AMD Ryzen 9 7900 DDR5-5200 ECC Memory

AMD Ryzen 9 7900 DDR5-5200 ECC memory benchmarks by Michael Larabel for a future article

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2405045-NE-AMDRYZEN948
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
Show Result Confidence Charts
Allow Limiting Results To Certain Suite(s)

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
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
Toggle/Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
2 x 32GB DDR5-4800
April 23
  3 Hours, 18 Minutes
2 x 32GB DDR5-5200
May 04
  3 Hours, 16 Minutes
Invert Behavior (Only Show Selected Data)
  3 Hours, 17 Minutes
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):


AMD Ryzen 9 7900 DDR5-5200 ECC MemoryOpenBenchmarking.orgPhoronix Test SuiteAMD Ryzen 9 7900 12-Core @ 5.48GHz (12 Cores / 24 Threads)ASRockRack 1U4LW-B650/2L2T B650D4U-2L2T/BCM (2.09 BIOS)AMD Device 14d82 x 32GB DDR5-4800MT/s Micron MTC20C2085S1EC48BA12 x 32GB DDR5-5200MT/s Kingston1024GB SOLIDIGM SSDPFKNU010TZASPEEDAMD Rembrandt Radeon HD AudioVA24312 x Intel I210 + 2 x Broadcom BCM57416 NetXtreme-E Dual-Media 10G RDMAUbuntu 24.046.8.0-31-generic (x86_64)GNOME Shell 46.0X ServerGCC 13.2.0ext41920x1200ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerCompilerFile-SystemScreen ResolutionAMD Ryzen 9 7900 DDR5-5200 ECC Memory BenchmarksSystem Logs- Transparent Huge Pages: madvise- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --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 - Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa601203 - Python 3.12.3- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Vulnerable: Safe RET no microcode + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected

2 x 32GB DDR5-4800 vs. 2 x 32GB DDR5-5200 ComparisonPhoronix Test SuiteBaseline+2.8%+2.8%+5.6%+5.6%+8.4%+8.4%11%8.5%7.5%7.2%6.4%6.3%5.3%4.3%4.2%4.2%4.1%4%4%3.9%3.9%3.9%3.9%3.7%3.7%3.7%3.7%3.7%3.6%3.6%3.6%3.1%3.1%3%3%3%2.8%2.6%2.5%2.4%2.3%2.2%2.1%2.1%2.1%2.1%2%2%M.C.F.B.S - 4096 MiBM.C.F.B.S - 8192 MiBllama-2-7b.Q4_0.ggufd.S.M.S - Mesh TimeM.C.F.B.S - 1024 MiBCG.CSP.B32FT.Ci.i.1.C.P.DIS.DAdd - IntegerIP Shapes 3D - CPUMemory Copy - 8192 MiBBT.CSP.CC.B.S.A - CPUd.S.M.S - Execution TimeChrysler Neon 1Mi.i.1.C.P.DScale - Integerllama-2-13b.Q4_0.ggufCPU - 1 - AlexNetllava-v1.5-7b-q4 - CPUCopy - IntegerAverage - Integermistral-7b-instruct-v0.2.Q8_0 - CPUMG.Cwizardcoder-python-34b-v1.0.Q6_K - CPUTomographic ModelWAV To WavPackTriad - Integerllama-2-70b-chat.Q5_0.gguf6Memory Copy - 1024 MiBW.l.H2.M.L.T.SR.S.A.F - CPU1R.S.A.F - CPUV.D.F - CPUV.D.F - CPUMBWMBWLlama.cppOpenFOAMMBWNAS Parallel BenchmarksNAS Parallel BenchmarkslibxsmmNAS Parallel BenchmarksXcompact3d Incompact3dNAS Parallel BenchmarksRAMspeed SMPoneDNNMBWNAS Parallel BenchmarksNAS Parallel BenchmarksoneDNNOpenFOAMOpenRadiossXcompact3d Incompact3dRAMspeed SMPLlama.cppTensorFlowLlamafileRAMspeed SMPRAMspeed SMPLlamafileNAS Parallel BenchmarksLlamafileSPECFEM3DWavPack Audio EncodingRAMspeed SMPLlama.cpplibavif avifencMBWSPECFEM3DRNNoiseOpenVINOJPEG-XL Decoding libjxlOpenVINOOpenVINOOpenVINO2 x 32GB DDR5-48002 x 32GB DDR5-5200

AMD Ryzen 9 7900 DDR5-5200 ECC Memorypytorch: CPU - 1 - ResNet-50quicksilver: CORAL2 P1quicksilver: CORAL2 P2ffmpeg: libx265 - Liveffmpeg: libx265 - Uploadffmpeg: libx265 - Platformffmpeg: libx265 - Video On Demandopenvino: Person Detection FP16 - CPUopenvino: Person Detection FP32 - CPUopenvino: Vehicle Detection FP16 - CPUopenvino: Face Detection FP16-INT8 - CPUopenvino: Road Segmentation ADAS FP16 - CPUopenvino: Road Segmentation ADAS FP16-INT8 - CPUopenvino: Machine Translation EN To DE FP16 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUembree: Pathtracer ISPC - Asian Dragonembree: Pathtracer ISPC - Asian Dragon Objsvt-av1: Preset 12 - Bosphorus 4Ksvt-av1: Preset 13 - Bosphorus 4Klibxsmm: 32tensorflow: CPU - 1 - AlexNettensorflow: CPU - 64 - AlexNettensorflow: CPU - 1 - GoogLeNettensorflow: CPU - 1 - ResNet-50tensorflow: CPU - 64 - GoogLeNettensorflow: CPU - 64 - ResNet-50ospray: gravity_spheres_volume/dim_512/ao/real_timegraphics-magick: Swirlgraphics-magick: Rotategraphics-magick: Resizinggraphics-magick: Noise-Gaussiangraphics-magick: HWB Color Spaceramspeed: Add - Integerramspeed: Copy - Integerramspeed: Scale - Integerramspeed: Triad - Integerramspeed: Average - Integersrsran: PDSCH Processor Benchmark, Throughput Totalsrsran: PUSCH Processor Benchmark, Throughput Totalmbw: Memory Copy - 1024 MiBmbw: Memory Copy - 4096 MiBmbw: Memory Copy - 8192 MiBmbw: Memory Copy, Fixed Block Size - 1024 MiBmbw: Memory Copy, Fixed Block Size - 4096 MiBmbw: Memory Copy, Fixed Block Size - 8192 MiBcompress-7zip: Compression Ratingcompress-7zip: Decompression Ratingjpegxl: PNG - 90jpegxl-decode: 1jpegxl-decode: Allstockfish: Chess Benchmarkgromacs: MPI CPU - water_GMX50_barenamd: ATPase with 327,506 Atomsnamd: STMV with 1,066,628 Atomsrocksdb: Read While Writingnginx: 1000llama-cpp: llama-2-7b.Q4_0.ggufllama-cpp: llama-2-13b.Q4_0.ggufllama-cpp: llama-2-70b-chat.Q5_0.ggufllamafile: llava-v1.5-7b-q4 - CPUllamafile: mistral-7b-instruct-v0.2.Q8_0 - CPUllamafile: wizardcoder-python-34b-v1.0.Q6_K - CPUnpb: BT.Cnpb: CG.Cnpb: FT.Cnpb: IS.Dnpb: LU.Cnpb: MG.Cnpb: SP.Bnpb: SP.Cpgbench: 100 - 1000 - Read Onlypgbench: 100 - 1000 - Read Writebrl-cad: VGR Performance Metricv-ray: CPUonednn: IP Shapes 3D - CPUonednn: Convolution Batch Shapes Auto - CPUonednn: Deconvolution Batch shapes_3d - CPUpgbench: 100 - 1000 - Read Only - Average Latencypgbench: 100 - 1000 - Read Write - Average Latencydraco: Liondraco: Church Facadeopenvino: Person Detection FP16 - CPUopenvino: Person Detection FP32 - CPUopenvino: Vehicle Detection FP16 - CPUopenvino: Face Detection FP16-INT8 - CPUopenvino: Road Segmentation ADAS FP16 - CPUopenvino: Road Segmentation ADAS FP16-INT8 - CPUopenvino: Machine Translation EN To DE FP16 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUincompact3d: input.i3d 129 Cells Per Directionincompact3d: input.i3d 193 Cells Per Directionopenfoam: drivaerFastback, Small Mesh Size - Mesh Timeopenfoam: drivaerFastback, Small Mesh Size - Execution Timeopenradioss: Bumper Beamopenradioss: Chrysler Neon 1Mopenradioss: Cell Phone Drop Testopenradioss: Bird Strike on Windshieldopenradioss: Rubber O-Ring Seal Installationopenradioss: INIVOL and Fluid Structure Interaction Drop Containerspecfem3d: Mount St. Helensspecfem3d: Layered Halfspacespecfem3d: Tomographic Modelspecfem3d: Homogeneous Halfspacespecfem3d: Water-layered Halfspaceavifenc: 2avifenc: 6avifenc: 10, Losslessbuild-godot: Time To Compilebuild-linux-kernel: defconfigbuild-linux-kernel: allmodconfigbuild-mesa: Time To Compilebuild-nodejs: Time To Compilebuild-php: Time To Compilernnoise: 26 Minute Long Talking Sampleblender: Junkshop - CPU-Onlyblender: Classroom - CPU-Onlyblender: Fishy Cat - CPU-Onlyblender: Barbershop - CPU-Onlyblender: Pabellon Barcelona - CPU-Onlyencode-wavpack: WAV To WavPack2 x 32GB DDR5-48002 x 32GB DDR5-520058.751855000022580000175.3033.6668.3068.2774.474.91645.6318.57281.64345.899.051090.8525.265821.5768145.78146.707107.114.48263.854.4114.68112.1335.445.498361871652899125954648.2962972.5563562.2955555.5159338.2211196.81559.317705.75617823.97217709.52616190.13316019.1616405.96813903512645741.17985.282201.87300728382.1512.267900.620073174105114755.213.677.31.2211.447.212.0236869.259531.4822528.921292.6641111.3922577.8818304.3911331.5172187233933280941318443.725716.336173.438031.38529.473903502280.5880.049.26322.6321.2717.3260.535.4816.669967779.429946928.682586196.382105.47913.4865.18182.1983.39338.3437.199417692102.82879717936.58730872247.018135857100.60128190643.0714.4574.145240.32976.578979.12222.061515.33344.6127.5499.78193.2794.72676.26233.854.40958.871848000022590000178.2633.8568.7168.8675.4875.64658.2818.51287.49349.13100.211104.9525.601421.8505147.738148.201111.715266.6654.7714.71113.8936.125.586481881672899226156856.0365210.165913.6257010.0561195.8211166.11558.718105.0918169.53118404.78517220.31817776.41917799.50114101512618941.74587.054205.306303118422.1822.270710.624593209986114660.4914.697.571.2511.857.432.0838315.110133.1323481.951345.741210.0623251.3419269.3711771.2273207834193282841318453.581116.100323.430411.36629.2463868495279.4179.289.08323.7720.8317.1659.825.4115.999898976.566825926.766308189.28942104.36880.5165.25180.3283.36335.5337.202661409101.4084400535.53709513946.5324412198.458997542.9434.3544.122239.44675.672971.78521.995512.54644.1817.382100.26193.8795.28676.53235.054.287OpenBenchmarking.org

PyTorch

This is a benchmark of PyTorch making use of pytorch-benchmark [https://github.com/LukasHedegaard/pytorch-benchmark]. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: ResNet-502 x 32GB DDR5-48002 x 32GB DDR5-5200132639526558.7558.87MIN: 53.89 / MAX: 59.52MIN: 54.65 / MAX: 59.36

Quicksilver

Quicksilver is a proxy application that represents some elements of the Mercury workload by solving a simplified dynamic Monte Carlo particle transport problem. Quicksilver is developed by Lawrence Livermore National Laboratory (LLNL) and this test profile currently makes use of the OpenMP CPU threaded code path. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFigure Of Merit, More Is BetterQuicksilver 20230818Input: CORAL2 P12 x 32GB DDR5-48002 x 32GB DDR5-52004M8M12M16M20M18550000184800001. (CXX) g++ options: -fopenmp -O3 -march=native

OpenBenchmarking.orgFigure Of Merit, More Is BetterQuicksilver 20230818Input: CORAL2 P22 x 32GB DDR5-48002 x 32GB DDR5-52005M10M15M20M25M22580000225900001. (CXX) g++ options: -fopenmp -O3 -march=native

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx265 - Scenario: Live2 x 32GB DDR5-48002 x 32GB DDR5-52004080120160200175.30178.261. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx265 - Scenario: Upload2 x 32GB DDR5-48002 x 32GB DDR5-520081624324033.6633.851. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx265 - Scenario: Platform2 x 32GB DDR5-48002 x 32GB DDR5-5200153045607568.3068.711. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx265 - Scenario: Video On Demand2 x 32GB DDR5-48002 x 32GB DDR5-5200153045607568.2768.861. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

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.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Detection FP16 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-52002040608010074.4075.481. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Detection FP32 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-52002040608010074.9175.641. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-5200140280420560700645.63658.281. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection FP16-INT8 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-520051015202518.5718.511. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-520060120180240300281.64287.491. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16-INT8 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-520080160240320400345.80349.131. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Machine Translation EN To DE FP16 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-52002040608010099.05100.211. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Vehicle Bike Detection FP16 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-520020040060080010001090.851104.951. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) 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 4.3Binary: Pathtracer ISPC - Model: Asian Dragon2 x 32GB DDR5-48002 x 32GB DDR5-520061218243025.2725.60MIN: 25.12 / MAX: 25.79MIN: 25.46 / MAX: 25.94

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.3Binary: Pathtracer ISPC - Model: Asian Dragon Obj2 x 32GB DDR5-48002 x 32GB DDR5-520051015202521.5821.85MIN: 21.45 / MAX: 21.94MIN: 21.73 / MAX: 22.2

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 2.0Encoder Mode: Preset 12 - Input: Bosphorus 4K2 x 32GB DDR5-48002 x 32GB DDR5-5200306090120150145.78147.741. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 13 - Input: Bosphorus 4K2 x 32GB DDR5-48002 x 32GB DDR5-5200306090120150146.71148.201. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

libxsmm

Libxsmm is an open-source library for specialized dense and sparse matrix operations and deep learning primitives. Libxsmm supports making use of Intel AMX, AVX-512, and other modern CPU instruction set capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 322 x 32GB DDR5-48002 x 32GB DDR5-5200306090120150107.1111.71. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -pedantic -O2 -fopenmp -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 1 - Model: AlexNet2 x 32GB DDR5-48002 x 32GB DDR5-52004812162014.4815.00

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 64 - Model: AlexNet2 x 32GB DDR5-48002 x 32GB DDR5-520060120180240300263.80266.66

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 1 - Model: GoogLeNet2 x 32GB DDR5-48002 x 32GB DDR5-5200122436486054.4154.77

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 1 - Model: ResNet-502 x 32GB DDR5-48002 x 32GB DDR5-52004812162014.6814.71

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 64 - Model: GoogLeNet2 x 32GB DDR5-48002 x 32GB DDR5-5200306090120150112.13113.89

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 64 - Model: ResNet-502 x 32GB DDR5-48002 x 32GB DDR5-520081624324035.4436.12

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity 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.orgItems Per Second, More Is BetterOSPRay 3.1Benchmark: gravity_spheres_volume/dim_512/ao/real_time2 x 32GB DDR5-48002 x 32GB DDR5-52001.2572.5143.7715.0286.2855.498365.58648

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample high resolution (currently 15400 x 6940) JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: Swirl2 x 32GB DDR5-48002 x 32GB DDR5-520040801201602001871881. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: Rotate2 x 32GB DDR5-48002 x 32GB DDR5-520040801201602001651671. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: Resizing2 x 32GB DDR5-48002 x 32GB DDR5-5200601201802403002892891. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: Noise-Gaussian2 x 32GB DDR5-48002 x 32GB DDR5-52002040608010091921. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: HWB Color Space2 x 32GB DDR5-48002 x 32GB DDR5-5200601201802403002592611. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

RAMspeed SMP

This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Add - Benchmark: Integer2 x 32GB DDR5-48002 x 32GB DDR5-520012K24K36K48K60K54648.2956856.031. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Copy - Benchmark: Integer2 x 32GB DDR5-48002 x 32GB DDR5-520014K28K42K56K70K62972.5565210.101. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Scale - Benchmark: Integer2 x 32GB DDR5-48002 x 32GB DDR5-520014K28K42K56K70K63562.2965913.621. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Triad - Benchmark: Integer2 x 32GB DDR5-48002 x 32GB DDR5-520012K24K36K48K60K55555.5157010.051. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Average - Benchmark: Integer2 x 32GB DDR5-48002 x 32GB DDR5-520013K26K39K52K65K59338.2261195.821. (CC) gcc options: -O3 -march=native

srsRAN Project

srsRAN Project is a complete ORAN-native 5G RAN solution created by Software Radio Systems (SRS). The srsRAN Project radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PDSCH Processor Benchmark, Throughput Total2 x 32GB DDR5-48002 x 32GB DDR5-52002K4K6K8K10K11196.811166.11. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -mavx512f -mavx512cd -mavx512bw -mavx512dq -fno-trapping-math -fno-math-errno -ldl

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PUSCH Processor Benchmark, Throughput Total2 x 32GB DDR5-48002 x 32GB DDR5-5200300600900120015001559.31558.7MIN: 954.9MIN: 954.91. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -mavx512f -mavx512cd -mavx512bw -mavx512dq -fno-trapping-math -fno-math-errno -ldl

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 1024 MiB2 x 32GB DDR5-48002 x 32GB DDR5-52004K8K12K16K20K17705.7618105.091. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 4096 MiB2 x 32GB DDR5-48002 x 32GB DDR5-52004K8K12K16K20K17823.9718169.531. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 8192 MiB2 x 32GB DDR5-48002 x 32GB DDR5-52004K8K12K16K20K17709.5318404.791. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 1024 MiB2 x 32GB DDR5-48002 x 32GB DDR5-52004K8K12K16K20K16190.1317220.321. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiB2 x 32GB DDR5-48002 x 32GB DDR5-52004K8K12K16K20K16019.1617776.421. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB2 x 32GB DDR5-48002 x 32GB DDR5-52004K8K12K16K20K16405.9717799.501. (CC) gcc options: -O3 -march=native

7-Zip Compression

This is a test of 7-Zip compression/decompression with its integrated benchmark feature. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Compression Rating2 x 32GB DDR5-48002 x 32GB DDR5-520030K60K90K120K150K1390351410151. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Decompression Rating2 x 32GB DDR5-48002 x 32GB DDR5-520030K60K90K120K150K1264571261891. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

JPEG-XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: PNG - Quality: 902 x 32GB DDR5-48002 x 32GB DDR5-5200102030405041.1841.751. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

JPEG-XL Decoding libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is suited for JPEG XL decode performance testing to PNG output file, the pts/jpexl test is for encode performance. The JPEG XL encoding/decoding is done using the libjxl codebase. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL Decoding libjxl 0.10.1CPU Threads: 12 x 32GB DDR5-48002 x 32GB DDR5-52002040608010085.2887.05

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL Decoding libjxl 0.10.1CPU Threads: All2 x 32GB DDR5-48002 x 32GB DDR5-520050100150200250201.87205.31

Stockfish

This is a test of Stockfish, an advanced open-source C++11 chess benchmark that can scale up to 1024 CPU threads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 16.1Chess Benchmark2 x 32GB DDR5-48002 x 32GB DDR5-52006M12M18M24M30M30072838303118421. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -funroll-loops -msse -msse3 -mpopcnt -mavx2 -mbmi -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2 -flto -flto-partition=one -flto=jobserver

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 2024Implementation: MPI CPU - Input: water_GMX50_bare2 x 32GB DDR5-48002 x 32GB DDR5-52000.4910.9821.4731.9642.4552.1512.1821. (CXX) g++ options: -O3 -lm

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.orgns/day, More Is BetterNAMD 3.0b6Input: ATPase with 327,506 Atoms2 x 32GB DDR5-48002 x 32GB DDR5-52000.51091.02181.53272.04362.55452.267902.27071

OpenBenchmarking.orgns/day, More Is BetterNAMD 3.0b6Input: STMV with 1,066,628 Atoms2 x 32GB DDR5-48002 x 32GB DDR5-52000.14050.2810.42150.5620.70250.620070.62459

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Read While Writing2 x 32GB DDR5-48002 x 32GB DDR5-5200700K1400K2100K2800K3500K317410532099861. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 10002 x 32GB DDR5-48002 x 32GB DDR5-520020K40K60K80K100K114755.20114660.491. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Llama.cpp

Llama.cpp is a port of Facebook's LLaMA model in C/C++ developed by Georgi Gerganov. Llama.cpp allows the inference of LLaMA and other supported models in C/C++. For CPU inference Llama.cpp supports AVX2/AVX-512, ARM NEON, and other modern ISAs along with features like OpenBLAS usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b1808Model: llama-2-7b.Q4_0.gguf2 x 32GB DDR5-48002 x 32GB DDR5-52004812162013.6714.691. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -march=native -mtune=native -lopenblas

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b1808Model: llama-2-13b.Q4_0.gguf2 x 32GB DDR5-48002 x 32GB DDR5-52002468107.307.571. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -march=native -mtune=native -lopenblas

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b1808Model: llama-2-70b-chat.Q5_0.gguf2 x 32GB DDR5-48002 x 32GB DDR5-52000.28130.56260.84391.12521.40651.221.251. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -march=native -mtune=native -lopenblas

Llamafile

Mozilla's Llamafile allows distributing and running large language models (LLMs) as a single file. Llamafile aims to make open-source LLMs more accessible to developers and users. Llamafile supports a variety of models, CPUs and GPUs, and other options. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.7Test: llava-v1.5-7b-q4 - Acceleration: CPU2 x 32GB DDR5-48002 x 32GB DDR5-5200369121511.4411.85

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.7Test: mistral-7b-instruct-v0.2.Q8_0 - Acceleration: CPU2 x 32GB DDR5-48002 x 32GB DDR5-52002468107.217.43

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.7Test: wizardcoder-python-34b-v1.0.Q6_K - Acceleration: CPU2 x 32GB DDR5-48002 x 32GB DDR5-52000.4680.9361.4041.8722.342.022.08

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.C2 x 32GB DDR5-48002 x 32GB DDR5-52008K16K24K32K40K36869.2538315.101. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.C2 x 32GB DDR5-48002 x 32GB DDR5-52002K4K6K8K10K9531.4810133.131. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: FT.C2 x 32GB DDR5-48002 x 32GB DDR5-52005K10K15K20K25K22528.9223481.951. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.D2 x 32GB DDR5-48002 x 32GB DDR5-5200300600900120015001292.661345.701. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.C2 x 32GB DDR5-48002 x 32GB DDR5-52009K18K27K36K45K41111.3941210.061. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.C2 x 32GB DDR5-48002 x 32GB DDR5-52005K10K15K20K25K22577.8823251.341. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.B2 x 32GB DDR5-48002 x 32GB DDR5-52004K8K12K16K20K18304.3919269.371. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.C2 x 32GB DDR5-48002 x 32GB DDR5-52003K6K9K12K15K11331.5111771.221. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

PostgreSQL

This is a benchmark of PostgreSQL using the integrated pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Only2 x 32GB DDR5-48002 x 32GB DDR5-5200160K320K480K640K800K7218727320781. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Write2 x 32GB DDR5-48002 x 32GB DDR5-52007K14K21K28K35K33933341931. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

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 7.38.2VGR Performance Metric2 x 32GB DDR5-48002 x 32GB DDR5-520060K120K180K240K300K2809412828411. (CXX) g++ options: -std=c++17 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -ltcl8.6 -lnetpbm -lregex_brl -lz_brl -lassimp -ldl -lm -ltk8.6

Chaos Group V-RAY

This is a test of Chaos Group's V-RAY benchmark. V-RAY is a commercial renderer that can integrate with various creator software products like SketchUp and 3ds Max. The V-RAY benchmark is standalone and supports CPU and NVIDIA CUDA/RTX based rendering. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgvsamples, More Is BetterChaos Group V-RAY 6.0Mode: CPU2 x 32GB DDR5-48002 x 32GB DDR5-52007K14K21K28K35K3184431845

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: IP Shapes 3D - Engine: CPU2 x 32GB DDR5-48002 x 32GB DDR5-52000.83831.67662.51493.35324.19153.725713.58111MIN: 3.69MIN: 3.541. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Convolution Batch Shapes Auto - Engine: CPU2 x 32GB DDR5-48002 x 32GB DDR5-52002468106.336176.10032MIN: 6.24MIN: 5.991. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Deconvolution Batch shapes_3d - Engine: CPU2 x 32GB DDR5-48002 x 32GB DDR5-52000.77361.54722.32083.09443.8683.438033.43041MIN: 3.21MIN: 3.211. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

PostgreSQL

This is a benchmark of PostgreSQL using the integrated pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency2 x 32GB DDR5-48002 x 32GB DDR5-52000.31160.62320.93481.24641.5581.3851.3661. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latency2 x 32GB DDR5-48002 x 32GB DDR5-520071421283529.4729.251. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Google Draco

Draco is a library developed by Google for compressing/decompressing 3D geometric meshes and point clouds. This test profile uses some Artec3D PLY models as the sample 3D model input formats for Draco compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: Lion2 x 32GB DDR5-48002 x 32GB DDR5-52008001600240032004000390338681. (CXX) g++ options: -O3

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: Church Facade2 x 32GB DDR5-48002 x 32GB DDR5-520011002200330044005500502249521. (CXX) g++ options: -O3

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 2024.0Model: Person Detection FP16 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-52002040608010080.5879.41MIN: 60.22 / MAX: 102.9MIN: 56.3 / MAX: 102.741. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Detection FP32 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-52002040608010080.0479.28MIN: 61.76 / MAX: 107.58MIN: 60.22 / MAX: 100.151. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-520036912159.269.08MIN: 4.6 / MAX: 15.49MIN: 4.23 / MAX: 17.051. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection FP16-INT8 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-520070140210280350322.63323.77MIN: 296.85 / MAX: 328.51MIN: 308.71 / MAX: 330.441. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-520051015202521.2720.83MIN: 9.77 / MAX: 37.47MIN: 12.6 / MAX: 29.811. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16-INT8 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-52004812162017.3217.16MIN: 8.73 / MAX: 23.67MIN: 13.76 / MAX: 231. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Machine Translation EN To DE FP16 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-5200142842567060.5359.82MIN: 33.29 / MAX: 82.21MIN: 48.42 / MAX: 67.391. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Vehicle Bike Detection FP16 - Device: CPU2 x 32GB DDR5-48002 x 32GB DDR5-52001.2332.4663.6994.9326.1655.485.41MIN: 3.89 / MAX: 9.6MIN: 3.92 / MAX: 8.181. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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 129 Cells Per Direction2 x 32GB DDR5-48002 x 32GB DDR5-52004812162016.6716.001. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per Direction2 x 32GB DDR5-48002 x 32GB DDR5-52002040608010079.4376.571. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

OpenFOAM

OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Mesh Time2 x 32GB DDR5-48002 x 32GB DDR5-520071421283528.6826.771. (CXX) g++ options: -std=c++14 -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 10Input: drivaerFastback, Small Mesh Size - Execution Time2 x 32GB DDR5-48002 x 32GB DDR5-52004080120160200196.38189.291. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/ and https://github.com/OpenRadioss/ModelExchange/tree/main/Examples. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: Bumper Beam2 x 32GB DDR5-48002 x 32GB DDR5-520020406080100105.47104.36

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: Chrysler Neon 1M2 x 32GB DDR5-48002 x 32GB DDR5-52002004006008001000913.48880.51

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: Cell Phone Drop Test2 x 32GB DDR5-48002 x 32GB DDR5-5200153045607565.1865.25

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: Bird Strike on Windshield2 x 32GB DDR5-48002 x 32GB DDR5-52004080120160200182.19180.32

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: Rubber O-Ring Seal Installation2 x 32GB DDR5-48002 x 32GB DDR5-52002040608010083.3983.36

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: INIVOL and Fluid Structure Interaction Drop Container2 x 32GB DDR5-48002 x 32GB DDR5-520070140210280350338.34335.53

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Mount St. Helens2 x 32GB DDR5-48002 x 32GB DDR5-520091827364537.2037.201. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Layered Halfspace2 x 32GB DDR5-48002 x 32GB DDR5-520020406080100102.83101.411. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Tomographic Model2 x 32GB DDR5-48002 x 32GB DDR5-520081624324036.5935.541. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Homogeneous Halfspace2 x 32GB DDR5-48002 x 32GB DDR5-5200112233445547.0246.531. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Water-layered Halfspace2 x 32GB DDR5-48002 x 32GB DDR5-520020406080100100.6098.461. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

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 1.0Encoder Speed: 22 x 32GB DDR5-48002 x 32GB DDR5-5200102030405043.0742.941. (CXX) g++ options: -O3 -fPIC -lm

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 62 x 32GB DDR5-48002 x 32GB DDR5-52001.00282.00563.00844.01125.0144.4574.3541. (CXX) g++ options: -O3 -fPIC -lm

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 10, Lossless2 x 32GB DDR5-48002 x 32GB DDR5-52000.93261.86522.79783.73044.6634.1454.1221. (CXX) g++ options: -O3 -fPIC -lm

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 4.0Time To Compile2 x 32GB DDR5-48002 x 32GB DDR5-520050100150200250240.33239.45

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 or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.8Build: defconfig2 x 32GB DDR5-48002 x 32GB DDR5-52002040608010076.5875.67

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.8Build: allmodconfig2 x 32GB DDR5-48002 x 32GB DDR5-52002004006008001000979.12971.79

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 24.0Time To Compile2 x 32GB DDR5-48002 x 32GB DDR5-520051015202522.0622.00

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 21.7.2Time To Compile2 x 32GB DDR5-48002 x 32GB DDR5-5200110220330440550515.33512.55

Timed PHP Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 8.3.4Time To Compile2 x 32GB DDR5-48002 x 32GB DDR5-5200102030405044.6144.18

RNNoise

RNNoise is a recurrent neural network for audio noise reduction developed by Mozilla and Xiph.Org. This test profile is a single-threaded test measuring the time to denoise a sample 26 minute long 16-bit RAW audio file using this recurrent neural network noise suppression library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRNNoise 0.2Input: 26 Minute Long Talking Sample2 x 32GB DDR5-48002 x 32GB DDR5-52002468107.5407.3821. (CC) gcc options: -O2 -pedantic -fvisibility=hidden

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: Junkshop - Compute: CPU-Only2 x 32GB DDR5-48002 x 32GB DDR5-52002040608010099.78100.26

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: Classroom - Compute: CPU-Only2 x 32GB DDR5-48002 x 32GB DDR5-52004080120160200193.27193.87

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: Fishy Cat - Compute: CPU-Only2 x 32GB DDR5-48002 x 32GB DDR5-52002040608010094.7295.28

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: Barbershop - Compute: CPU-Only2 x 32GB DDR5-48002 x 32GB DDR5-5200150300450600750676.26676.53

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.1Blend File: Pabellon Barcelona - Compute: CPU-Only2 x 32GB DDR5-48002 x 32GB DDR5-520050100150200250233.85235.05

WavPack Audio Encoding

This test times how long it takes to encode a sample WAV file to WavPack format with very high quality settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.7WAV To WavPack2 x 32GB DDR5-48002 x 32GB DDR5-52000.9921.9842.9763.9684.964.4094.287

120 Results Shown

PyTorch
Quicksilver:
  CORAL2 P1
  CORAL2 P2
FFmpeg:
  libx265 - Live
  libx265 - Upload
  libx265 - Platform
  libx265 - Video On Demand
OpenVINO:
  Person Detection FP16 - CPU
  Person Detection FP32 - CPU
  Vehicle Detection FP16 - CPU
  Face Detection FP16-INT8 - CPU
  Road Segmentation ADAS FP16 - CPU
  Road Segmentation ADAS FP16-INT8 - CPU
  Machine Translation EN To DE FP16 - CPU
  Person Vehicle Bike Detection FP16 - CPU
Embree:
  Pathtracer ISPC - Asian Dragon
  Pathtracer ISPC - Asian Dragon Obj
SVT-AV1:
  Preset 12 - Bosphorus 4K
  Preset 13 - Bosphorus 4K
libxsmm
TensorFlow:
  CPU - 1 - AlexNet
  CPU - 64 - AlexNet
  CPU - 1 - GoogLeNet
  CPU - 1 - ResNet-50
  CPU - 64 - GoogLeNet
  CPU - 64 - ResNet-50
OSPRay
GraphicsMagick:
  Swirl
  Rotate
  Resizing
  Noise-Gaussian
  HWB Color Space
RAMspeed SMP:
  Add - Integer
  Copy - Integer
  Scale - Integer
  Triad - Integer
  Average - Integer
srsRAN Project:
  PDSCH Processor Benchmark, Throughput Total
  PUSCH Processor Benchmark, Throughput Total
MBW:
  Memory Copy - 1024 MiB
  Memory Copy - 4096 MiB
  Memory Copy - 8192 MiB
  Memory Copy, Fixed Block Size - 1024 MiB
  Memory Copy, Fixed Block Size - 4096 MiB
  Memory Copy, Fixed Block Size - 8192 MiB
7-Zip Compression:
  Compression Rating
  Decompression Rating
JPEG-XL libjxl
JPEG-XL Decoding libjxl:
  1
  All
Stockfish
GROMACS
NAMD:
  ATPase with 327,506 Atoms
  STMV with 1,066,628 Atoms
RocksDB
nginx
Llama.cpp:
  llama-2-7b.Q4_0.gguf
  llama-2-13b.Q4_0.gguf
  llama-2-70b-chat.Q5_0.gguf
Llamafile:
  llava-v1.5-7b-q4 - CPU
  mistral-7b-instruct-v0.2.Q8_0 - CPU
  wizardcoder-python-34b-v1.0.Q6_K - CPU
NAS Parallel Benchmarks:
  BT.C
  CG.C
  FT.C
  IS.D
  LU.C
  MG.C
  SP.B
  SP.C
PostgreSQL:
  100 - 1000 - Read Only
  100 - 1000 - Read Write
BRL-CAD
Chaos Group V-RAY
oneDNN:
  IP Shapes 3D - CPU
  Convolution Batch Shapes Auto - CPU
  Deconvolution Batch shapes_3d - CPU
PostgreSQL:
  100 - 1000 - Read Only - Average Latency
  100 - 1000 - Read Write - Average Latency
Google Draco:
  Lion
  Church Facade
OpenVINO:
  Person Detection FP16 - CPU
  Person Detection FP32 - CPU
  Vehicle Detection FP16 - CPU
  Face Detection FP16-INT8 - CPU
  Road Segmentation ADAS FP16 - CPU
  Road Segmentation ADAS FP16-INT8 - CPU
  Machine Translation EN To DE FP16 - CPU
  Person Vehicle Bike Detection FP16 - CPU
Xcompact3d Incompact3d:
  input.i3d 129 Cells Per Direction
  input.i3d 193 Cells Per Direction
OpenFOAM:
  drivaerFastback, Small Mesh Size - Mesh Time
  drivaerFastback, Small Mesh Size - Execution Time
OpenRadioss:
  Bumper Beam
  Chrysler Neon 1M
  Cell Phone Drop Test
  Bird Strike on Windshield
  Rubber O-Ring Seal Installation
  INIVOL and Fluid Structure Interaction Drop Container
SPECFEM3D:
  Mount St. Helens
  Layered Halfspace
  Tomographic Model
  Homogeneous Halfspace
  Water-layered Halfspace
libavif avifenc:
  2
  6
  10, Lossless
Timed Godot Game Engine Compilation
Timed Linux Kernel Compilation:
  defconfig
  allmodconfig
Timed Mesa Compilation
Timed Node.js Compilation
Timed PHP Compilation
RNNoise
Blender:
  Junkshop - CPU-Only
  Classroom - CPU-Only
  Fishy Cat - CPU-Only
  Barbershop - CPU-Only
  Pabellon Barcelona - CPU-Only
WavPack Audio Encoding