Avg: 46557.7 / Max: 46897.3Min: 46531.2 / Avg: 46650.07 / Max: 46837.2Min: 46324 / Avg: 46349.77 / Max: 46400.9

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet FloatUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye9K18K27K36K45KSE +/- 161.47, N = 3SE +/- 107.46, N = 3SE +/- 96.21, N = 3SE +/- 101.03, N = 343688.443885.744222.943711.5
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet FloatUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye8K16K24K32K40KMin: 43422 / Avg: 43688.4 / Max: 43979.7Min: 43675.1 / Avg: 43885.73 / Max: 44028Min: 44080.8 / Avg: 44222.87 / Max: 44406.3Min: 43607.2 / Avg: 43711.47 / Max: 43913.5

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2021.1Model: Age Gender Recognition Retail 0013 FP16 - Device: CPUUbuntu 20.04.2 LTS0.1710.3420.5130.6840.855SE +/- 0.00, N = 30.76

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2021.1Model: Age Gender Recognition Retail 0013 FP16 - Device: CPUUbuntu 20.04.2 LTS4K8K12K16K20KSE +/- 155.65, N = 320273.78

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2021.1Model: Age Gender Recognition Retail 0013 FP32 - Device: CPUUbuntu 20.04.2 LTS0.1710.3420.5130.6840.855SE +/- 0.00, N = 30.76

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2021.1Model: Age Gender Recognition Retail 0013 FP32 - Device: CPUUbuntu 20.04.2 LTS4K8K12K16K20KSE +/- 8.07, N = 320458.87

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.10.20Time To CompileUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye816243240SE +/- 0.36, N = 5SE +/- 0.40, N = 4SE +/- 0.22, N = 10SE +/- 0.46, N = 335.1434.0229.4932.03
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.10.20Time To CompileUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye816243240Min: 34.73 / Avg: 35.14 / Max: 36.57Min: 33.6 / Avg: 34.02 / Max: 35.23Min: 28.65 / Avg: 29.49 / Max: 31.16Min: 31.51 / Avg: 32.03 / Max: 32.95

Timed Godot Game Engine Compilation

This test times how long it takes to compile the Godot Game Engine. Godot is a popular, open-source, cross-platform 2D/3D game engine and is built using the SCons build system and targeting the X11 platform. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 3.2.3Time To CompileUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye1428425670SE +/- 0.28, N = 3SE +/- 0.09, N = 3SE +/- 0.36, N = 3SE +/- 0.17, N = 357.9154.9563.2953.43
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 3.2.3Time To CompileUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye1224364860Min: 57.36 / Avg: 57.91 / Max: 58.27Min: 54.83 / Avg: 54.95 / Max: 55.13Min: 62.71 / Avg: 63.29 / Max: 63.96Min: 53.1 / Avg: 53.43 / Max: 53.68

GROMACS

The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing with the water_GMX50 data. This test profile allows selecting between CPU and GPU-based GROMACS builds. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2021.2Implementation: MPI CPU - Input: water_GMX50_bareUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye0.84511.69022.53533.38044.2255SE +/- 0.010, N = 3SE +/- 0.007, N = 3SE +/- 0.005, N = 33.6393.6403.7561. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2021.2Implementation: MPI CPU - Input: water_GMX50_bareUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye246810Min: 3.62 / Avg: 3.64 / Max: 3.66Min: 3.63 / Avg: 3.64 / Max: 3.66Min: 3.75 / Avg: 3.76 / Max: 3.771. (CXX) g++ options: -O3 -pthread

Timed Wasmer Compilation

This test times how long it takes to compile Wasmer. Wasmer is written in the Rust programming language and is a WebAssembly runtime implementation that supports WASI and EmScripten. This test profile builds Wasmer with the Cranelift and Singlepast compiler features enabled. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Wasmer Compilation 1.0.2Time To CompileUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye1428425670SE +/- 0.35, N = 3SE +/- 0.19, N = 3SE +/- 0.57, N = 361.2441.5060.641. (CC) gcc options: -m64 -pie -nodefaultlibs -ldl -lrt -lpthread -lgcc_s -lc -lm -lutil
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Wasmer Compilation 1.0.2Time To CompileUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye1224364860Min: 60.62 / Avg: 61.24 / Max: 61.82Min: 41.21 / Avg: 41.5 / Max: 41.86Min: 59.62 / Avg: 60.64 / Max: 61.611. (CC) gcc options: -m64 -pie -nodefaultlibs -ldl -lrt -lpthread -lgcc_s -lc -lm -lutil

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL, NVIDIA OptiX, and NVIDIA CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: CPU-OnlyUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye1224364860SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 352.8553.1352.4452.55
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: CPU-OnlyUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye1122334455Min: 52.78 / Avg: 52.85 / Max: 52.97Min: 52.98 / Avg: 53.13 / Max: 53.3Min: 52.35 / Avg: 52.44 / Max: 52.53Min: 52.48 / Avg: 52.55 / Max: 52.65

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: Path TracerUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye1.01252.0253.03754.055.0625SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 34.474.464.484.50MIN: 4.24 / MAX: 4.5MIN: 4.05 / MAX: 4.5MIN: 4.37 / MAX: 4.5MIN: 3.82 / MAX: 4.52
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: Path TracerUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810Min: 4.46 / Avg: 4.47 / Max: 4.48Min: 4.46 / Avg: 4.46 / Max: 4.46Min: 4.48 / Avg: 4.48 / Max: 4.48Min: 4.48 / Avg: 4.5 / Max: 4.5

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 5 - Input: Bosphorus 4KUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye48121620SE +/- 0.07, N = 3SE +/- 0.13, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 313.6316.2313.039.311. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.10.0Speed: Speed 5 - Input: Bosphorus 4KUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye48121620Min: 13.5 / Avg: 13.63 / Max: 13.73Min: 16.06 / Avg: 16.23 / Max: 16.49Min: 12.93 / Avg: 13.03 / Max: 13.1Min: 9.25 / Avg: 9.31 / Max: 9.41. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

NAMD

NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye0.15150.3030.45450.6060.7575SE +/- 0.00041, N = 3SE +/- 0.00051, N = 3SE +/- 0.00019, N = 3SE +/- 0.00045, N = 30.672520.673550.671930.67131
OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810Min: 0.67 / Avg: 0.67 / Max: 0.67Min: 0.67 / Avg: 0.67 / Max: 0.67Min: 0.67 / Avg: 0.67 / Max: 0.67Min: 0.67 / Avg: 0.67 / Max: 0.67

Xmrig

Xmrig is an open-source cross-platform CPU/GPU miner for RandomX, KawPow, CryptoNight and AstroBWT. This test profile is setup to measure the Xmlrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.12.1Variant: Monero - Hash Count: 1MUbuntu 20.04.2 LTSDebian 10.10Debian Bullseye5K10K15K20K25KSE +/- 33.78, N = 3SE +/- 52.41, N = 3SE +/- 32.24, N = 321914.922334.523576.31. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
OpenBenchmarking.orgH/s, More Is BetterXmrig 6.12.1Variant: Monero - Hash Count: 1MUbuntu 20.04.2 LTSDebian 10.10Debian Bullseye4K8K12K16K20KMin: 21879.4 / Avg: 21914.87 / Max: 21982.4Min: 22235.1 / Avg: 22334.5 / Max: 22413Min: 23513.4 / Avg: 23576.3 / Max: 236201. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye20K40K60K80K100KSE +/- 609.71, N = 3SE +/- 42.83, N = 3SE +/- 824.08, N = 3102552.8421406.45105456.95-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian Bullseye: Open MPI 4.1.0
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye20K40K60K80K100KMin: 101833.42 / Avg: 102552.84 / Max: 103765.24Min: 21326 / Avg: 21406.45 / Max: 21472.15Min: 103841.68 / Avg: 105456.95 / Max: 106548.291. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian Bullseye: Open MPI 4.1.0

LuxCoreRender

LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Rainbow Colors and Prism - Acceleration: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye48121620SE +/- 0.19, N = 15SE +/- 0.10, N = 3SE +/- 0.30, N = 15SE +/- 0.26, N = 1516.3816.1117.1416.75MIN: 14.52 / MAX: 18.74MIN: 14.68 / MAX: 16.32MIN: 14.81 / MAX: 19.08MIN: 14.5 / MAX: 18.42
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Rainbow Colors and Prism - Acceleration: CPUUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye48121620Min: 15.85 / Avg: 16.38 / Max: 18.46Min: 15.91 / Avg: 16.11 / Max: 16.24Min: 16.01 / Avg: 17.14 / Max: 18.8Min: 15.69 / Avg: 16.75 / Max: 18.35

Xmrig

Xmrig is an open-source cross-platform CPU/GPU miner for RandomX, KawPow, CryptoNight and AstroBWT. This test profile is setup to measure the Xmlrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.12.1Variant: Wownero - Hash Count: 1MUbuntu 20.04.2 LTSDebian 10.10Debian Bullseye5K10K15K20K25KSE +/- 36.32, N = 3SE +/- 79.23, N = 3SE +/- 30.16, N = 323123.123003.324478.31. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
OpenBenchmarking.orgH/s, More Is BetterXmrig 6.12.1Variant: Wownero - Hash Count: 1MUbuntu 20.04.2 LTSDebian 10.10Debian Bullseye4K8K12K16K20KMin: 23075 / Avg: 23123.1 / Max: 23194.3Min: 22869.7 / Avg: 23003.33 / Max: 23143.9Min: 24436.7 / Avg: 24478.27 / Max: 24536.91. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

Intel Open Image Denoise

Open Image Denoise is a denoising library for ray-tracing and part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RT.ldr_alb_nrm.3840x2160Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye0.2160.4320.6480.8641.08SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 60.900.960.870.94
OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RT.ldr_alb_nrm.3840x2160Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810Min: 0.9 / Avg: 0.9 / Max: 0.91Min: 0.96 / Avg: 0.96 / Max: 0.96Min: 0.87 / Avg: 0.87 / Max: 0.88Min: 0.89 / Avg: 0.94 / Max: 0.96

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian DragonUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye918273645SE +/- 0.26, N = 3SE +/- 0.15, N = 3SE +/- 0.64, N = 15SE +/- 0.22, N = 337.1937.3832.4437.32MIN: 36.46 / MAX: 37.99MIN: 36.85 / MAX: 38.17MIN: 29.11 / MAX: 37.15MIN: 36.86 / MAX: 38.33
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian DragonUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye816243240Min: 36.68 / Avg: 37.19 / Max: 37.48Min: 37.09 / Avg: 37.38 / Max: 37.55Min: 29.36 / Avg: 32.44 / Max: 36.54Min: 37.09 / Avg: 37.32 / Max: 37.76

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 8 - Input: Bosphorus 4KUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye510152025SE +/- 0.12, N = 3SE +/- 0.11, N = 3SE +/- 0.11, N = 3SE +/- 0.12, N = 320.8518.3520.7214.881. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 8 - Input: Bosphorus 4KUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye510152025Min: 20.66 / Avg: 20.85 / Max: 21.08Min: 18.22 / Avg: 18.35 / Max: 18.56Min: 20.57 / Avg: 20.72 / Max: 20.93Min: 14.65 / Avg: 14.88 / Max: 15.081. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

Intel Open Image Denoise

Open Image Denoise is a denoising library for ray-tracing and part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RT.hdr_alb_nrm.3840x2160Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye0.2160.4320.6480.8641.08SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 30.900.960.870.89
OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.4.0Run: RT.hdr_alb_nrm.3840x2160Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810Min: 0.9 / Avg: 0.9 / Max: 0.91Min: 0.95 / Avg: 0.96 / Max: 0.96Min: 0.86 / Avg: 0.87 / Max: 0.87Min: 0.86 / Avg: 0.89 / Max: 0.9

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.DUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye400800120016002000SE +/- 8.29, N = 3SE +/- 11.91, N = 4SE +/- 10.78, N = 31786.821092.181731.34-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian Bullseye: Open MPI 4.1.0
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.DUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye30060090012001500Min: 1770.26 / Avg: 1786.82 / Max: 1795.89Min: 1060.76 / Avg: 1092.18 / Max: 1117.81Min: 1714.2 / Avg: 1731.34 / Max: 1751.231. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian Bullseye: Open MPI 4.1.0

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial timeseries data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Bayesian ChangepointUbuntu 20.04.2 LTSDebian 10.10816243240SE +/- 0.15, N = 3SE +/- 0.29, N = 326.8435.81
OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Bayesian ChangepointUbuntu 20.04.2 LTSDebian 10.10816243240Min: 26.54 / Avg: 26.84 / Max: 27.03Min: 35.25 / Avg: 35.81 / Max: 36.16

OpenFOAM

OpenFOAM is the leading free, open source software for computational fluid dynamics (CFD). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 30MUbuntu 20.04.2 LTSDebian Bullseye510152025SE +/- 0.05, N = 3SE +/- 0.07, N = 322.2421.771. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 30MUbuntu 20.04.2 LTSDebian Bullseye510152025Min: 22.14 / Avg: 22.24 / Max: 22.3Min: 21.68 / Avg: 21.77 / Max: 21.911. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6, LosslessUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye714212835SE +/- 0.20, N = 3SE +/- 0.02, N = 3SE +/- 0.31, N = 3SE +/- 0.09, N = 330.2227.8827.2427.401. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6, LosslessUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye714212835Min: 29.86 / Avg: 30.22 / Max: 30.55Min: 27.84 / Avg: 27.88 / Max: 27.92Min: 26.84 / Avg: 27.24 / Max: 27.85Min: 27.23 / Avg: 27.4 / Max: 27.551. (CXX) g++ options: -O3 -fPIC -lm

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.0Preset: ExhaustiveUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye612182430SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 323.6823.5223.461. (CXX) g++ options: -O3 -flto -pthread
OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.0Preset: ExhaustiveUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye612182430Min: 23.66 / Avg: 23.68 / Max: 23.69Min: 23.51 / Avg: 23.52 / Max: 23.52Min: 23.44 / Avg: 23.46 / Max: 23.471. (CXX) g++ options: -O3 -flto -pthread

SVT-HEVC

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 1 - Input: Bosphorus 1080pUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye612182430SE +/- 0.09, N = 3SE +/- 0.10, N = 3SE +/- 0.04, N = 3SE +/- 0.08, N = 324.6524.7125.1924.961. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 1 - Input: Bosphorus 1080pUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye612182430Min: 24.48 / Avg: 24.65 / Max: 24.75Min: 24.53 / Avg: 24.71 / Max: 24.87Min: 25.11 / Avg: 25.19 / Max: 25.23Min: 24.79 / Avg: 24.96 / Max: 25.061. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: MobileNet v2Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye70140210280350SE +/- 0.71, N = 3SE +/- 2.56, N = 3SE +/- 2.75, N = 3SE +/- 3.50, N = 4290.34296.55286.25307.53MIN: 286.82 / MAX: 344.05MIN: 285.14 / MAX: 355.51MIN: 280.64 / MAX: 370.21MIN: 283.29 / MAX: 381.191. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: MobileNet v2Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye60120180240300Min: 289.51 / Avg: 290.34 / Max: 291.76Min: 291.45 / Avg: 296.55 / Max: 299.49Min: 283.4 / Avg: 286.25 / Max: 291.76Min: 299.89 / Avg: 307.53 / Max: 316.631. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

Timed Mesa Compilation

This test profile times how long it takes to compile Mesa with Meson/Ninja. For minimizing build dependencies and avoid versioning conflicts, test this is just the core Mesa build without LLVM or the extra Gallium3D/Mesa drivers enabled. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Mesa Compilation 21.0Time To CompileUbuntu 20.04.2 LTSDebian Bullseye510152025SE +/- 0.32, N = 3SE +/- 0.10, N = 322.7520.27
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Mesa Compilation 21.0Time To CompileUbuntu 20.04.2 LTSDebian Bullseye510152025Min: 22.27 / Avg: 22.75 / Max: 23.35Min: 20.09 / Avg: 20.27 / Max: 20.45

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: CrownUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye816243240SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.12, N = 3SE +/- 0.07, N = 333.3233.3932.5933.52MIN: 32.96 / MAX: 33.96MIN: 32.92 / MAX: 34.19MIN: 32.08 / MAX: 33.4MIN: 33.02 / MAX: 34.41
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: CrownUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye714212835Min: 33.31 / Avg: 33.32 / Max: 33.34Min: 33.31 / Avg: 33.39 / Max: 33.48Min: 32.43 / Avg: 32.59 / Max: 32.82Min: 33.38 / Avg: 33.52 / Max: 33.61

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: Path TracerUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye3691215SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 312.3512.3512.5012.66MIN: 9.26 / MAX: 12.5MIN: 9.26 / MAX: 12.5MIN: 9.8 / MAX: 12.66MIN: 9.8 / MAX: 12.82
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: Path TracerUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye48121620Min: 12.35 / Avg: 12.35 / Max: 12.35Min: 12.35 / Avg: 12.35 / Max: 12.35Min: 12.5 / Avg: 12.5 / Max: 12.5Min: 12.66 / Avg: 12.66 / Max: 12.66

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v1.1Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye60120180240300SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.36, N = 3SE +/- 0.57, N = 3269.18271.39268.00269.36MIN: 268.98 / MAX: 269.45MIN: 271.08 / MAX: 271.69MIN: 267.47 / MAX: 295.77MIN: 268.62 / MAX: 270.751. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v1.1Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye50100150200250Min: 269.08 / Avg: 269.18 / Max: 269.34Min: 271.22 / Avg: 271.39 / Max: 271.51Min: 267.59 / Avg: 268 / Max: 268.71Min: 268.73 / Avg: 269.36 / Max: 270.51. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.CUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye11K22K33K44K55KSE +/- 83.51, N = 3SE +/- 467.47, N = 15SE +/- 60.07, N = 349265.3623154.5450128.60-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian Bullseye: Open MPI 4.1.0
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.CUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye9K18K27K36K45KMin: 49105.03 / Avg: 49265.36 / Max: 49386.05Min: 17561.91 / Avg: 23154.54 / Max: 24365.05Min: 50012.22 / Avg: 50128.6 / Max: 50212.591. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian Bullseye: Open MPI 4.1.0

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: FT.CUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye11K22K33K44K55KSE +/- 40.78, N = 3SE +/- 43.58, N = 3SE +/- 134.89, N = 347059.6817996.0949076.24-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian Bullseye: Open MPI 4.1.0
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: FT.CUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye9K18K27K36K45KMin: 46997.65 / Avg: 47059.68 / Max: 47136.55Min: 17925.22 / Avg: 17996.09 / Max: 18075.48Min: 48806.56 / Avg: 49076.24 / Max: 49217.541. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian Bullseye: Open MPI 4.1.0

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial timeseries data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Relative EntropyUbuntu 20.04.2 LTSDebian 10.103691215SE +/- 0.14, N = 3SE +/- 0.03, N = 313.0512.32
OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Relative EntropyUbuntu 20.04.2 LTSDebian 10.1048121620Min: 12.83 / Avg: 13.05 / Max: 13.3Min: 12.25 / Avg: 12.32 / Max: 12.36

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10, LosslessUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810SE +/- 0.060, N = 6SE +/- 0.006, N = 3SE +/- 0.079, N = 15SE +/- 0.015, N = 36.2876.1855.3235.2981. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10, LosslessUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye3691215Min: 6.01 / Avg: 6.29 / Max: 6.43Min: 6.18 / Avg: 6.18 / Max: 6.2Min: 5.09 / Avg: 5.32 / Max: 5.95Min: 5.27 / Avg: 5.3 / Max: 5.321. (CXX) g++ options: -O3 -fPIC -lm

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye0.85881.71762.57643.43524.294SE +/- 0.018, N = 3SE +/- 0.030, N = 9SE +/- 0.037, N = 3SE +/- 0.021, N = 153.8173.6202.7852.8731. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye246810Min: 3.78 / Avg: 3.82 / Max: 3.84Min: 3.51 / Avg: 3.62 / Max: 3.73Min: 2.74 / Avg: 2.79 / Max: 2.86Min: 2.76 / Avg: 2.87 / Max: 2.981. (CXX) g++ options: -O3 -fPIC -lm

SVT-HEVC

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080pUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye120240360480600SE +/- 12.16, N = 15SE +/- 6.12, N = 15SE +/- 6.17, N = 4SE +/- 8.24, N = 15458.18434.07537.01429.711. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080pUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye90180270360450Min: 396.04 / Avg: 458.18 / Max: 521.29Min: 407.61 / Avg: 434.07 / Max: 494.64Min: 522.65 / Avg: 537.01 / Max: 550.96Min: 383.88 / Avg: 429.71 / Max: 489.81. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye7K14K21K28K35KSE +/- 271.73, N = 3SE +/- 10.91, N = 3SE +/- 246.35, N = 329891.7113587.6130352.21-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian Bullseye: Open MPI 4.1.0
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CUbuntu 20.04.2 LTSFedora Server 34Debian Bullseye5K10K15K20K25KMin: 29384.41 / Avg: 29891.71 / Max: 30314.17Min: 13566.81 / Avg: 13587.61 / Max: 13603.74Min: 29881.94 / Avg: 30352.21 / Max: 30714.591. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 20.04.2 LTS: Open MPI 4.0.33. Debian Bullseye: Open MPI 4.1.0

SVT-VP9

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample YUV input video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 1080pUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye80160240320400SE +/- 4.22, N = 12SE +/- 1.79, N = 3SE +/- 2.58, N = 15SE +/- 2.14, N = 3349.16291.76344.93256.351. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 1080pUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye60120180240300Min: 303.39 / Avg: 349.16 / Max: 358.61Min: 288.3 / Avg: 291.76 / Max: 294.26Min: 311.18 / Avg: 344.93 / Max: 353.22Min: 252.17 / Avg: 256.35 / Max: 259.251. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial timeseries data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Windowed GaussianUbuntu 20.04.2 LTSDebian 10.10246810SE +/- 0.070, N = 3SE +/- 0.035, N = 36.8056.671
OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Windowed GaussianUbuntu 20.04.2 LTSDebian 10.103691215Min: 6.67 / Avg: 6.81 / Max: 6.91Min: 6.63 / Avg: 6.67 / Max: 6.74

SVT-VP9

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample YUV input video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye80160240320400SE +/- 2.18, N = 3SE +/- 2.17, N = 15SE +/- 1.49, N = 3SE +/- 0.27, N = 3353.98305.14353.73259.561. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye60120180240300Min: 350 / Avg: 353.98 / Max: 357.53Min: 296.81 / Avg: 305.14 / Max: 329.62Min: 352.03 / Avg: 353.73 / Max: 356.69Min: 259.2 / Avg: 259.56 / Max: 260.091. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v2Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye1428425670SE +/- 0.31, N = 3SE +/- 0.06, N = 3SE +/- 0.67, N = 5SE +/- 0.47, N = 362.6064.8162.2262.17MIN: 61.94 / MAX: 63.37MIN: 64.62 / MAX: 65.08MIN: 60.94 / MAX: 65.02MIN: 61.16 / MAX: 62.911. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl
OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v2Ubuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye1326395265Min: 62.15 / Avg: 62.6 / Max: 63.19Min: 64.7 / Avg: 64.81 / Max: 64.88Min: 61.13 / Avg: 62.22 / Max: 64.81Min: 61.24 / Avg: 62.17 / Max: 62.81. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: Path TracerUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye110220330440550SE +/- 21.00, N = 15SE +/- 22.22, N = 15500.00500.00444.44422.22MIN: 250MIN: 250MIN: 250 / MAX: 500MIN: 200 / MAX: 500
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: Path TracerUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye90180270360450Min: 333.33 / Avg: 444.44 / Max: 500Min: 333.33 / Avg: 422.22 / Max: 500

SVT-VP9

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample YUV input video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: Visual Quality Optimized - Input: Bosphorus 1080pUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye60120180240300SE +/- 2.34, N = 3SE +/- 2.18, N = 3SE +/- 2.94, N = 3SE +/- 2.63, N = 5264.89272.55297.41236.321. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: Visual Quality Optimized - Input: Bosphorus 1080pUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye50100150200250Min: 260.29 / Avg: 264.89 / Max: 267.93Min: 268.31 / Avg: 272.55 / Max: 275.55Min: 292.75 / Avg: 297.41 / Max: 302.84Min: 229.27 / Avg: 236.32 / Max: 241.761. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

SVT-HEVC

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080pUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye60120180240300SE +/- 3.95, N = 3SE +/- 2.22, N = 3SE +/- 3.39, N = 3SE +/- 3.08, N = 4282.38281.37296.57273.701. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080pUbuntu 20.04.2 LTSFedora Server 34Debian 10.10Debian Bullseye50100150200250Min: 275.48 / Avg: 282.38 / Max: 289.16Min: 278.04 / Avg: 281.37 / Max: 285.58Min: 289.86 / Avg: 296.57 / Max: 300.75Min: 270.03 / Avg: 273.7 / Max: 282.891. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

102 Results Shown

Numenta Anomaly Benchmark
NWChem
High Performance Conjugate Gradient
OpenVKL
ONNX Runtime:
  bertsquad-10 - OpenMP CPU
  fcn-resnet101-11 - OpenMP CPU
  yolov4 - OpenMP CPU
Mobile Neural Network:
  inception-v3
  mobilenet-v1-1.0
  MobileNetV2_224
  SqueezeNetV1.0
  resnet-v2-50
  squeezenetv1.1
  mobilenetV3
OpenFOAM
Timed LLVM Compilation
TensorFlow Lite
PlaidML
BRL-CAD
NAS Parallel Benchmarks
ONNX Runtime
TNN
Timed LLVM Compilation
Kripke
Blender:
  Barbershop - CPU-Only
  Classroom - CPU-Only
Timed Node.js Compilation
NAS Parallel Benchmarks
miniFE
ONNX Runtime
GPAW
PlaidML
VP9 libvpx Encoding
SVT-AV1
PlaidML
OSPray
Xcompact3d Incompact3d
Numenta Anomaly Benchmark
OpenVINO:
  Person Detection 0106 FP16 - CPU:
    ms
    FPS
  Person Detection 0106 FP32 - CPU:
    ms
    FPS
NAS Parallel Benchmarks
OpenVINO:
  Face Detection 0106 FP16 - CPU:
    ms
    FPS
  Face Detection 0106 FP32 - CPU:
    ms
    FPS
Intel Open Image Denoise
LuxCoreRender:
  Orange Juice - CPU
  Danish Mood - CPU
  LuxCore Benchmark - CPU
TensorFlow Lite
LuxCoreRender
TensorFlow Lite:
  Inception ResNet V2
  SqueezeNet
  Mobilenet Quant
  Mobilenet Float
OpenVINO:
  Age Gender Recognition Retail 0013 FP16 - CPU:
    ms
    FPS
  Age Gender Recognition Retail 0013 FP32 - CPU:
    ms
    FPS
Timed Linux Kernel Compilation
Timed Godot Game Engine Compilation
GROMACS
Timed Wasmer Compilation
Blender
OSPray
VP9 libvpx Encoding
NAMD
Xmrig
NAS Parallel Benchmarks
LuxCoreRender
Xmrig
Intel Open Image Denoise
Embree
SVT-AV1
Intel Open Image Denoise
NAS Parallel Benchmarks
Numenta Anomaly Benchmark
OpenFOAM
libavif avifenc
ASTC Encoder
SVT-HEVC
TNN
Timed Mesa Compilation
Embree
OSPray
TNN
NAS Parallel Benchmarks:
  MG.C
  FT.C
Numenta Anomaly Benchmark
libavif avifenc:
  10, Lossless
  10
SVT-HEVC
NAS Parallel Benchmarks
SVT-VP9
Numenta Anomaly Benchmark
SVT-VP9
TNN
OSPray
SVT-VP9
SVT-HEVC