2024 Server CPUs

Benchmarks by Michael Larabel for a future article. 2 x Intel Xeon Max 9480 testing with a Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS (3B05.TEL4P1 BIOS) and ASPEED on Ubuntu 24.04 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2405073-NE-UPLOAD15789
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

Limit displaying results to tests within:

AV1 2 Tests
BLAS (Basic Linear Algebra Sub-Routine) Tests 4 Tests
C++ Boost Tests 5 Tests
Chess Test Suite 2 Tests
Timed Code Compilation 11 Tests
C/C++ Compiler Tests 18 Tests
Compression Tests 3 Tests
CPU Massive 28 Tests
Creator Workloads 26 Tests
Cryptography 5 Tests
Database Test Suite 6 Tests
Encoding 7 Tests
Fortran Tests 7 Tests
Game Development 6 Tests
HPC - High Performance Computing 30 Tests
Imaging 5 Tests
Java Tests 2 Tests
Common Kernel Benchmarks 3 Tests
LAPACK (Linear Algebra Pack) Tests 3 Tests
Linear Algebra 2 Tests
Machine Learning 7 Tests
Molecular Dynamics 10 Tests
MPI Benchmarks 10 Tests
Multi-Core 47 Tests
NVIDIA GPU Compute 3 Tests
Intel oneAPI 7 Tests
OpenMPI Tests 19 Tests
Programmer / Developer System Benchmarks 16 Tests
Python 2 Tests
Raytracing 2 Tests
Renderers 5 Tests
Scientific Computing 15 Tests
Software Defined Radio 2 Tests
Server 11 Tests
Server CPU Tests 19 Tests
Single-Threaded 3 Tests
Texture Compression 2 Tests
Video Encoding 7 Tests
Common Workstation Benchmarks 3 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
Disable Color Branding
Prefer Vertical Bar Graphs

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable
Show Perf Per Clock Calculation Graphs Where Applicable

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
Xeon Platinum 8490H
April 16
  1 Day, 1 Hour, 45 Minutes
Xeon Platinum 8490H 2P
April 17
  1 Day, 3 Hours, 9 Minutes
Xeon Platinum 8592+
April 11
  1 Day, 2 Hours, 43 Minutes
Xeon Platinum 8592+ 2P
April 08
  1 Day, 4 Hours, 29 Minutes
Xeon Max 9468
April 21
  1 Day, 8 Hours, 22 Minutes
Xeon Max 9468 2P
April 22
  1 Day, 5 Hours, 47 Minutes
Xeon Max 9480
April 26
  1 Day, 6 Hours, 43 Minutes
Xeon Max 9480 2P
April 27
  1 Day, 12 Hours, 13 Minutes
Invert Hiding All Results Option
  1 Day, 5 Hours, 39 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):


2024 Server CPUsProcessorMotherboardChipsetMemoryDiskGraphicsNetworkOSKernelCompilerFile-SystemScreen ResolutionXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2PXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PIntel Xeon Platinum 8490H @ 3.50GHz (60 Cores / 120 Threads)Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS (3B05.TEL4P1 BIOS)Intel Device 1bce512GB2 x 1920GB KIOXIA KCD8XPUG1T92ASPEEDUbuntu 24.046.9.0-060900rc3-generic (x86_64)GCC 13.2.0ext41920x12002 x Intel Xeon Platinum 8490H @ 3.50GHz (120 Cores / 240 Threads)1008GB2 x Intel X710 for 10GBASE-TINTEL XEON PLATINUM 8592+ @ 3.90GHz (64 Cores / 128 Threads)512GB2 x INTEL XEON PLATINUM 8592+ @ 3.90GHz (128 Cores / 256 Threads)1008GB2 x Intel X710 for 10GBASE-TIntel Xeon Max 9468 @ 3.50GHz (48 Cores / 96 Threads)576GB2 x Intel Xeon Max 9468 @ 3.50GHz (96 Cores / 192 Threads)1136GB2 x Intel X710 for 10GBASE-TIntel Xeon Max 9480 @ 3.50GHz (56 Cores / 112 Threads)576GB2 x Intel Xeon Max 9480 @ 3.50GHz (112 Cores / 224 Threads)1136GB2 x Intel X710 for 10GBASE-TOpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --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-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/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 Processor Details- Xeon Platinum 8490H: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b000590- Xeon Platinum 8490H 2P: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b000590- Xeon Platinum 8592+: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x21000200- Xeon Platinum 8592+ 2P: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x21000200- Xeon Max 9468: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2c000290- Xeon Max 9468 2P: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2c000290- Xeon Max 9480: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2c000290- Xeon Max 9480 2P: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2c000290Java Details- OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)Python Details- Python 3.12.2Security Details- 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: Not affected + 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 RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected

Xeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2PXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PLogarithmic Result OverviewPhoronix Test SuiteBRL-CADASKAPLlamafileStockfishCoremarkLULESHNAS Parallel BenchmarksOSPRay StudioXmrigoneDNNACES DGEMMminiBUDEOpenSSLAircrack-ngm-queensJohn The RipperPrimesieveHigh Performance Conjugate GradientEmbreeRELIONASTC EncoderOpenVKLHelsingminiFEOSPRayeasyWaveBlenderXcompact3d Incompact3dGROMACS7-Zip CompressionPennantNAMDSpeedbAlgebraic Multi-Grid BenchmarkWRFIntel Open Image DenoiseRocksDBPostgreSQLLAMMPS Molecular Dynamics SimulatorTimed LLVM CompilationGNU Octave BenchmarkTimed Linux Kernel CompilationOpenVINOTimed Node.js CompilationParallel BZIP2 CompressionPyTorchCockroachDBQuantLibCloverLeafKvazaarOpenFOAMONNX RuntimeY-Cruncheruvg266Liquid-DSPTimed FFmpeg CompilationsrsRAN ProjectMemcachedTimed Godot Game Engine Compilationx265GPAWGoogle DracoSVT-AV1Timed Gem5 CompilationNWChemTimed ImageMagick CompilationAppleseedDaCapo BenchmarkRawTherapeeVVenCJPEG-XL libjxlGraphicsMagickDuckDBOpenRadiossZstd CompressionTimed Mesa CompilationLuxCoreRenderQMCPACKlibavif avifencTimed PHP CompilationNode.js V8 Web Tooling BenchmarkTimed Wasmer CompilationTensorFlowNumpy BenchmarkFFmpegApache IoTDBWebP Image EncodeClickHouseTimed CPython CompilationSecureMarkGoogle SynthMarkPyBench

2024 Server CPUswrf: conus 2.5kmopenvkl: vklBenchmarkCPU ISPCaskap: tConvolve MT - Degriddingnwchem: C240 Buckyballrelion: Basic - CPUeasywave: e2Asean Grid + BengkuluSept2007 Source - 2400askap: tConvolve MT - Griddingcockroach: KV, 95% Reads - 512brl-cad: VGR Performance Metriccockroach: KV, 60% Reads - 512pgbench: 100 - 1000 - Read Only - Average Latencypgbench: 100 - 1000 - Read Onlytensorflow: CPU - 512 - ResNet-50cloverleaf: clover_bm16namd: STMV with 1,066,628 Atomshpcg: 144 144 144 - 60apache-iotdb: 500 - 100 - 800 - 400apache-iotdb: 500 - 100 - 800 - 400clickhouse: 100M Rows Hits Dataset, Third Runclickhouse: 100M Rows Hits Dataset, Second Runclickhouse: 100M Rows Hits Dataset, First Run / Cold Cacheopenssl: RSA4096openssl: RSA4096tensorflow: CPU - 256 - ResNet-50build-linux-kernel: allmodconfigincompact3d: X3D-benchmarking input.i3dstockfish: Chess Benchmarksecuremark: SecureMark-TLSapache-iotdb: 800 - 100 - 800 - 100apache-iotdb: 800 - 100 - 800 - 100cockroach: KV, 50% Reads - 512pytorch: CPU - 64 - ResNet-152luxcorerender: Danish Mood - CPUbuild-gem5: Time To Compilerocksdb: Read While Writinglammps: 20k Atomsblender: Barbershop - CPU-Onlyllamafile: wizardcoder-python-34b-v1.0.Q6_K - CPUbuild-nodejs: Time To Compileonnx: fcn-resnet101-11 - CPU - Standardonnx: fcn-resnet101-11 - CPU - Standardduckdb: TPC-H Parqueteasywave: e2Asean Grid + BengkuluSept2007 Source - 1200openradioss: Chrysler Neon 1Mluxcorerender: DLSC - CPUffmpeg: libx265 - Uploadonnx: yolov4 - CPU - Standardonnx: yolov4 - CPU - Standardopenssl: AES-128-GCMopenssl: AES-256-GCMopenssl: ChaCha20-Poly1305openssl: SHA512openssl: ChaCha20openssl: SHA256speedb: Rand Readffmpeg: libx265 - Video On Demandffmpeg: libx265 - Platformpytorch: CPU - 512 - ResNet-152compress-zstd: 19 - Decompression Speedcompress-zstd: 19 - Compression Speedonednn: Recurrent Neural Network Inference - CPUjohn-the-ripper: MD5pytorch: CPU - 256 - ResNet-152ospray: particle_volume/scivis/real_timeduckdb: IMDBospray: particle_volume/pathtracer/real_timeospray: gravity_spheres_volume/dim_512/scivis/real_timeluxcorerender: Orange Juice - CPUbuild-llvm: Ninjanamd: ATPase with 327,506 Atomsvvenc: Bosphorus 4K - Fastnumpy: pytorch: CPU - 512 - ResNet-50luxcorerender: LuxCore Benchmark - CPUopenfoam: drivaerFastback, Medium Mesh Size - Execution Timeopenfoam: drivaerFastback, Medium Mesh Size - Mesh Timepgbench: 100 - 1000 - Read Write - Average Latencypgbench: 100 - 1000 - Read Writeopenradioss: Rubber O-Ring Seal Installationopenradioss: INIVOL and Fluid Structure Interaction Drop Containercompress-zstd: 19, Long Mode - Decompression Speedcompress-zstd: 19, Long Mode - Compression Speedbuild-godot: Time To Compilexmrig: GhostRider - 1Mtensorflow: CPU - 512 - GoogLeNetonnx: CaffeNet 12-int8 - CPU - Standardonnx: CaffeNet 12-int8 - CPU - Standardonnx: GPT-2 - CPU - Standardonnx: GPT-2 - CPU - Standardonnx: T5 Encoder - CPU - Parallelonnx: T5 Encoder - CPU - Parallelopenradioss: Bird Strike on Windshieldqmcpack: Li2_STO_aeappleseed: Material Testeronnx: ArcFace ResNet-100 - CPU - Standardonnx: ArcFace ResNet-100 - CPU - Standardospray: gravity_spheres_volume/dim_512/ao/real_timeospray: particle_volume/ao/real_timeonednn: Recurrent Neural Network Training - CPUspeedb: Read While Writingmemcached: 1:100dacapobench: Tradesoaponnx: ResNet50 v1-12-int8 - CPU - Standardonnx: ResNet50 v1-12-int8 - CPU - Standardonnx: super-resolution-10 - CPU - Standardonnx: super-resolution-10 - CPU - Standardpytorch: CPU - 1 - ResNet-152onnx: bertsquad-12 - CPU - Standardonnx: bertsquad-12 - CPU - Standardopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUonnx: T5 Encoder - CPU - Standardonnx: T5 Encoder - CPU - Standardbuild-python: Released Build, PGO + LTO Optimizedappleseed: Emilyonnx: fcn-resnet101-11 - CPU - Parallelonnx: fcn-resnet101-11 - CPU - Parallelopenradioss: Bumper Beamaircrack-ng: helsing: 14 digitdacapobench: Tradebeanstensorflow: CPU - 1 - ResNet-50tensorflow: CPU - 64 - ResNet-50blender: Pabellon Barcelona - CPU-Onlyavifenc: 0ospray-studio: 3 - 4K - 1 - Path Tracer - CPUtensorflow: CPU - 256 - GoogLeNetgraphics-magick: Rotateospray-studio: 1 - 4K - 1 - Path Tracer - CPUopenvino: Face Detection FP16-INT8 - CPUopenvino: Face Detection FP16-INT8 - CPUospray-studio: 3 - 4K - 32 - Path Tracer - CPUminibude: OpenMP - BM2minibude: OpenMP - BM2ospray: gravity_spheres_volume/dim_512/pathtracer/real_timeospray-studio: 1 - 4K - 32 - Path Tracer - CPUospray-studio: 1 - 4K - 16 - Path Tracer - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Noise Suppression Poconet-Like FP16 - CPUopenvino: Noise Suppression Poconet-Like FP16 - CPUospray-studio: 3 - 4K - 16 - Path Tracer - CPUonnx: CaffeNet 12-int8 - CPU - Parallelonnx: CaffeNet 12-int8 - CPU - Parallelopenvino: Person Vehicle Bike Detection FP16 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUopenvino: Person Re-Identification Retail FP16 - CPUopenvino: Person Re-Identification Retail FP16 - CPUopenvino: Road Segmentation ADAS FP16-INT8 - CPUopenvino: Road Segmentation ADAS FP16-INT8 - CPUquantlib: Multi-Threadedonnx: Faster R-CNN R-50-FPN-int8 - CPU - Standardonnx: Faster R-CNN R-50-FPN-int8 - CPU - Standardonnx: Faster R-CNN R-50-FPN-int8 - CPU - Parallelonnx: Faster R-CNN R-50-FPN-int8 - CPU - Paralleljpegxl: JPEG - 90openvino: Person Detection FP16 - CPUopenvino: Person Detection FP16 - CPUonnx: GPT-2 - CPU - Parallelonnx: GPT-2 - CPU - Parallelopenvino: Machine Translation EN To DE FP16 - CPUopenvino: Machine Translation EN To DE FP16 - CPUnode-web-tooling: openvino: Vehicle Detection FP16-INT8 - CPUopenvino: Vehicle Detection FP16-INT8 - CPUonnx: yolov4 - CPU - Parallelonnx: yolov4 - CPU - Parallelonnx: bertsquad-12 - CPU - Parallelonnx: bertsquad-12 - CPU - Parallelopenvino: Face Detection Retail FP16-INT8 - CPUopenvino: Face Detection Retail FP16-INT8 - CPUonnx: ArcFace ResNet-100 - CPU - Parallelonnx: ArcFace ResNet-100 - CPU - Parallelopenvino: Handwritten English Recognition FP16-INT8 - CPUopenvino: Handwritten English Recognition FP16-INT8 - CPUgraphics-magick: Sharpengraphics-magick: Noise-Gaussiangraphics-magick: Resizingonnx: ResNet50 v1-12-int8 - CPU - Parallelonnx: ResNet50 v1-12-int8 - CPU - Parallelgraphics-magick: HWB Color Spacegraphics-magick: Enhancedonnx: super-resolution-10 - CPU - Parallelonnx: super-resolution-10 - CPU - Parallelgraphics-magick: Swirlrocksdb: Rand Readblender: Classroom - CPU-Onlypytorch: CPU - 256 - ResNet-50pytorch: CPU - 64 - ResNet-50cloverleaf: clover_bm64_shortllamafile: llava-v1.5-7b-q4 - CPUsrsran: PDSCH Processor Benchmark, Throughput Totalrawtherapee: Total Benchmark Timegpaw: Carbon Nanotubevvenc: Bosphorus 4K - Fastercoremark: CoreMark Size 666 - Iterations Per Secondminife: Smallbuild-php: Time To Compilejohn-the-ripper: WPA PSKcompress-7zip: Decompression Ratingcompress-7zip: Compression Ratingluxcorerender: Rainbow Colors and Prism - CPUliquid-dsp: 1 - 256 - 512build-linux-kernel: defconfigwebp: Quality 100, Lossless, Highest Compressiontensorflow: CPU - 512 - AlexNetjpegxl: PNG - 90primesieve: 1e13graph500: 26graph500: 26graph500: 26graph500: 26liquid-dsp: 1 - 256 - 57srsran: PUSCH Processor Benchmark, Throughput Totalavifenc: 2askap: tConvolve MPI - Griddingaskap: tConvolve MPI - Degriddingopenradioss: Cell Phone Drop Testliquid-dsp: 64 - 256 - 512liquid-dsp: 1 - 256 - 32build-wasmer: Time To Compilepytorch: CPU - 1 - ResNet-50appleseed: Disney Materialastcenc: Very Thoroughliquid-dsp: 64 - 256 - 57uvg266: Bosphorus 4K - Slowastcenc: Exhaustiveliquid-dsp: 128 - 256 - 512svt-av1: Preset 13 - Bosphorus 4Kquantlib: Single-Threadedliquid-dsp: 128 - 256 - 57nginx: 1000liquid-dsp: 256 - 256 - 512blender: Junkshop - CPU-Onlynpb: EP.Djohn-the-ripper: bcryptjohn-the-ripper: Blowfishliquid-dsp: 256 - 256 - 57liquid-dsp: 256 - 256 - 32uvg266: Bosphorus 4K - Mediumliquid-dsp: 128 - 256 - 32synthmark: VoiceMark_100liquid-dsp: 64 - 256 - 32blender: Fishy Cat - CPU-Onlytensorflow: CPU - 1 - GoogLeNetdacapobench: Apache Lucene Search Indexembree: Pathtracer ISPC - Asian Dragon Objgromacs: MPI CPU - water_GMX50_baresvt-av1: Preset 4 - Bosphorus 4Ktensorflow: CPU - 64 - GoogLeNetaskap: tConvolve OpenMP - Degriddingaskap: tConvolve OpenMP - Griddingkvazaar: Bosphorus 4K - Slowkvazaar: Bosphorus 4K - Mediumsvt-av1: Preset 8 - Bosphorus 4Kuvg266: Bosphorus 4K - Ultra Fasttensorflow: CPU - 256 - AlexNetminibude: OpenMP - BM1minibude: OpenMP - BM1tensorflow: CPU - 1 - AlexNetblender: BMW27 - CPU-Onlyy-cruncher: 1Bopenfoam: drivaerFastback, Small Mesh Size - Execution Timeopenfoam: drivaerFastback, Small Mesh Size - Mesh Timenpb: IS.Dkvazaar: Bosphorus 4K - Super Fastamg: incompact3d: input.i3d 193 Cells Per Directionbuild-ffmpeg: Time To Compileoidn: RTLightmap.hdr.4096x4096 - CPU-Onlyonednn: Deconvolution Batch shapes_1d - CPUx265: Bosphorus 4Kuvg266: Bosphorus 4K - Very Fastlulesh: onednn: IP Shapes 1D - CPUtensorflow: CPU - 64 - AlexNetnpb: SP.Cpennant: sedovbigkvazaar: Bosphorus 4K - Ultra Fastuvg266: Bosphorus 4K - Super Fastdacapobench: Jythonjpegxl: PNG - 100build-mesa: Time To Compilepybench: Total For Average Test Timesdacapobench: Apache Kafkawebp: Quality 100, Losslessbuild-imagemagick: Time To Compilejpegxl: JPEG - 100astcenc: Thoroughnpb: LU.Cm-queens: Time To Solvesrsran: PUSCH Processor Benchmark, Throughput Thready-cruncher: 500Mnpb: CG.Ckvazaar: Bosphorus 4K - Very Fastembree: Pathtracer ISPC - Asian Dragonoctave-benchmark: oidn: RT.hdr_alb_nrm.3840x2160 - CPU-Onlypennant: leblancbigoidn: RT.ldr_alb_nrm.3840x2160 - CPU-Onlyembree: Pathtracer ISPC - Crownwebp: Quality 100, Highest Compressiononednn: IP Shapes 3D - CPUdraco: Church Facadeonednn: Convolution Batch Shapes Auto - CPUmt-dgemm: Sustained Floating-Point Rateastcenc: Mediumavifenc: 6, Losslesssvt-av1: Preset 12 - Bosphorus 4Kdraco: Lionlammps: Rhodopsin Proteinavifenc: 10, Losslesssrsran: PDSCH Processor Benchmark, Throughput Threadprimesieve: 1e12avifenc: 6npb: MG.Ccompress-pbzip2: FreeBSD-13.0-RELEASE-amd64-memstick.img Compressiononednn: Deconvolution Batch shapes_3d - CPUwebp: Quality 100build-python: DefaultXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2PXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2P11980.80617342070.3373.182119.8378088.11131999.7833892116603.70.5721748146148.45328.471.0527431.0738294.3387797206481.66486.05465.50974189.927431.1143.67362.454357.0097129688747927720585.1690871770108953.620.136.85210.934784173834.892261.975.85254.003116.3158.59719145.57446.289207.208.7224.9750.352219.8593768804693697700239352717225038850730222199277233161264211736180699452327617961749.8449.7920.461032.014.6409.592958300020.2521.893299.493160.24219.576213.46187.0172.970216.719455.5756.447.10420.6467160.7268514.0407122759.91140.431069.17.94140.2335808.8461.231.17292852.3264.49976222.0293.06500326.143117.41109.30126.15224718.931952.816719.845821.9324623.604106953604481667.9541153.17002315.4104.39649227.42319.4149.637620.14530.43115106.302.79694357.355277.39120.750285329.0363.0392279.54147469.32893.348727010.53120.9387.9269.9311977448.461961657356.38167.836800686.9672174.19224.644558026313803.5815943.1713.954209.10366491.27368784.30416.503606.129.795943.6655.991068.43179647.529.408034.002130.916532.343343.202100.39298.565.21520191.56262.46477.7715.7617.843325.9055.120418.141362.127616.09566.439004.2321.343646.849136.201643.431571582293.19298313.1002652164.86604205.52550226536617068.6154.8456.0139.4820.9524688.546.75747.77615.0602194630.60044935230.148.00740127828786238819013.541688233339.4500.521278.4542.90551.820670746672616.937.75221314.217891.631.497444100003650600034.93351.4653.4902767.0885253920000015.714.2689918566667162.4743209.63181266667102083333335.967562.8094551945133500833333324380000017.853200600000711.378189056666734.8333.56358179.38058.7627.061380.1430471.519798.520.2320.9770.90860.871205.3082.8482071.18956.5325.897.04742.21941931.3845852865.7266.74160509700012.929341324.0561.296.1585637.8649.4223072.3020.539467969.6271245.2616.0308074.5755.54380128.10019.22676551061.3114.72428.75451.2971135125.1515.211129.73.26733811.6350.0391.99257.5322.714.9558652.7170.05403.080.77150656700.65061823.543705333.99026.278164.250473128.7164.923680.24.5763.16983448.702.1448571.0774310.1119.2016674.1227161960.9203.407126.3887630.0682341.2413696972110.11.293778141174.95214.801.1090060.8515317.3680169620501.63507.56485.631916162.954601.1161.06218.570196.80811618776176927547292.238365178568524.714.037.00178.2821232130050.521141.543.82172.99092.347310.83623155.73050.834115.8911.1224.5664.285815.5555153302630072713695343284704422025329774284625447762440297294711910680210755128695848.7748.7614.59963.615.2633.2881816600014.0838.2269125.882149.08835.602618.70122.3142.219486.480456.1837.558.82196.49644162.0266914.8566731588.39107.36992.18.16118.57211263.2576.081.43573696.9805.55101180.0324.05881246.335124.44101.629225.3469227.594436.249336.152938.2235635.154172970273245908.4042155.62517177.7634.72530212.92415.7760.358916.57050.29189893.313.46571288.471279.692130.618252352.7382.8349881.3053914.93350.83180504.39115.4746.3867.3751063528.93172894283.11423.1838813140.8283520.70542.818833411141382.5047654.9515.967356.63168371.76353566.37715.047862.587.7215519.9265.021843.61350187.529.753633.608233.289930.036842.04045.78654.846.90619144.64037.58793.6415.4921.735483.8178.398912.754668.829914.52805.7220937.5834.305429.148725.124744.05265182956.39031156.4612363425.29287188.89061752504999835.9137.4437.0826.5810.4430108.753.75841.56312.5363902076.24361246877.546.6447932535570545927295.961684333329.9810.511899.6440.07527.107669913334853.637.31141773.034614.128.227994966673646100035.45535.5348.5641813.7518265550000024.718.37011126466667154.5023210.13723666667161696666721.1514946.761836031835835361266667611656666727.653558600000712.697202500000019.1313.503689132.280615.4506.698331.7651476.217010.831.2432.0465.22455.261730.25133.6283340.69437.5713.745.00828.55603735.6040973960.2869.0531244026676.3916981719.5062.1816.550737.0454.8843724.0620.783180996.47147311.1411.2843872.0155.40389227.65918.29376751131.3013.26228.16386.9845261127.097.881129.72.36458128.2467.64152.74728.0364.412.5298074.48124.70843.070.71469956720.43815445.965955484.54996.172156.516473342.2815.074675.52.3253.044163333.211.4692100.73617010.0518.93110602.40820041878.3290.44875.7325062.07135861.9916904121465.20.5891699046140.53410.591.4378735.1361280.2692709751458.23455.79437.201036735.631594.4129.02316.902323.24569710976634530989681.7394616807115715.823.497.63187.507892689337.460236.734.08214.431109.6709.11885132.87328.698157.569.3826.9343.535522.9689889567506540744387683670223625738013238860547703441509169376657395792031517838153.9953.9223.411221.818.0488.9221092266723.3330.679999.184193.13823.175514.74165.0484.648147.392520.2263.308.06302.0641148.4024813.2647539654.14128.741242.69.80122.7188378.3462.781.10042908.1873.69649270.3292.15980462.855110.32105.98118.77905915.037666.516223.729330.8404703.474121684134752673.7973213.18392314.0573.85774259.18627.2139.745125.15910.4981279.792.16966460.775244.824119.803392302.2663.3083678.71153584.20389.070110046.4391.5180.2964.4751709423.592031441224.64284.6459522108.1242703.08827.101050798231252.3626943.429.406757.26319931.20298830.72511.175725.036.729515.8655.071159.66195648.829.018534.459529.956633.379748.36039.62403.204.01687248.68226.85592.2017.4213.734656.9247.895320.879550.830719.67735.1912317.9416.553460.409421.872924.661671712303.23780308.8012852284.26352234.54052929751794663.6163.6063.0639.3816.6126378.142.77543.04815.9662369998.35694031830.241.84642811633814644163614.281877433334.5910.601486.1246.31748.40641652300053961300012443000001299170000731406673088.734.79137265.931177.227.907837700004024400030.17477.5248.4250517.6240247550000016.064.63031152933333174.8323600.53503200000399698.78117876666732.088089.141057811058313806833333316720000016.643148233333799.717192776666731.9121.67324297.186110.4777.876319.7616936.819141.824.2324.8079.43739.311277.46106.3952659.88335.5023.726.70133.06539331.7049054059.7039.20181476000010.196706621.3571.586.6969238.9335.1038433.9730.398446819.89109416.4216.9691047.0735.56342431.34817.25869155601.5012.59532.23954.4547232012.7714.257129.53.39163760.4538.02112.66116.4583.314.1827073.3282.96633.470.84160845650.58580429.141078327.78295.784177.861386134.3014.456742.94.2502.971107321.891.8475370.90716111.3617.0765601.10132221809.7155.741102.1256038.9285613.1518667773640.91.149881881169.75378.052.2542269.2976286.7889231850438.60443.58423.302057021.661499.6150.33189.739171.37577822261037830843482.739277163977327.116.257.70157.3631242551956.400125.873.05145.413100.7389.92684144.25837.66086.9312.7426.8058.939416.9672176931314601314975299393774440991442704652076070768428438276013197605012058614787153.2953.2216.171165.817.8538.2251843860016.0053.9348127.567176.39541.911119.91106.8155.852676.960519.8940.838.88137.03802139.6171330.2033311086.85100.001189.79.67103.37616324.7577.761.32454754.8054.29531236.6482.81739354.754113.1597.911210.81585626.086638.424042.129554.1195626.138175857373780240.5674925.83954171.2313.98013251.30517.5744.599622.42510.46128566.812.80263356.764244.952129.956841327.8753.0508586.1164181.29348.091127133.8594.6641.8560.526898501.39182759234.24545.2133391191.8184795.46248.757624249121242.4848122.1613.159360.89143581.64919605.58911.6910934.176.9518375.6357.132237.69380867.528.848134.662831.532431.710942.97443.41736.235.10948195.45328.171128.7817.0214.678719.3770.437714.198064.789115.43415.1824676.8631.790731.458922.965570.63284194966.49693153.9222563744.65431214.81268957839280033.6141.4540.7731.907.4930547.652.68938.04913.2564966250.77798342114.640.3817970935734106742397.221820053325.8720.592039.6339.52425.02148497700066736500012045200001267480000704990005493.433.65072004.059139.226.888837012503895060030.47245.6943.69539914.6334262006666727.118.90911493833333162.9793586.34450525000381352.69202700000018.3116004.072067782054996284300000607730000030.373520766667799.491215463333317.2912.693345173.220818.5407.360291.3748686.820204.940.9241.5870.42657.131681.44199.7344993.34330.7612.765.11023.55893229.3251257437.8166.7836096463335.3912596717.3892.4038.081339.3153.6670826.7340.580664858.84228184.689.96452673.8154.95346630.96616.18369256111.5011.49331.61097.9383455312.097.402129.52.65998764.9365.38199.60896.8565.032.1594595.09153.16223.460.73844246930.30679558.253340550.01685.613165.495389455.4964.579755.52.1872.805220893.781.4323620.62425011.3116.92813273.69412438868.202530.5412.371173.0143375.32128125.9713937110666.90.5591792636117.53351.291.4522325.2423314.4482455664470.02465.06446.40697707.222861.0119.73428.387407.5894066801908827764790.0285744542103089.819.016.03220.569743966328.278320.111.47270.380139.0967.24187146.86863.428228.747.4623.3354.959918.2179641200848170507456728087161321058213161992612772265973541534443918518721145704847.3447.3219.471126.414.4734.804762633319.2016.0869100.089144.48115.502811.86212.4825.971306.324461.6151.056.30440.83113153.1246714.2926996959.58160.561131.17.76149.3785446.8383.571.37831728.4695.67700176.3332.97094337.004126.89121.19136.49240621.217947.270315.398416.05021200.7587954084384040.5238873.02897330.1044.58912217.88019.9947.241721.17020.33135075.074.57581219.532275.385132.132632380.9022.6254483.44117588.992123.832613011.29104.23106.9676.9152357375.731971988236.52202.598074270.2131755.33319.910268973370552.5019183.8311.374208.49428371.45280687.67112.543802.617.136717.6555.74860.64143755.333.334129.997335.066528.516439.10041.93285.955.22781191.11128.73415.9615.5916.332935.4761.825616.174261.254716.32515.279087.1224.553540.728923.122074.301141502453.13366319.0432861605.10914195.70539420534615486.6752.9751.5541.437.6721286.646.95954.37614.1011614763.70323720091.148.70828655220773831471612.801690633343.4210.551047.9643.36267.324671303332170.741.12521766.51852233.756323966673653766735.65750.9958.3791165.1342213876666713.363.1204800186667144.6983225.7267003333383853000042.855366.6967875678932848933333238180000015.172351900000711.101164026666742.7836.13337864.95977.5456.490333.736758.814173.4617.5218.1362.75048.91986.0463.6201590.50148.2831.668.30045.94743531.7818743050.1653.13158849666713.969237626.4581.085.3095329.0539.5623848.5400.725974841.6858815.3621.0211263.1044.05378526.24920.17576351081.4016.07526.86437.0051126171.5221.030129.73.78135116.5540.2375.46907.6222.265.9309572.2758.42303.104.3246567079.1943419.768930247.29796.836146.097521727.6294.916702.66.1603.73085910.952.6320771.2640710.1319.1177157.961177614665.01785214.416182.1233347.9380804.7365879071680.51.027976035145.29263.331.8867954.4054328.4177501248501.07499.18491.891395446.445531.0141.95246.936189.68611613797720327758994.848077046569221.415.846.23184.298962441746.705169.441.77175.625121.6708.39962156.13674.866125.1510.1722.3469.450114.483112775833827571011898993323320396720447323042229934499084949278863575341341994357045.2945.9516.681040.213.81498.5841295800016.3029.3496127.703137.50315.340516.91134.4197.300656.023435.2141.656.81201.10956134.7821316.9635895270.73114.291041.87.62123.88210975.8467.911.65270604.7557.04111141.9544.05270246.680116.49102.12243.23657824.651040.567115.977729.50181966.88138592543373323.9241995.74491174.6774.85818205.88517.9160.112016.63940.28174862.565.02267199.079276.457137.457437460.6732.1746378.2783665.75865.02666815.11111.7155.2669.6471237447.501481049256.63373.5045211127.8453196.12335.932438926166452.5337710.6017.155565.23197012.02659492.88513.347127.287.3213102.2856.711691.79288129.136.228327.607440.453424.727439.59442.16568.817.34848135.97629.98795.6314.9917.195576.8676.422713.086972.801813.73665.5517252.2730.473332.814124.003998.021991501086.00525166.4851932695.74921173.90950940553317344.7141.6842.3532.257.7730236.055.44242.87311.8972993136.60183219876.546.8715628754017074917299.641690466732.1430.541604.1134.82633.921670640003787.038.65342939.236900.527.517325600003649766736.19746.4951.24589810.2051249405000019.756.21261091460000122.6173207.03596400000144406666724.9910970.451338661345624825133333467096666722.253254033333710.916196323333323.1718.243643104.842513.3925.991341.514437.63916.7826.0026.6954.82346.131479.53108.3972709.93037.7616.556.33428.99311932.4819933816.7052.6331562616677.2934911720.7881.7213.450027.2644.7552102.6680.662387932.59135939.2513.2112554.9845.09392626.84218.47476351011.3914.40227.32572.3297246822.5210.840129.73.11065396.2251.91125.749412.6173.512.8810993.49107.62993.094.2839968217.7543139.345732441.65187.121122.089534247.5125.317651.93.1213.776172614.101.9829320.94507510.0118.81212410.2913569204.872245.2384.384145.2363406.96120494.3767812109388.20.6221607424125.14357.571.5021430.5872309.0383475645476.91474.68461.62813667.625441.5123.69384.063351.6441657359223027726089.5386354518112265.319.246.29215.496736650032.157290.561.54252.648155.7886.54965145.23864.883212.998.0023.7759.179516.9166747710702660592014115453187796174440188939807832637406900675184043676324407067448.1448.3519.291127.514.5632.146895700019.5818.485199.619143.48016.265212.50197.8445.925016.501464.7251.866.82410.29005150.6684813.8457222559.32146.031133.37.72143.8086122.3398.401.41662705.6065.91961168.8592.88734346.213120.76117.81134.38800923.662542.274716.321918.55331180.8599881764315467.1738843.37034296.7994.67269213.98518.0749.974020.01030.37140144.014.64757215.127273.515128.603449382.1092.6172181.72134048.964107.23759949.99103.5697.8073.0602123386.001941792257.50217.117336678.1051952.60622.362562540337902.6720910.5312.724386.48392381.47249678.48313.594093.737.767207.7462.53894.87162727.433.987729.425035.111028.479843.62959.69301.305.29027188.93038.50467.0815.6219.442877.3261.010716.389863.535315.73945.779686.3923.656542.271124.942243.841301582133.32030301.1362771815.26075190.06343223446993976.8751.5651.3542.228.1315081.047.08648.48714.3621855135.32609818621.247.95133407624334734236312.971688533340.5010.551108.9540.34758.159671276672514.039.87623250.220561.431.926890033333653900035.20148.5956.4016985.9702230896666713.893.6276882200000149.9963229.12972400000367199.0694455000039.176445.0979144791523222533333276906666715.682739633333712.013167970000038.4831.47339671.82538.2826.572335.587046.564438.5818.2418.8465.35950.091058.9970.8431771.06543.6728.497.95941.74619231.9982483076.6654.47159799333312.799758224.8401.146.0974731.4040.8225831.8380.617190852.3669776.4119.8824663.9945.32370727.08719.65076251021.3915.37627.75042.9834131606.8618.102129.73.70435234.0641.3483.49217.7482.385.1141612.3965.33183.104.2748166938.7510522.539990284.33366.605151.718518830.8434.867705.55.3323.55086407.462.2877661.1964310.1318.8476840.057185214463.41873.1201.382173.5504382.9776524.7407534271696.51.098920725138.39269.201.7677862.6357325.9778057728471.79467.96460.661627941.448534.8135.28231.033187.22400414940908227729895.398013956666816.914.965.91181.6291122332951.816156.701.54171.174121.7808.26016158.31068.860117.9110.4423.3069.201414.4583148466962705011811489573533727352152973756452205352291240334710338286011349790749046.7946.5914.741039.213.3719.2291460941714.9233.1275131.795135.43615.820716.64128.0662.563196.077439.8437.847.41186.32539159.3117116.6556004681.65110.561046.07.46122.70412115.0477.331.71875581.7697.53259132.7123.95819252.671119.03106.27239.72677329.444833.962115.374333.11252776.52153544003392842.3840696.17796161.8865.00222199.89614.8655.901917.98730.30176900.325.39123185.46276.229137.004266475.2542.1048080.3265663.82358.24866724.37107.3051.2968.4371143449.91143961278.52401.4541971133.9903349.75639.338936179153572.7940032.8118.725933.65181112.08780478.46714.427689.327.9714033.9462.111801.89316359.236.003927.777240.517824.681834.63464.91569.607.30822136.68941.57889.1814.5120.245525.5679.377012.597473.968413.51906.0218561.9134.250729.195525.724351.59227164976.51744153.4072073015.88628169.85055847375594640.3837.9138.3032.447.2611866.756.77643.91111.9673279262.53034921583.246.5716578894691315255146.511688400030.5300.541677.6233.27430.196662660004386.638.13645195.140368.627.637571766673652133335.70035.4649.13617511.8875249134000022.147.20751206066667124.2633222.83896500000151893333323.4512793.011559591566585036000000536703333324.653337066667712.085209046666721.4514.673537112.723114.1276.052308.957304.583836.7328.9629.6255.97945.791528.41114.9022872.56233.2115.296.26627.57729128.8401573819.8354.1932051610006.6638751020.0081.7416.090127.9046.2553269.4690.759434940.40141954.1512.1069058.1444.40392126.99418.13876350881.3914.09427.26083.3395257384.559.285129.72.96166492.4452.23129.748212.5783.652.5273633.66116.60533.094.1755367777.0604344.668302483.10497.038127.546532252.2445.311663.82.6913.738171436.471.7112960.86435810.0318.824OpenBenchmarking.org

WRF

WRF, the Weather Research and Forecasting Model, is a "next-generation mesoscale numerical weather prediction system designed for both atmospheric research and operational forecasting applications. It features two dynamical cores, a data assimilation system, and a software architecture supporting parallel computation and system extensibility." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWRF 4.2.2Input: conus 2.5kmXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3K6K9K12K15K13273.697157.9612410.296840.0611980.816674.1210602.415601.101. (F9X) gfortran options: -O2 -ftree-vectorize -funroll-loops -ffree-form -fconvert=big-endian -frecord-marker=4 -fallow-invalid-boz -lesmf_time -lwrfio_nf -lnetcdff -lnetcdf -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

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 2.0.0Benchmark: vklBenchmarkCPU ISPCXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P7001400210028003500SE +/- 6.66, N = 3SE +/- 8.62, N = 3SE +/- 13.96, N = 3SE +/- 4.36, N = 3SE +/- 1.53, N = 3SE +/- 6.74, N = 3SE +/- 2.08, N = 3SE +/- 13.62, N = 312431776135618521734271620043222MIN: 95 / MAX: 19286MIN: 177 / MAX: 27371MIN: 107 / MAX: 20909MIN: 191 / MAX: 28947MIN: 104 / MAX: 26188MIN: 188 / MAX: 30888MIN: 134 / MAX: 29750MIN: 248 / MAX: 36519

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - DegriddingXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2P3K6K9K12K15KSE +/- 150.41, N = 12SE +/- 577.93, N = 3SE +/- 228.53, N = 12SE +/- 295.26, N = 88868.2014665.009204.8714463.401. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

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 BuckyballXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50010001500200025002530.51785.02245.21873.12070.31960.91878.31809.71. (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 -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz -lcomex -m64 -ffast-math -std=legacy -fdefault-integer-8 -finline-functions -O2

RELION

RELION - REgularised LIkelihood OptimisatioN - is a stand-alone computer program for Maximum A Posteriori refinement of (multiple) 3D reconstructions or 2D class averages in cryo-electron microscopy (cryo-EM). It is developed in the research group of Sjors Scheres at the MRC Laboratory of Molecular Biology. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRELION 4.0.1Test: Basic - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P90180270360450SE +/- 2.98, N = 3SE +/- 3.24, N = 12SE +/- 4.65, N = 9SE +/- 2.66, N = 12SE +/- 7.14, N = 9SE +/- 2.65, N = 3SE +/- 2.68, N = 7SE +/- 1.41, N = 3412.37214.42384.38201.38373.18203.41290.45155.741. (CXX) g++ options: -fopenmp -std=c++11 -O3 -rdynamic -ldl -ltiff -lfftw3f -lfftw3 -lpng -ljpeg -lmpi_cxx -lmpi

easyWave

The easyWave software allows simulating tsunami generation and propagation in the context of early warning systems. EasyWave supports making use of OpenMP for CPU multi-threading and there are also GPU ports available but not currently incorporated as part of this test profile. The easyWave tsunami generation software is run with one of the example/reference input files for measuring the CPU execution time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BettereasyWave r34Input: e2Asean Grid + BengkuluSept2007 Source - Time: 2400Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 12.12, N = 12SE +/- 15.80, N = 12SE +/- 5.17, N = 9SE +/- 18.26, N = 12SE +/- 0.67, N = 3SE +/- 0.99, N = 12SE +/- 0.96, N = 15SE +/- 9.88, N = 12173.01182.12145.24173.55119.84126.3975.73102.131. (CXX) g++ options: -O3 -fopenmp

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - GriddingXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2K4K6K8K10KSE +/- 87.89, N = 12SE +/- 23.56, N = 3SE +/- 32.87, N = 12SE +/- 195.28, N = 8SE +/- 4.15, N = 3SE +/- 75.19, N = 5SE +/- 54.27, N = 5SE +/- 60.72, N = 33375.323347.933406.964382.978088.117630.065062.076038.921. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 95% Reads - Concurrency: 512Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30K60K90K120K150KSE +/- 2758.29, N = 15SE +/- 1830.36, N = 15SE +/- 2047.28, N = 12SE +/- 1792.91, N = 12SE +/- 3058.00, N = 15SE +/- 1606.68, N = 15SE +/- 2853.06, N = 12SE +/- 1875.64, N = 15128125.980804.7120494.376524.7131999.782341.2135861.985613.1

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 MetricXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1.1M2.2M3.3M4.4M5.5M71393736587907678124075342833892413696991690451866771. (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

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 60% Reads - Concurrency: 512Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30K60K90K120K150KSE +/- 1362.07, N = 3SE +/- 889.12, N = 15SE +/- 1384.66, N = 15SE +/- 970.10, N = 15SE +/- 1885.96, N = 15SE +/- 886.06, N = 15SE +/- 2440.36, N = 12SE +/- 1250.17, N = 13110666.971680.5109388.271696.5116603.772110.1121465.273640.9

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 LatencyXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.29090.58180.87271.16361.4545SE +/- 0.008, N = 12SE +/- 0.017, N = 9SE +/- 0.007, N = 3SE +/- 0.038, N = 9SE +/- 0.007, N = 3SE +/- 0.030, N = 12SE +/- 0.005, N = 12SE +/- 0.040, N = 120.5591.0270.6221.0980.5721.2930.5891.1491. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read OnlyXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P400K800K1200K1600K2000KSE +/- 24978.23, N = 12SE +/- 16726.04, N = 9SE +/- 17697.00, N = 3SE +/- 36627.33, N = 9SE +/- 20350.34, N = 3SE +/- 18635.21, N = 12SE +/- 13863.97, N = 12SE +/- 30933.90, N = 1217926369760351607424920725174814677814116990468818811. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

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: 512 - Model: ResNet-50Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 1.17, N = 3SE +/- 0.79, N = 3SE +/- 0.61, N = 3SE +/- 1.80, N = 3SE +/- 0.01, N = 3SE +/- 1.39, N = 3SE +/- 0.48, N = 3SE +/- 0.61, N = 3117.53145.29125.14138.39148.45174.95140.53169.75

CloverLeaf

CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeaf 1.3Input: clover_bm16Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P90180270360450SE +/- 1.43, N = 3SE +/- 1.56, N = 3SE +/- 2.03, N = 3SE +/- 2.11, N = 9SE +/- 0.13, N = 3SE +/- 0.74, N = 3SE +/- 0.54, N = 3SE +/- 2.02, N = 3351.29263.33357.57269.20328.47214.80410.59378.051. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

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: STMV with 1,066,628 AtomsXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.50721.01441.52162.02882.536SE +/- 0.05087, N = 15SE +/- 0.05644, N = 12SE +/- 0.05862, N = 15SE +/- 0.05463, N = 15SE +/- 0.01396, N = 15SE +/- 0.04403, N = 15SE +/- 0.02648, N = 15SE +/- 0.08185, N = 131.452231.886791.502141.767781.052741.109001.437872.25422

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 3.1X Y Z: 144 144 144 - RT: 60Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1530456075SE +/- 0.08, N = 3SE +/- 0.29, N = 3SE +/- 0.37, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.94, N = 325.2454.4130.5962.6431.0760.8535.1469.301. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

Apache IoTDB

Apache IotDB is a time series database and this benchmark is facilitated using the IoT Benchmaark [https://github.com/thulab/iot-benchmark/]. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P70140210280350SE +/- 5.15, N = 3SE +/- 2.59, N = 12SE +/- 5.07, N = 3SE +/- 3.74, N = 12SE +/- 2.20, N = 3SE +/- 3.79, N = 8SE +/- 3.41, N = 3SE +/- 3.40, N = 12314.44328.41309.03325.97294.33317.36280.26286.78MAX: 27244.84MAX: 29309.84MAX: 26720.06MAX: 30042.47MAX: 27740.38MAX: 32341.07MAX: 27450.82MAX: 28496.67

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20M40M60M80M100MSE +/- 662950.65, N = 3SE +/- 565297.90, N = 12SE +/- 1037167.36, N = 3SE +/- 574699.97, N = 12SE +/- 308756.34, N = 3SE +/- 692701.79, N = 8SE +/- 746822.47, N = 3SE +/- 761276.42, N = 128245566477501248834756457805772887797206801696209270975189231850

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ / https://github.com/ClickHouse/ClickBench/tree/main/clickhouse with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all separate queries performed as an aggregate. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third RunXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P110220330440550SE +/- 6.03, N = 3SE +/- 3.08, N = 3SE +/- 3.60, N = 3SE +/- 3.54, N = 12SE +/- 3.45, N = 3SE +/- 2.55, N = 3SE +/- 3.62, N = 6SE +/- 5.39, N = 5470.02501.07476.91471.79481.66501.63458.23438.60MIN: 32.59 / MAX: 5454.55MIN: 54.25 / MAX: 5000MIN: 36.01 / MAX: 5000MIN: 53.43 / MAX: 6000MIN: 40.16 / MAX: 5454.55MIN: 64.45 / MAX: 6666.67MIN: 42.16 / MAX: 6000MIN: 58.2 / MAX: 6000

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Second RunXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P110220330440550SE +/- 1.94, N = 3SE +/- 0.19, N = 3SE +/- 4.95, N = 3SE +/- 3.17, N = 12SE +/- 2.93, N = 3SE +/- 3.26, N = 3SE +/- 2.86, N = 6SE +/- 3.10, N = 5465.06499.18474.68467.96486.05507.56455.79443.58MIN: 32.59 / MAX: 5000MIN: 54.2 / MAX: 4615.38MIN: 36.43 / MAX: 6000MIN: 52.36 / MAX: 6000MIN: 40.43 / MAX: 6666.67MIN: 66.45 / MAX: 5454.55MIN: 42.19 / MAX: 6666.67MIN: 62.43 / MAX: 6666.67

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, First Run / Cold CacheXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P110220330440550SE +/- 2.85, N = 3SE +/- 1.35, N = 3SE +/- 5.88, N = 3SE +/- 3.85, N = 12SE +/- 3.41, N = 3SE +/- 4.37, N = 3SE +/- 4.18, N = 6SE +/- 4.12, N = 5446.40491.89461.62460.66465.50485.63437.20423.30MIN: 32.64 / MAX: 4615.38MIN: 51.86 / MAX: 5454.55MIN: 35.59 / MAX: 5454.55MIN: 53.62 / MAX: 5454.55MIN: 40.93 / MAX: 5454.55MIN: 63.63 / MAX: 5000MIN: 41.7 / MAX: 5454.55MIN: 61.54 / MAX: 6000

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgverify/s, More Is BetterOpenSSL 3.3Algorithm: RSA4096Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P400K800K1200K1600K2000KSE +/- 165.80, N = 3SE +/- 355.69, N = 3SE +/- 200.66, N = 3SE +/- 277.84, N = 3SE +/- 92.46, N = 3SE +/- 862.96, N = 3SE +/- 92.62, N = 3SE +/- 6745.42, N = 3697707.21395446.4813667.61627941.4974189.91916162.91036735.62057021.61. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgsign/s, More Is BetterOpenSSL 3.3Algorithm: RSA4096Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P13K26K39K52K65KSE +/- 19.45, N = 3SE +/- 6.92, N = 3SE +/- 20.13, N = 3SE +/- 83.09, N = 3SE +/- 9.84, N = 3SE +/- 14.34, N = 3SE +/- 15.15, N = 3SE +/- 56.68, N = 322861.045531.025441.548534.827431.154601.131594.461499.61. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

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: 256 - Model: ResNet-50Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 0.14, N = 3SE +/- 1.40, N = 3SE +/- 0.63, N = 3SE +/- 1.04, N = 12SE +/- 0.03, N = 3SE +/- 1.57, N = 3SE +/- 0.58, N = 3SE +/- 0.21, N = 3119.73141.95123.69135.28143.67161.06129.02150.33

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: allmodconfigXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P90180270360450SE +/- 0.61, N = 3SE +/- 0.56, N = 3SE +/- 0.67, N = 3SE +/- 0.73, N = 3SE +/- 0.80, N = 3SE +/- 0.86, N = 3SE +/- 0.72, N = 3SE +/- 0.56, N = 3428.39246.94384.06231.03362.45218.57316.90189.74

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: X3D-benchmarking input.i3dXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P90180270360450SE +/- 1.25, N = 3SE +/- 0.04, N = 3SE +/- 0.14, N = 3SE +/- 0.01, N = 3SE +/- 3.88, N = 4SE +/- 0.05, N = 3SE +/- 2.51, N = 3SE +/- 0.37, N = 3407.59189.69351.64187.22357.01196.81323.25171.381. (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

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 BenchmarkXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50M100M150M200M250MSE +/- 1751369.24, N = 12SE +/- 4344961.90, N = 10SE +/- 629927.17, N = 3SE +/- 2925715.37, N = 11SE +/- 1304590.97, N = 3SE +/- 2148606.38, N = 4SE +/- 2966712.90, N = 9SE +/- 2482737.87, N = 36801908813797720373592230149409082968874791877617691097663452226103781. (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

SecureMark

SecureMark is an objective, standardized benchmarking framework for measuring the efficiency of cryptographic processing solutions developed by EEMBC. SecureMark-TLS is benchmarking Transport Layer Security performance with a focus on IoT/edge computing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmarks, More Is BetterSecureMark 1.0.4Benchmark: SecureMark-TLSXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P70K140K210K280K350KSE +/- 256.85, N = 3SE +/- 384.08, N = 3SE +/- 670.43, N = 3SE +/- 196.24, N = 3SE +/- 625.56, N = 3SE +/- 330.95, N = 3SE +/- 209.97, N = 3SE +/- 411.11, N = 32776472775892772602772982772052754723098963084341. (CC) gcc options: -pedantic -O3

Apache IoTDB

Apache IotDB is a time series database and this benchmark is facilitated using the IoT Benchmaark [https://github.com/thulab/iot-benchmark/]. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.2Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 100Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.48, N = 3SE +/- 1.01, N = 3SE +/- 0.87, N = 3SE +/- 1.11, N = 4SE +/- 0.15, N = 3SE +/- 1.18, N = 3SE +/- 0.68, N = 3SE +/- 0.80, N = 790.0294.8489.5395.3985.1692.2381.7382.73MAX: 23857.34MAX: 23842.98MAX: 23819.37MAX: 23846.38MAX: 23810.86MAX: 23842.9MAX: 24081.78MAX: 24148.34

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 100Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20M40M60M80M100MSE +/- 435900.66, N = 3SE +/- 594234.55, N = 3SE +/- 782176.98, N = 3SE +/- 872659.60, N = 4SE +/- 276414.35, N = 3SE +/- 928557.10, N = 3SE +/- 739343.51, N = 3SE +/- 862582.24, N = 78574454280770465863545188013956690871770836517859461680792771639

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 50% Reads - Concurrency: 512Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20K40K60K80K100KSE +/- 297.86, N = 3SE +/- 833.19, N = 3SE +/- 1352.20, N = 3SE +/- 556.62, N = 3SE +/- 1037.18, N = 15SE +/- 639.27, N = 15SE +/- 1659.04, N = 15SE +/- 325.76, N = 3103089.869221.4112265.366816.9108953.668524.7115715.877327.1

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: 64 - Model: ResNet-152Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P612182430SE +/- 0.19, N = 5SE +/- 0.21, N = 3SE +/- 0.17, N = 12SE +/- 0.05, N = 3SE +/- 0.18, N = 3SE +/- 0.17, N = 3SE +/- 0.34, N = 3SE +/- 0.15, N = 1219.0115.8419.2414.9620.1314.0323.4916.25MIN: 11.15 / MAX: 20.08MIN: 6.82 / MAX: 16.51MIN: 10.95 / MAX: 20.39MIN: 10.89 / MAX: 15.42MIN: 17.06 / MAX: 21.01MIN: 8.98 / MAX: 15.41MIN: 19.73 / MAX: 24.56MIN: 6.19 / MAX: 17.5

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 2.6Scene: Danish Mood - Acceleration: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.02, N = 3SE +/- 0.22, N = 15SE +/- 0.07, N = 15SE +/- 0.20, N = 15SE +/- 0.08, N = 3SE +/- 0.18, N = 12SE +/- 0.10, N = 12SE +/- 0.11, N = 156.036.236.295.916.857.007.637.70MIN: 2.52 / MAX: 7MIN: 2.27 / MAX: 9.18MIN: 2.44 / MAX: 7.56MIN: 2.31 / MAX: 9.24MIN: 3.05 / MAX: 8.02MIN: 2.74 / MAX: 9.4MIN: 3.26 / MAX: 9.01MIN: 3.44 / MAX: 9.79

Timed Gem5 Compilation

This test times how long it takes to compile Gem5. Gem5 is a simulator for computer system architecture research. Gem5 is widely used for computer architecture research within the industry, academia, and more. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Gem5 Compilation 23.0.1Time To CompileXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50100150200250SE +/- 2.59, N = 4SE +/- 1.72, N = 3SE +/- 2.11, N = 6SE +/- 1.01, N = 3SE +/- 0.92, N = 3SE +/- 0.58, N = 3SE +/- 2.17, N = 3SE +/- 0.89, N = 3220.57184.30215.50181.63210.93178.28187.51157.36

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 WritingXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3M6M9M12M15MSE +/- 83772.27, N = 13SE +/- 218479.19, N = 15SE +/- 26369.56, N = 3SE +/- 254193.37, N = 15SE +/- 61518.27, N = 10SE +/- 343737.14, N = 15SE +/- 28201.44, N = 3SE +/- 215352.04, N = 15743966396244177366500112233297841738123213008926893124255191. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

LAMMPS Molecular Dynamics Simulator

LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: 20k AtomsXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1326395265SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 328.2846.7132.1651.8234.8950.5237.4656.401. (CXX) g++ options: -O3 -lm -ldl

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: Barbershop - Compute: CPU-OnlyXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P70140210280350SE +/- 0.12, N = 3SE +/- 0.23, N = 3SE +/- 1.09, N = 3SE +/- 0.06, N = 3SE +/- 0.18, N = 3SE +/- 0.42, N = 3SE +/- 0.69, N = 3SE +/- 0.21, N = 3320.11169.44290.56156.70261.97141.54236.73125.87

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: wizardcoder-python-34b-v1.0.Q6_K - Acceleration: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1.31632.63263.94895.26526.5815SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.12, N = 6SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 31.471.771.541.545.853.824.083.05

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 CompileXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P60120180240300SE +/- 0.53, N = 3SE +/- 0.84, N = 3SE +/- 0.74, N = 3SE +/- 0.52, N = 3SE +/- 0.71, N = 3SE +/- 0.48, N = 3SE +/- 0.52, N = 3SE +/- 0.12, N = 3270.38175.63252.65171.17254.00172.99214.43145.41

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: fcn-resnet101-11 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150SE +/- 3.36, N = 15SE +/- 5.82, N = 12SE +/- 6.41, N = 15SE +/- 2.59, N = 15SE +/- 0.09, N = 3SE +/- 0.69, N = 15SE +/- 0.71, N = 3SE +/- 0.44, N = 3139.10121.67155.79121.78116.3292.35109.67100.741. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: fcn-resnet101-11 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3691215SE +/- 0.15651, N = 15SE +/- 0.34632, N = 12SE +/- 0.22869, N = 15SE +/- 0.16487, N = 15SE +/- 0.00664, N = 3SE +/- 0.07619, N = 15SE +/- 0.05824, N = 3SE +/- 0.04271, N = 37.241878.399626.549658.260168.5971910.836239.118859.926841. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

DuckDB

DuckDB is an in-progress SQL OLAP database management system optimized for analytics and features a vectorized and parallel engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDuckDB 0.9.1Benchmark: TPC-H ParquetXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150SE +/- 0.15, N = 3SE +/- 0.57, N = 3SE +/- 0.58, N = 3SE +/- 0.95, N = 3SE +/- 0.12, N = 3SE +/- 0.55, N = 3SE +/- 0.34, N = 3SE +/- 0.44, N = 3146.87156.14145.24158.31145.57155.73132.87144.261. (CXX) g++ options: -O3 -rdynamic -lssl -lcrypto -ldl

easyWave

The easyWave software allows simulating tsunami generation and propagation in the context of early warning systems. EasyWave supports making use of OpenMP for CPU multi-threading and there are also GPU ports available but not currently incorporated as part of this test profile. The easyWave tsunami generation software is run with one of the example/reference input files for measuring the CPU execution time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BettereasyWave r34Input: e2Asean Grid + BengkuluSept2007 Source - Time: 1200Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 3.12, N = 12SE +/- 4.11, N = 15SE +/- 3.75, N = 15SE +/- 3.75, N = 15SE +/- 0.10, N = 3SE +/- 0.24, N = 3SE +/- 0.32, N = 4SE +/- 0.23, N = 363.4374.8764.8868.8646.2950.8328.7037.661. (CXX) g++ options: -O3 -fopenmp

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: Chrysler Neon 1MXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50100150200250SE +/- 0.08, N = 3SE +/- 0.13, N = 3SE +/- 0.18, N = 3SE +/- 0.14, N = 3SE +/- 0.15, N = 3SE +/- 0.11, N = 3SE +/- 0.20, N = 3SE +/- 0.14, N = 3228.74125.15212.99117.91207.20115.89157.5686.93

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 2.6Scene: DLSC - Acceleration: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3691215SE +/- 0.08, N = 3SE +/- 0.12, N = 3SE +/- 0.10, N = 3SE +/- 0.20, N = 15SE +/- 0.06, N = 15SE +/- 0.23, N = 15SE +/- 0.02, N = 3SE +/- 0.24, N = 157.4610.178.0010.448.7211.129.3812.74MIN: 7.17 / MAX: 8.21MIN: 9.34 / MAX: 14.75MIN: 7.65 / MAX: 9.12MIN: 8.86 / MAX: 16.08MIN: 8.12 / MAX: 10.6MIN: 9.18 / MAX: 17.51MIN: 8.94 / MAX: 11.49MIN: 10.69 / MAX: 19.19

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: UploadXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P612182430SE +/- 0.00, N = 3SE +/- 0.12, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.04, N = 3SE +/- 0.10, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 323.3322.3423.7723.3024.9724.5626.9326.801. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: yolov4 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1530456075SE +/- 0.55, N = 15SE +/- 1.43, N = 15SE +/- 0.55, N = 15SE +/- 0.47, N = 14SE +/- 0.12, N = 3SE +/- 0.28, N = 3SE +/- 0.10, N = 3SE +/- 0.35, N = 354.9669.4559.1869.2050.3564.2943.5458.941. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: yolov4 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P612182430SE +/- 0.17, N = 15SE +/- 0.29, N = 15SE +/- 0.15, N = 15SE +/- 0.10, N = 14SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.05, N = 3SE +/- 0.10, N = 318.2214.4816.9214.4619.8615.5622.9716.971. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: AES-128-GCMXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P400000M800000M1200000M1600000M2000000MSE +/- 43545680.66, N = 3SE +/- 268465891.11, N = 3SE +/- 56355870.99, N = 3SE +/- 697208281.49, N = 3SE +/- 28627701.54, N = 3SE +/- 159598753.20, N = 3SE +/- 1253503549.61, N = 3SE +/- 1622921637.03, N = 364120084817012775833827577477107026601484669627050768804693697153302630072788956750654017693131460131. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: AES-256-GCMXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P300000M600000M900000M1200000M1500000MSE +/- 119857064.56, N = 3SE +/- 108288170.03, N = 3SE +/- 63026881.96, N = 3SE +/- 112278705.55, N = 3SE +/- 2848140790.67, N = 3SE +/- 1397764228.85, N = 3SE +/- 3966858436.56, N = 3SE +/- 1320809584.03, N = 350745672808710118989933235920141154531181148957353700239352717136953432847074438768367014975299393771. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: ChaCha20-Poly1305Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P100000M200000M300000M400000M500000MSE +/- 6370607.14, N = 3SE +/- 4148679.28, N = 3SE +/- 37795783.50, N = 3SE +/- 36941390.18, N = 3SE +/- 766713911.50, N = 3SE +/- 132227134.30, N = 3SE +/- 120899798.30, N = 3SE +/- 82161991.11, N = 31613210582133203967204471877961744403727352152972250388507304422025329772236257380134440991442701. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: SHA512Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P10000M20000M30000M40000M50000MSE +/- 500176.69, N = 3SE +/- 3500318.33, N = 3SE +/- 1250834.62, N = 3SE +/- 14339684.98, N = 3SE +/- 203213936.42, N = 3SE +/- 481040738.21, N = 3SE +/- 37654528.63, N = 3SE +/- 14734422.48, N = 316199261277323042229931889398078337564522053222199277234284625447723886054770465207607071. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: ChaCha20Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P150000M300000M450000M600000M750000MSE +/- 415487944.57, N = 3SE +/- 329510485.23, N = 3SE +/- 312824840.64, N = 3SE +/- 863978337.15, N = 3SE +/- 232288636.76, N = 3SE +/- 271318554.29, N = 3SE +/- 3335467.26, N = 3SE +/- 303075210.03, N = 32265973541534499084949272637406900675229124033473161264211736244029729473441509169376842843827601. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: SHA256Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30000M60000M90000M120000M150000MSE +/- 27466135.54, N = 3SE +/- 126029257.64, N = 3SE +/- 33334799.97, N = 3SE +/- 70647143.49, N = 3SE +/- 159515825.53, N = 3SE +/- 66583540.43, N = 3SE +/- 23323764.84, N = 3SE +/- 29142783.86, N = 344439185187886357534135184043676310338286011361806994523119106802107665739579201319760501201. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

Speedb

Speedb is a next-generation key value storage engine that is RocksDB compatible and aiming for stability, efficiency, and performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Random ReadXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P130M260M390M520M650MSE +/- 3698537.65, N = 15SE +/- 8491098.40, N = 15SE +/- 2763560.64, N = 13SE +/- 2955608.21, N = 3SE +/- 168680.90, N = 3SE +/- 5393159.45, N = 3SE +/- 242146.75, N = 3SE +/- 5399265.88, N = 152114570484199435702440706744979074902761796175512869583151783815861478711. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

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: Video On DemandXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1224364860SE +/- 0.10, N = 3SE +/- 0.43, N = 3SE +/- 0.07, N = 3SE +/- 0.13, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.14, N = 347.3445.2948.1446.7949.8448.7753.9953.291. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 7.0Encoder: libx265 - Scenario: PlatformXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1224364860SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.08, N = 3SE +/- 0.27, N = 3SE +/- 0.16, N = 3SE +/- 0.12, N = 3SE +/- 0.14, N = 3SE +/- 0.15, N = 347.3245.9548.3546.5949.7948.7653.9253.221. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

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: 512 - Model: ResNet-152Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P612182430SE +/- 0.18, N = 7SE +/- 0.09, N = 3SE +/- 0.18, N = 7SE +/- 0.16, N = 3SE +/- 0.04, N = 3SE +/- 0.09, N = 3SE +/- 0.16, N = 3SE +/- 0.18, N = 319.4716.6819.2914.7420.4614.5923.4116.17MIN: 11.79 / MAX: 20.53MIN: 7.02 / MAX: 17.07MIN: 10.66 / MAX: 20.07MIN: 10.63 / MAX: 15.36MIN: 17.25 / MAX: 21.16MIN: 10.11 / MAX: 16.06MIN: 22.41 / MAX: 24.3MIN: 11.54 / MAX: 16.69

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19 - Decompression SpeedXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30060090012001500SE +/- 2.12, N = 3SE +/- 0.66, N = 15SE +/- 1.08, N = 15SE +/- 0.65, N = 15SE +/- 4.45, N = 3SE +/- 4.96, N = 3SE +/- 3.01, N = 4SE +/- 1.88, N = 31126.41040.21127.51039.21032.0963.61221.81165.81. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19 - Compression SpeedXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P48121620SE +/- 0.15, N = 3SE +/- 0.09, N = 15SE +/- 0.11, N = 15SE +/- 0.13, N = 15SE +/- 0.06, N = 3SE +/- 0.12, N = 3SE +/- 0.21, N = 4SE +/- 0.03, N = 314.413.814.513.314.615.218.017.81. (CC) gcc options: -O3 -pthread -lz -llzma

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: Recurrent Neural Network Inference - Engine: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30060090012001500SE +/- 21.06, N = 12SE +/- 113.47, N = 12SE +/- 20.21, N = 15SE +/- 7.22, N = 3SE +/- 0.29, N = 3SE +/- 2.30, N = 3SE +/- 2.24, N = 3SE +/- 3.17, N = 3734.801498.58632.15719.23409.59633.29488.92538.23MIN: 546.95MIN: 641.2MIN: 459.4MIN: 696.95MIN: 390.22MIN: 621.42MIN: 457.65MIN: 529.581. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: MD5Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4M8M12M16M20MSE +/- 54649.77, N = 15SE +/- 601801.63, N = 12SE +/- 5000.00, N = 3SE +/- 695042.48, N = 12SE +/- 3511.88, N = 3SE +/- 22338.31, N = 3SE +/- 1201.85, N = 3SE +/- 317498.76, N = 1576263331295800089570001460941795830001816600010922667184386001. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

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: 256 - Model: ResNet-152Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P612182430SE +/- 0.15, N = 3SE +/- 0.09, N = 3SE +/- 0.15, N = 3SE +/- 0.18, N = 4SE +/- 0.08, N = 3SE +/- 0.13, N = 7SE +/- 0.29, N = 3SE +/- 0.16, N = 319.2016.3019.5814.9220.2514.0823.3316.00MIN: 10.93 / MAX: 20MIN: 11.86 / MAX: 16.95MIN: 13.13 / MAX: 20.27MIN: 8.28 / MAX: 15.85MIN: 17.03 / MAX: 21.21MIN: 6.98 / MAX: 15.46MIN: 19.78 / MAX: 24.54MIN: 11.43 / MAX: 16.66

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: particle_volume/scivis/real_timeXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1224364860SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 316.0929.3518.4933.1321.8938.2330.6853.93

DuckDB

DuckDB is an in-progress SQL OLAP database management system optimized for analytics and features a vectorized and parallel engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDuckDB 0.9.1Benchmark: IMDBXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150SE +/- 0.23, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.41, N = 3SE +/- 0.10, N = 3SE +/- 0.67, N = 3SE +/- 0.59, N = 3SE +/- 0.30, N = 3100.09127.7099.62131.8099.49125.8899.18127.571. (CXX) g++ options: -O3 -rdynamic -lssl -lcrypto -ldl

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: particle_volume/pathtracer/real_timeXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 0.09, N = 3SE +/- 0.42, N = 3SE +/- 0.29, N = 3SE +/- 0.81, N = 3SE +/- 0.17, N = 3SE +/- 0.49, N = 3SE +/- 0.25, N = 3SE +/- 0.89, N = 3144.48137.50143.48135.44160.24149.09193.14176.40

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.1Benchmark: gravity_spheres_volume/dim_512/scivis/real_timeXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1020304050SE +/- 0.22, N = 15SE +/- 0.24, N = 12SE +/- 0.18, N = 15SE +/- 0.22, N = 15SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.09, N = 315.5015.3416.2715.8219.5835.6023.1841.91

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 2.6Scene: Orange Juice - Acceleration: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P510152025SE +/- 0.10, N = 3SE +/- 0.16, N = 15SE +/- 0.04, N = 3SE +/- 0.28, N = 15SE +/- 0.07, N = 3SE +/- 0.12, N = 3SE +/- 0.05, N = 3SE +/- 0.30, N = 1511.8616.9112.5016.6413.4618.7014.7419.91MIN: 9.93 / MAX: 13.28MIN: 12.9 / MAX: 23.07MIN: 10.38 / MAX: 14.93MIN: 12.39 / MAX: 24.93MIN: 11.09 / MAX: 16.56MIN: 16.22 / MAX: 25.05MIN: 12.17 / MAX: 18.21MIN: 15 / MAX: 30.59

Timed LLVM Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: NinjaXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50100150200250SE +/- 0.19, N = 3SE +/- 0.17, N = 3SE +/- 0.36, N = 3SE +/- 0.79, N = 3SE +/- 0.07, N = 3SE +/- 0.15, N = 3SE +/- 0.12, N = 3SE +/- 0.27, N = 3212.48134.42197.84128.07187.02122.31165.05106.82

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 AtomsXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.16058, N = 15SE +/- 0.26203, N = 15SE +/- 0.14537, N = 15SE +/- 0.44989, N = 15SE +/- 0.21567, N = 12SE +/- 0.17435, N = 15SE +/- 0.54361, N = 15SE +/- 0.34017, N = 125.971307.300655.925012.563192.970212.219484.648145.85267

VVenC

VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.11Video Input: Bosphorus 4K - Video Preset: FastXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.010, N = 3SE +/- 0.081, N = 3SE +/- 0.033, N = 3SE +/- 0.043, N = 15SE +/- 0.022, N = 3SE +/- 0.047, N = 3SE +/- 0.023, N = 3SE +/- 0.038, N = 36.3246.0236.5016.0776.7196.4807.3926.9601. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects

Numpy Benchmark

This is a test to obtain the general Numpy performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P110220330440550SE +/- 0.60, N = 3SE +/- 1.13, N = 3SE +/- 0.79, N = 3SE +/- 1.30, N = 3SE +/- 0.33, N = 3SE +/- 0.26, N = 3SE +/- 1.84, N = 3SE +/- 1.42, N = 3461.61435.21464.72439.84455.57456.18520.22519.89

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: 512 - Model: ResNet-50Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1428425670SE +/- 0.72, N = 3SE +/- 0.26, N = 3SE +/- 0.43, N = 15SE +/- 0.34, N = 15SE +/- 0.12, N = 3SE +/- 0.21, N = 3SE +/- 0.19, N = 3SE +/- 0.30, N = 1551.0541.6551.8637.8456.4437.5563.3040.83MIN: 40.71 / MAX: 53.89MIN: 31.6 / MAX: 43.65MIN: 30.55 / MAX: 55.87MIN: 19.63 / MAX: 40.77MIN: 48.09 / MAX: 58.6MIN: 24.5 / MAX: 40.69MIN: 48.94 / MAX: 65.25MIN: 15.4 / MAX: 43.41

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 2.6Scene: LuxCore Benchmark - Acceleration: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.03, N = 3SE +/- 0.22, N = 12SE +/- 0.05, N = 3SE +/- 0.16, N = 15SE +/- 0.04, N = 3SE +/- 0.12, N = 3SE +/- 0.11, N = 3SE +/- 0.17, N = 126.306.816.827.417.108.828.068.88MIN: 2.53 / MAX: 7.3MIN: 2.55 / MAX: 9.19MIN: 2.81 / MAX: 7.94MIN: 2.67 / MAX: 9.42MIN: 3.06 / MAX: 8.2MIN: 3.72 / MAX: 10.41MIN: 3.62 / MAX: 9.37MIN: 3.73 / MAX: 11.48

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, Medium Mesh Size - Execution TimeXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P100200300400500440.83201.11410.29186.33420.65196.50302.06137.041. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Mesh TimeXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200153.12134.78150.67159.31160.73162.03148.40139.621. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm

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 Write - Average LatencyXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P714212835SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.12, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.10, N = 314.2916.9613.8516.6614.0414.8613.2630.201. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read WriteXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P16K32K48K64K80KSE +/- 95.84, N = 3SE +/- 195.77, N = 3SE +/- 143.25, N = 3SE +/- 440.63, N = 3SE +/- 78.02, N = 3SE +/- 137.82, N = 3SE +/- 198.41, N = 3SE +/- 106.16, N = 369969589527222560046712276731575396331101. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -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: Rubber O-Ring Seal InstallationXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.60, N = 3SE +/- 0.02, N = 3SE +/- 0.32, N = 3SE +/- 0.51, N = 15SE +/- 0.12, N = 3SE +/- 0.06, N = 3SE +/- 0.34, N = 3SE +/- 0.72, N = 859.5870.7359.3281.6559.9188.3954.1486.85

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: INIVOL and Fluid Structure Interaction Drop ContainerXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 0.08, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 0.19, N = 3SE +/- 0.17, N = 3SE +/- 0.12, N = 3SE +/- 0.30, N = 3160.56114.29146.03110.56140.43107.36128.74100.00

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19, Long Mode - Decompression SpeedXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30060090012001500SE +/- 0.56, N = 3SE +/- 1.00, N = 3SE +/- 1.45, N = 3SE +/- 1.78, N = 15SE +/- 0.43, N = 3SE +/- 3.69, N = 3SE +/- 0.84, N = 3SE +/- 1.60, N = 31131.11041.81133.31046.01069.1992.11242.61189.71. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19, Long Mode - Compression SpeedXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3691215SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 15SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 37.767.627.727.467.948.169.809.671. (CC) gcc options: -O3 -pthread -lz -llzma

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 CompileXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150SE +/- 0.18, N = 3SE +/- 0.56, N = 3SE +/- 0.32, N = 3SE +/- 1.69, N = 3SE +/- 0.76, N = 3SE +/- 0.56, N = 3SE +/- 0.91, N = 3SE +/- 0.29, N = 3149.38123.88143.81122.70140.23118.57122.72103.38

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 Xmrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.21Variant: GhostRider - Hash Count: 1MXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3K6K9K12K15KSE +/- 12.93, N = 3SE +/- 14.02, N = 3SE +/- 14.78, N = 3SE +/- 7.44, N = 3SE +/- 23.16, N = 3SE +/- 4.03, N = 3SE +/- 44.68, N = 3SE +/- 64.69, N = 35446.810975.86122.312115.05808.811263.28378.316324.71. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

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: 512 - Model: GoogLeNetXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P120240360480600SE +/- 0.41, N = 3SE +/- 1.14, N = 3SE +/- 0.37, N = 3SE +/- 2.46, N = 3SE +/- 0.34, N = 3SE +/- 2.62, N = 3SE +/- 3.68, N = 3SE +/- 1.52, N = 3383.57467.91398.40477.33461.23576.08462.78577.76

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: CaffeNet 12-int8 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.38670.77341.16011.54681.9335SE +/- 0.02719, N = 15SE +/- 0.00498, N = 3SE +/- 0.01456, N = 3SE +/- 0.01853, N = 5SE +/- 0.01594, N = 3SE +/- 0.01405, N = 15SE +/- 0.00291, N = 3SE +/- 0.01541, N = 31.378311.652701.416621.718751.172921.435731.100421.324541. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: CaffeNet 12-int8 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2004006008001000SE +/- 12.91, N = 15SE +/- 1.83, N = 3SE +/- 7.34, N = 3SE +/- 6.21, N = 5SE +/- 11.46, N = 3SE +/- 6.50, N = 15SE +/- 2.43, N = 3SE +/- 8.88, N = 3728.47604.76705.61581.77852.33696.98908.19754.811. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: GPT-2 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.06596, N = 15SE +/- 0.02902, N = 3SE +/- 0.06620, N = 3SE +/- 0.06105, N = 3SE +/- 0.03075, N = 3SE +/- 0.03218, N = 3SE +/- 0.01631, N = 3SE +/- 0.15083, N = 155.677007.041115.919617.532594.499765.551013.696494.295311. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: GPT-2 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P60120180240300SE +/- 2.01, N = 15SE +/- 0.58, N = 3SE +/- 1.88, N = 3SE +/- 1.08, N = 3SE +/- 1.50, N = 3SE +/- 1.05, N = 3SE +/- 1.19, N = 3SE +/- 8.14, N = 15176.33141.95168.86132.71222.03180.03270.33236.651. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: T5 Encoder - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.91321.82642.73963.65284.566SE +/- 0.03338, N = 15SE +/- 0.04072, N = 3SE +/- 0.01443, N = 3SE +/- 0.02633, N = 15SE +/- 0.00807, N = 3SE +/- 0.04801, N = 3SE +/- 0.01801, N = 3SE +/- 0.01381, N = 32.970944.052702.887343.958193.065004.058812.159802.817391. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: T5 Encoder - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P100200300400500SE +/- 3.48, N = 15SE +/- 2.44, N = 3SE +/- 1.74, N = 3SE +/- 1.67, N = 15SE +/- 0.86, N = 3SE +/- 2.94, N = 3SE +/- 3.84, N = 3SE +/- 1.73, N = 3337.00246.68346.21252.67326.14246.34462.86354.751. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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: Bird Strike on WindshieldXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150SE +/- 0.45, N = 3SE +/- 0.69, N = 3SE +/- 0.03, N = 3SE +/- 0.38, N = 3SE +/- 0.24, N = 3SE +/- 0.29, N = 3SE +/- 0.14, N = 3SE +/- 0.83, N = 3126.89116.49120.76119.03117.41124.44110.32113.15

QMCPACK

QMCPACK is a modern high-performance open-source Quantum Monte Carlo (QMC) simulation code making use of MPI for this benchmark of the H20 example code. QMCPACK is an open-source production level many-body ab initio Quantum Monte Carlo code for computing the electronic structure of atoms, molecules, and solids. QMCPACK is supported by the U.S. Department of Energy. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.17.1Input: Li2_STO_aeXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150SE +/- 1.20, N = 3SE +/- 0.88, N = 3SE +/- 1.04, N = 3SE +/- 1.11, N = 5SE +/- 0.89, N = 3SE +/- 1.11, N = 3SE +/- 1.03, N = 3SE +/- 0.93, N = 3121.19102.12117.81106.27109.30101.63105.9897.911. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -march=native -O3 -lm -ldl

Appleseed

Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: Material TesterXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50100150200250136.49243.24134.39239.73126.15225.35118.78210.82

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P714212835SE +/- 0.32, N = 15SE +/- 0.21, N = 3SE +/- 0.33, N = 3SE +/- 0.18, N = 3SE +/- 0.06, N = 3SE +/- 0.36, N = 3SE +/- 0.19, N = 3SE +/- 0.39, N = 1321.2224.6523.6629.4418.9327.5915.0426.091. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1530456075SE +/- 0.69, N = 15SE +/- 0.34, N = 3SE +/- 0.59, N = 3SE +/- 0.21, N = 3SE +/- 0.16, N = 3SE +/- 0.48, N = 3SE +/- 0.82, N = 3SE +/- 0.51, N = 1347.2740.5742.2733.9652.8236.2566.5238.421. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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_timeXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1020304050SE +/- 0.17, N = 3SE +/- 0.22, N = 3SE +/- 0.18, N = 15SE +/- 0.32, N = 15SE +/- 0.10, N = 3SE +/- 0.11, N = 3SE +/- 0.05, N = 3SE +/- 0.36, N = 315.4015.9816.3215.3719.8536.1523.7342.13

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 3.1Benchmark: particle_volume/ao/real_timeXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1224364860SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.07, N = 316.0529.5018.5533.1121.9338.2230.8454.12

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: Recurrent Neural Network Training - Engine: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P6001200180024003000SE +/- 12.69, N = 3SE +/- 119.20, N = 12SE +/- 8.14, N = 3SE +/- 31.34, N = 3SE +/- 0.67, N = 3SE +/- 2.79, N = 3SE +/- 5.57, N = 3SE +/- 4.22, N = 31200.751966.881180.852776.52623.60635.15703.47626.14MIN: 1127.94MIN: 1146.52MIN: 1153.29MIN: 2715.64MIN: 598.72MIN: 612.85MIN: 649.83MIN: 600.371. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

Speedb

Speedb is a next-generation key value storage engine that is RocksDB compatible and aiming for stability, efficiency, and performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Read While WritingXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4M8M12M16M20MSE +/- 88257.34, N = 3SE +/- 171807.74, N = 3SE +/- 93731.24, N = 7SE +/- 197596.27, N = 3SE +/- 62510.72, N = 3SE +/- 239363.92, N = 3SE +/- 143570.67, N = 4SE +/- 195912.00, N = 15879540813859254998817615354400106953601729702712168413175857371. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Memcached

Memcached is a high performance, distributed memory object caching system. This Memcached test profiles makes use of memtier_benchmark for excuting this CPU/memory-focused server benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:100Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1000K2000K3000K4000K5000KSE +/- 21428.06, N = 3SE +/- 26891.33, N = 3SE +/- 52575.97, N = 3SE +/- 24083.78, N = 3SE +/- 25335.31, N = 3SE +/- 50660.01, N = 15SE +/- 35182.67, N = 3SE +/- 35300.21, N = 34384040.523373323.924315467.173392842.384481667.953245908.404752673.793780240.561. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance of various popular real-world Java workloads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: TradesoapXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P16003200480064008000SE +/- 36.36, N = 15SE +/- 48.92, N = 14SE +/- 40.72, N = 15SE +/- 48.60, N = 15SE +/- 29.41, N = 3SE +/- 31.29, N = 3SE +/- 57.43, N = 15SE +/- 52.55, N = 1538874199388440694115421573217492

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: ResNet50 v1-12-int8 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.02187, N = 3SE +/- 0.09530, N = 15SE +/- 0.03815, N = 5SE +/- 0.06816, N = 3SE +/- 0.01810, N = 3SE +/- 0.03225, N = 3SE +/- 0.02572, N = 3SE +/- 0.01358, N = 33.028975.744913.370346.177963.170025.625173.183925.839541. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: ResNet50 v1-12-int8 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P70140210280350SE +/- 2.40, N = 3SE +/- 2.72, N = 15SE +/- 3.31, N = 5SE +/- 1.78, N = 3SE +/- 1.80, N = 3SE +/- 1.01, N = 3SE +/- 2.52, N = 3SE +/- 0.40, N = 3330.10174.68296.80161.89315.41177.76314.06171.231. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: super-resolution-10 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1.12552.2513.37654.5025.6275SE +/- 0.00187, N = 3SE +/- 0.05347, N = 4SE +/- 0.00532, N = 3SE +/- 0.02558, N = 3SE +/- 0.00313, N = 3SE +/- 0.10855, N = 15SE +/- 0.00044, N = 3SE +/- 0.05342, N = 34.589124.858184.672695.002224.396494.725303.857743.980131. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: super-resolution-10 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P60120180240300SE +/- 0.09, N = 3SE +/- 2.30, N = 4SE +/- 0.25, N = 3SE +/- 1.02, N = 3SE +/- 0.16, N = 3SE +/- 4.15, N = 15SE +/- 0.03, N = 3SE +/- 3.39, N = 3217.88205.89213.99199.90227.42212.92259.19251.311. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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-152Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P612182430SE +/- 0.26, N = 3SE +/- 0.18, N = 3SE +/- 0.16, N = 3SE +/- 0.13, N = 8SE +/- 0.09, N = 3SE +/- 0.16, N = 4SE +/- 0.08, N = 3SE +/- 0.19, N = 519.9917.9118.0714.8619.4115.7727.2117.57MIN: 0.34 / MAX: 23.05MIN: 5.47 / MAX: 18.9MIN: 0.27 / MAX: 21.93MIN: 0.53 / MAX: 17.18MIN: 14.68 / MAX: 22.3MIN: 8.33 / MAX: 16.91MIN: 13.91 / MAX: 28.31MIN: 5.14 / MAX: 18.58

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: bertsquad-12 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1428425670SE +/- 0.44, N = 3SE +/- 0.68, N = 3SE +/- 0.23, N = 3SE +/- 1.12, N = 15SE +/- 0.17, N = 3SE +/- 0.62, N = 3SE +/- 0.04, N = 3SE +/- 0.44, N = 347.2460.1149.9755.9049.6460.3639.7544.601. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: bertsquad-12 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P612182430SE +/- 0.20, N = 3SE +/- 0.19, N = 3SE +/- 0.09, N = 3SE +/- 0.35, N = 15SE +/- 0.07, N = 3SE +/- 0.17, N = 3SE +/- 0.02, N = 3SE +/- 0.22, N = 321.1716.6420.0117.9920.1516.5725.1622.431. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.11030.22060.33090.44120.5515SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 15SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 30.330.280.370.300.430.290.490.46MIN: 0.26 / MAX: 19.2MIN: 0.25 / MAX: 41.4MIN: 0.28 / MAX: 40.85MIN: 0.26 / MAX: 42.1MIN: 0.23 / MAX: 17.03MIN: 0.24 / MAX: 39.69MIN: 0.18 / MAX: 16.07MIN: 0.18 / MAX: 22.291. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P40K80K120K160K200KSE +/- 383.22, N = 3SE +/- 18.38, N = 3SE +/- 1761.33, N = 3SE +/- 1422.88, N = 15SE +/- 75.30, N = 3SE +/- 2494.14, N = 3SE +/- 988.12, N = 3SE +/- 1158.60, N = 3135075.07174862.56140144.01176900.32115106.30189893.3181279.79128566.811. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: T5 Encoder - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1.2132.4263.6394.8526.065SE +/- 0.08827, N = 15SE +/- 0.03283, N = 3SE +/- 0.01103, N = 3SE +/- 0.02608, N = 3SE +/- 0.00659, N = 3SE +/- 0.00316, N = 3SE +/- 0.00472, N = 3SE +/- 0.02371, N = 34.575815.022674.647575.391232.796943.465712.169662.802631. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: T5 Encoder - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P100200300400500SE +/- 3.95, N = 15SE +/- 1.31, N = 3SE +/- 0.51, N = 3SE +/- 0.89, N = 3SE +/- 0.84, N = 3SE +/- 0.26, N = 3SE +/- 1.00, N = 3SE +/- 3.01, N = 3219.53199.08215.13185.46357.36288.47460.78356.761. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

Timed CPython Compilation

This test times how long it takes to build the reference Python implementation, CPython, with optimizations and LTO enabled for a release build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed CPython Compilation 3.10.6Build Configuration: Released Build, PGO + LTO OptimizedXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P60120180240300275.39276.46273.52276.23277.39279.69244.82244.95

Appleseed

Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: EmilyXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150132.13137.46128.60137.00120.75130.62119.80129.96

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: fcn-resnet101-11 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P100200300400500SE +/- 1.73, N = 3SE +/- 5.57, N = 13SE +/- 2.21, N = 3SE +/- 6.06, N = 3SE +/- 0.94, N = 3SE +/- 0.96, N = 3SE +/- 0.74, N = 3SE +/- 4.07, N = 3380.90460.67382.11475.25329.04352.74302.27327.881. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: fcn-resnet101-11 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.74441.48882.23322.97763.722SE +/- 0.01189, N = 3SE +/- 0.02702, N = 13SE +/- 0.01517, N = 3SE +/- 0.02652, N = 3SE +/- 0.00871, N = 3SE +/- 0.00768, N = 3SE +/- 0.00805, N = 3SE +/- 0.03826, N = 32.625442.174632.617212.104803.039222.834983.308363.050851. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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 BeamXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.03, N = 3SE +/- 0.23, N = 3SE +/- 0.09, N = 3SE +/- 0.22, N = 3SE +/- 0.04, N = 3SE +/- 0.27, N = 3SE +/- 0.09, N = 3SE +/- 0.22, N = 383.4478.2781.7280.3279.5481.3078.7186.11

Aircrack-ng

Aircrack-ng is a tool for assessing WiFi/WLAN network security. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgk/s, More Is BetterAircrack-ng 1.7Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P30K60K90K120K150KSE +/- 36.42, N = 3SE +/- 1027.07, N = 15SE +/- 122.31, N = 3SE +/- 1131.21, N = 15SE +/- 1953.08, N = 3SE +/- 1197.41, N = 15SE +/- 840.59, N = 3SE +/- 773.70, N = 3117588.9983665.76134048.9665663.82147469.3353914.93153584.2064181.291. (CXX) g++ options: -std=gnu++17 -O3 -fvisibility=hidden -fcommon -rdynamic -lnl-3 -lnl-genl-3 -lpcre -lsqlite3 -lpthread -lz -lssl -lcrypto -lhwloc -ldl -lm -pthread

Helsing

Helsing is an open-source POSIX vampire number generator. This test profile measures the time it takes to generate vampire numbers between varying numbers of digits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHelsing 1.0-betaDigit Range: 14 digitXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150SE +/- 0.53, N = 3SE +/- 0.11, N = 3SE +/- 0.12, N = 3SE +/- 0.42, N = 3SE +/- 0.26, N = 3SE +/- 0.06, N = 3SE +/- 0.08, N = 3SE +/- 0.07, N = 3123.8365.03107.2458.2593.3550.8389.0748.091. (CC) gcc options: -O2 -pthread

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance of various popular real-world Java workloads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: TradebeansXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3K6K9K12K15KSE +/- 51.68, N = 15SE +/- 66.27, N = 3SE +/- 54.76, N = 15SE +/- 10.41, N = 3SE +/- 79.99, N = 3SE +/- 64.94, N = 3SE +/- 226.65, N = 15SE +/- 16.01, N = 36130668159946672727080501100412713

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: ResNet-50Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3691215SE +/- 0.09, N = 4SE +/- 0.07, N = 3SE +/- 0.10, N = 4SE +/- 0.04, N = 15SE +/- 0.05, N = 4SE +/- 0.06, N = 15SE +/- 0.05, N = 9SE +/- 0.03, N = 1511.295.119.994.3710.534.396.433.85

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 64 - Model: ResNet-50Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150SE +/- 0.13, N = 3SE +/- 0.39, N = 3SE +/- 0.93, N = 3SE +/- 1.28, N = 3SE +/- 0.04, N = 3SE +/- 1.12, N = 3SE +/- 0.50, N = 3SE +/- 0.70, N = 3104.23111.71103.56107.30120.93115.4791.5194.66

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: Pabellon Barcelona - Compute: CPU-OnlyXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.31, N = 3SE +/- 0.08, N = 3SE +/- 0.44, N = 3SE +/- 0.20, N = 3SE +/- 0.15, N = 3SE +/- 0.26, N = 3SE +/- 0.09, N = 3SE +/- 0.06, N = 3106.9655.2697.8051.2987.9246.3880.2941.85

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: 0Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.15, N = 3SE +/- 0.29, N = 3SE +/- 0.22, N = 3SE +/- 0.32, N = 3SE +/- 0.32, N = 3SE +/- 0.05, N = 3SE +/- 0.08, N = 3SE +/- 0.03, N = 376.9269.6573.0668.4469.9367.3864.4860.531. (CXX) g++ options: -O3 -fPIC -lm

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P5001000150020002500SE +/- 1.20, N = 3SE +/- 2.65, N = 3SE +/- 1.20, N = 3SE +/- 0.33, N = 3SE +/- 0.88, N = 3SE +/- 0.88, N = 3SE +/- 1.45, N = 3SE +/- 0.00, N = 32357123721231143197710631709898

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: 256 - Model: GoogLeNetXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P110220330440550SE +/- 1.67, N = 3SE +/- 2.29, N = 3SE +/- 1.91, N = 3SE +/- 4.80, N = 3SE +/- 0.24, N = 3SE +/- 4.87, N = 3SE +/- 2.30, N = 3SE +/- 2.68, N = 3375.73447.50386.00449.91448.46528.93423.59501.39

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: RotateXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 1.93, N = 6SE +/- 0.33, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 31971481941431961722031821. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P400800120016002000SE +/- 0.33, N = 3SE +/- 1.76, N = 3SE +/- 0.58, N = 3SE +/- 0.58, N = 3SE +/- 0.58, N = 3SE +/- 0.58, N = 3SE +/- 0.88, N = 3SE +/- 0.88, N = 319881049179296116578941441759

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: Face Detection FP16-INT8 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P80160240320400SE +/- 0.15, N = 3SE +/- 0.40, N = 3SE +/- 0.47, N = 3SE +/- 0.35, N = 3SE +/- 0.28, N = 3SE +/- 0.00, N = 3SE +/- 0.69, N = 3SE +/- 0.33, N = 3236.52256.63257.50278.52356.38283.11224.64234.24MIN: 172.47 / MAX: 265.77MIN: 191.95 / MAX: 301.21MIN: 187.62 / MAX: 295.06MIN: 196.58 / MAX: 401.59MIN: 109.11 / MAX: 395.69MIN: 203.37 / MAX: 332.08MIN: 153.29 / MAX: 553.64MIN: 148.68 / MAX: 448.541. (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: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P120240360480600SE +/- 0.14, N = 3SE +/- 0.58, N = 3SE +/- 0.41, N = 3SE +/- 0.54, N = 3SE +/- 0.12, N = 3SE +/- 0.01, N = 3SE +/- 0.69, N = 3SE +/- 0.77, N = 3202.59373.50217.11401.45167.83423.18284.64545.211. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20K40K60K80K100KSE +/- 41.33, N = 3SE +/- 155.80, N = 3SE +/- 65.97, N = 3SE +/- 141.04, N = 3SE +/- 58.68, N = 3SE +/- 125.33, N = 3SE +/- 162.52, N = 3SE +/- 47.16, N = 38074245211733664197168006388135952233391

miniBUDE

MiniBUDE is a mini application for the the core computation of the Bristol University Docking Engine (BUDE). This test profile currently makes use of the OpenMP implementation of miniBUDE for CPU benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 0.31, N = 3SE +/- 1.48, N = 3SE +/- 0.96, N = 4SE +/- 0.93, N = 3SE +/- 0.31, N = 3SE +/- 0.99, N = 3SE +/- 0.03, N = 3SE +/- 1.93, N = 1570.21127.8578.11133.9986.97140.83108.12191.821. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P10002000300040005000SE +/- 7.81, N = 3SE +/- 36.96, N = 3SE +/- 23.93, N = 4SE +/- 23.14, N = 3SE +/- 7.77, N = 3SE +/- 24.63, N = 3SE +/- 0.70, N = 3SE +/- 48.20, N = 151755.333196.121952.613349.762174.193520.712703.094795.461. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

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/pathtracer/real_timeXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1122334455SE +/- 0.01, N = 3SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.16, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 319.9135.9322.3639.3424.6442.8227.1048.76

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P15K30K45K60K75KSE +/- 134.33, N = 3SE +/- 37.55, N = 3SE +/- 64.91, N = 3SE +/- 13.86, N = 3SE +/- 102.40, N = 3SE +/- 64.24, N = 3SE +/- 90.37, N = 3SE +/- 16.05, N = 36897338926625403617958026334115079824249

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P8K16K24K32K40KSE +/- 28.06, N = 3SE +/- 72.42, N = 3SE +/- 83.47, N = 3SE +/- 51.39, N = 3SE +/- 62.46, N = 3SE +/- 34.10, N = 3SE +/- 37.49, N = 3SE +/- 7.13, N = 33705516645337901535731380141382312512124

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: Weld Porosity Detection FP16-INT8 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.80551.6112.41653.2224.0275SE +/- 0.03, N = 4SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 32.502.532.672.793.582.502.362.48MIN: 2.22 / MAX: 27.89MIN: 2.23 / MAX: 31.44MIN: 2.39 / MAX: 34.22MIN: 2.44 / MAX: 34.21MIN: 1.4 / MAX: 12.71MIN: 2.28 / MAX: 40.8MIN: 2.02 / MAX: 21.33MIN: 1.98 / MAX: 34.011. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16-INT8 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P10K20K30K40K50KSE +/- 234.43, N = 4SE +/- 80.58, N = 3SE +/- 21.27, N = 3SE +/- 233.95, N = 3SE +/- 2.71, N = 3SE +/- 53.98, N = 3SE +/- 47.41, N = 3SE +/- 39.08, N = 319183.8337710.6020910.5340032.8115943.1747654.9526943.4248122.161. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Noise Suppression Poconet-Like FP16 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P510152025SE +/- 0.04, N = 3SE +/- 0.12, N = 3SE +/- 0.02, N = 3SE +/- 0.09, N = 3SE +/- 0.00, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 311.3717.1512.7218.7213.9515.969.4013.15MIN: 9.17 / MAX: 32.12MIN: 10.81 / MAX: 59.34MIN: 9.56 / MAX: 30.66MIN: 12.15 / MAX: 64.81MIN: 8.67 / MAX: 65.92MIN: 10.38 / MAX: 43.64MIN: 7.66 / MAX: 25.07MIN: 7.28 / MAX: 47.971. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Noise Suppression Poconet-Like FP16 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2K4K6K8K10KSE +/- 15.13, N = 3SE +/- 37.79, N = 3SE +/- 7.60, N = 3SE +/- 28.98, N = 3SE +/- 1.87, N = 3SE +/- 13.31, N = 3SE +/- 10.14, N = 3SE +/- 24.76, N = 34208.495565.234386.485933.654209.107356.636757.269360.891. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OSPRay Studio

Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity 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.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P9K18K27K36K45KSE +/- 36.47, N = 3SE +/- 25.12, N = 3SE +/- 74.33, N = 3SE +/- 10.97, N = 3SE +/- 61.20, N = 3SE +/- 29.96, N = 3SE +/- 26.00, N = 3SE +/- 8.00, N = 34283719701392381811136649168373199314358

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: CaffeNet 12-int8 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.46980.93961.40941.87922.349SE +/- 0.00884, N = 3SE +/- 0.00720, N = 3SE +/- 0.01053, N = 3SE +/- 0.00871, N = 3SE +/- 0.00936, N = 3SE +/- 0.01492, N = 3SE +/- 0.01491, N = 4SE +/- 0.00701, N = 31.452802.026591.472492.087801.273681.763531.202981.649191. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: CaffeNet 12-int8 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2004006008001000SE +/- 4.16, N = 3SE +/- 1.77, N = 3SE +/- 4.87, N = 3SE +/- 1.98, N = 3SE +/- 5.72, N = 3SE +/- 4.81, N = 3SE +/- 10.06, N = 4SE +/- 2.57, N = 3687.67492.89678.48478.47784.30566.38830.73605.591. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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 Vehicle Bike Detection FP16 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P48121620SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 312.5413.3413.5914.4216.5015.0411.1711.69MIN: 9.17 / MAX: 21.01MIN: 9.67 / MAX: 39.84MIN: 9.53 / MAX: 21.61MIN: 9.74 / MAX: 43.95MIN: 11.95 / MAX: 28.87MIN: 12.32 / MAX: 32.87MIN: 10.18 / MAX: 19.36MIN: 10.28 / MAX: 42.121. (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: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2K4K6K8K10KSE +/- 7.17, N = 3SE +/- 4.05, N = 3SE +/- 9.37, N = 3SE +/- 3.66, N = 3SE +/- 4.52, N = 3SE +/- 8.98, N = 3SE +/- 4.97, N = 3SE +/- 3.37, N = 33802.617127.284093.737689.323606.127862.585725.0310934.171. (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 Re-Identification Retail FP16 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3691215SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 37.137.327.767.979.797.726.726.95MIN: 6.64 / MAX: 12.22MIN: 6.78 / MAX: 18.72MIN: 6.86 / MAX: 12.92MIN: 7.04 / MAX: 21.51MIN: 4.37 / MAX: 18.18MIN: 7.16 / MAX: 18.31MIN: 6.22 / MAX: 11.15MIN: 6.28 / MAX: 20.421. (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 Re-Identification Retail FP16 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4K8K12K16K20KSE +/- 3.72, N = 3SE +/- 3.67, N = 3SE +/- 0.46, N = 3SE +/- 7.22, N = 3SE +/- 3.97, N = 3SE +/- 4.14, N = 3SE +/- 5.57, N = 3SE +/- 7.74, N = 36717.6513102.287207.7414033.945943.6615519.929515.8618375.631. (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: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1530456075SE +/- 0.15, N = 3SE +/- 0.05, N = 3SE +/- 0.21, N = 3SE +/- 0.06, N = 3SE +/- 0.14, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.28, N = 355.7456.7162.5362.1155.9965.0255.0757.13MIN: 44.83 / MAX: 69.99MIN: 45.43 / MAX: 89.46MIN: 48.23 / MAX: 78.43MIN: 47.82 / MAX: 101.26MIN: 27.63 / MAX: 76.45MIN: 55.15 / MAX: 104.43MIN: 46.78 / MAX: 80.73MIN: 39.58 / MAX: 104.911. (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: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P5001000150020002500SE +/- 2.29, N = 3SE +/- 1.56, N = 3SE +/- 2.94, N = 3SE +/- 1.74, N = 3SE +/- 2.65, N = 3SE +/- 0.66, N = 3SE +/- 0.47, N = 3SE +/- 11.53, N = 3860.641691.79894.871801.891068.431843.611159.662237.691. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

QuantLib

QuantLib is an open-source library/framework around quantitative finance for modeling, trading and risk management scenarios. QuantLib is written in C++ with Boost and its built-in benchmark used reports the QuantLib Benchmark Index benchmark score. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.32Configuration: Multi-ThreadedXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P80K160K240K320K400KSE +/- 155.19, N = 3SE +/- 286.86, N = 3SE +/- 40.48, N = 3SE +/- 160.59, N = 3SE +/- 105.60, N = 3SE +/- 362.38, N = 3SE +/- 63.14, N = 3SE +/- 248.00, N = 3143755.3288129.1162727.4316359.2179647.5350187.5195648.8380867.51. (CXX) g++ options: -O3 -march=native -fPIE -pie

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P816243240SE +/- 0.11, N = 3SE +/- 0.40, N = 3SE +/- 0.31, N = 3SE +/- 0.33, N = 3SE +/- 0.04, N = 3SE +/- 0.12, N = 3SE +/- 0.09, N = 3SE +/- 0.09, N = 333.3336.2333.9936.0029.4129.7529.0228.851. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: StandardXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P816243240SE +/- 0.09, N = 3SE +/- 0.31, N = 3SE +/- 0.27, N = 3SE +/- 0.26, N = 3SE +/- 0.05, N = 3SE +/- 0.13, N = 3SE +/- 0.10, N = 3SE +/- 0.11, N = 330.0027.6129.4327.7834.0033.6134.4634.661. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P918273645SE +/- 0.13, N = 3SE +/- 0.55, N = 3SE +/- 0.08, N = 3SE +/- 0.32, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 335.0740.4535.1140.5230.9233.2929.9631.531. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P816243240SE +/- 0.10, N = 3SE +/- 0.34, N = 3SE +/- 0.07, N = 3SE +/- 0.19, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 328.5224.7328.4824.6832.3430.0433.3831.711. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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: JPEG - Quality: 90Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1122334455SE +/- 0.34, N = 8SE +/- 0.01, N = 3SE +/- 0.21, N = 3SE +/- 0.36, N = 5SE +/- 0.35, N = 3SE +/- 0.06, N = 3SE +/- 0.53, N = 15SE +/- 0.67, N = 1239.1039.5943.6334.6343.2042.0448.3642.971. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

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: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.18, N = 3SE +/- 0.10, N = 3SE +/- 0.10, N = 3SE +/- 0.06, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.09, N = 3SE +/- 0.09, N = 341.9342.1659.6964.91100.3945.7839.6243.41MIN: 27.14 / MAX: 54.05MIN: 32.47 / MAX: 70.84MIN: 42.86 / MAX: 84.25MIN: 49.88 / MAX: 107.03MIN: 41.76 / MAX: 139.86MIN: 33.94 / MAX: 85.21MIN: 26.71 / MAX: 62.24MIN: 27.3 / MAX: 82.851. (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 FP16 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P160320480640800SE +/- 1.23, N = 3SE +/- 1.28, N = 3SE +/- 0.52, N = 3SE +/- 0.55, N = 3SE +/- 0.04, N = 3SE +/- 0.69, N = 3SE +/- 0.95, N = 3SE +/- 1.42, N = 3285.95568.81301.30569.60298.56654.84403.20736.231. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: GPT-2 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.02823, N = 3SE +/- 0.08283, N = 3SE +/- 0.07483, N = 3SE +/- 0.01417, N = 3SE +/- 0.01428, N = 3SE +/- 0.03717, N = 3SE +/- 0.00434, N = 3SE +/- 0.02114, N = 35.227817.348485.290277.308225.215206.906194.016875.109481. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: GPT-2 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50100150200250SE +/- 1.04, N = 3SE +/- 1.53, N = 3SE +/- 2.63, N = 3SE +/- 0.26, N = 3SE +/- 0.53, N = 3SE +/- 0.78, N = 3SE +/- 0.27, N = 3SE +/- 0.81, N = 3191.11135.98188.93136.69191.56144.64248.68195.451. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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: Machine Translation EN To DE FP16 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1428425670SE +/- 0.12, N = 3SE +/- 0.07, N = 3SE +/- 0.34, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.16, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 328.7329.9838.5041.5762.4637.5826.8528.17MIN: 19.52 / MAX: 209.26MIN: 20.48 / MAX: 222.7MIN: 25.18 / MAX: 161.13MIN: 28.11 / MAX: 155.06MIN: 46.12 / MAX: 79.06MIN: 20.58 / MAX: 347.67MIN: 20.18 / MAX: 307.38MIN: 20.12 / MAX: 261.311. (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: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2004006008001000SE +/- 1.21, N = 3SE +/- 1.99, N = 3SE +/- 4.09, N = 3SE +/- 1.13, N = 3SE +/- 0.17, N = 3SE +/- 3.48, N = 3SE +/- 0.54, N = 3SE +/- 0.85, N = 3415.96795.63467.08889.18477.77793.64592.201128.781. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Node.js V8 Web Tooling Benchmark

Running the V8 project's Web-Tooling-Benchmark under Node.js. The Web-Tooling-Benchmark stresses JavaScript-related workloads common to web developers like Babel and TypeScript and Babylon. This test profile can test the system's JavaScript performance with Node.js. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgruns/s, More Is BetterNode.js V8 Web Tooling BenchmarkXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P48121620SE +/- 0.11, N = 3SE +/- 0.11, N = 3SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 315.5914.9915.6214.5115.7615.4917.4217.02

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: Vehicle Detection FP16-INT8 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P510152025SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 316.3317.1919.4420.2417.8421.7313.7314.67MIN: 13.49 / MAX: 26.98MIN: 13.16 / MAX: 42.02MIN: 15.1 / MAX: 32.19MIN: 13.69 / MAX: 36.9MIN: 9.08 / MAX: 37.11MIN: 15.08 / MAX: 39.82MIN: 12.23 / MAX: 23.01MIN: 12.73 / MAX: 41.281. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16-INT8 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2K4K6K8K10KSE +/- 8.09, N = 3SE +/- 4.11, N = 3SE +/- 5.68, N = 3SE +/- 8.70, N = 3SE +/- 0.81, N = 3SE +/- 11.51, N = 3SE +/- 0.73, N = 3SE +/- 6.25, N = 32935.475576.862877.325525.563325.905483.814656.928719.371. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: yolov4 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.22, N = 3SE +/- 0.74, N = 3SE +/- 0.01, N = 3SE +/- 0.13, N = 3SE +/- 0.05, N = 3SE +/- 0.09, N = 3SE +/- 0.30, N = 3SE +/- 0.57, N = 361.8376.4261.0179.3855.1278.4047.9070.441. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: yolov4 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P510152025SE +/- 0.06, N = 3SE +/- 0.13, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.13, N = 3SE +/- 0.11, N = 316.1713.0916.3912.6018.1412.7520.8814.201. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: bertsquad-12 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1632486480SE +/- 0.23, N = 3SE +/- 0.50, N = 3SE +/- 0.31, N = 3SE +/- 0.23, N = 3SE +/- 0.18, N = 3SE +/- 0.11, N = 3SE +/- 0.57, N = 3SE +/- 0.12, N = 361.2572.8063.5473.9762.1368.8350.8364.791. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: bertsquad-12 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P510152025SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.22, N = 3SE +/- 0.03, N = 316.3313.7415.7413.5216.1014.5319.6815.431. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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: Face Detection Retail FP16-INT8 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 35.275.555.776.026.435.725.195.18MIN: 4.98 / MAX: 13.16MIN: 5.12 / MAX: 24.99MIN: 5.26 / MAX: 13.86MIN: 5.37 / MAX: 27.77MIN: 3.84 / MAX: 15.83MIN: 5.1 / MAX: 17.93MIN: 4.72 / MAX: 11.02MIN: 4.74 / MAX: 26.351. (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 Retail FP16-INT8 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P5K10K15K20K25KSE +/- 18.13, N = 3SE +/- 12.64, N = 3SE +/- 7.39, N = 3SE +/- 3.89, N = 3SE +/- 5.45, N = 3SE +/- 12.64, N = 3SE +/- 4.30, N = 3SE +/- 17.69, N = 39087.1217252.279686.3918561.919004.2320937.5812317.9424676.861. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: ArcFace ResNet-100 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P816243240SE +/- 0.17, N = 3SE +/- 0.11, N = 3SE +/- 0.12, N = 3SE +/- 0.13, N = 3SE +/- 0.04, N = 3SE +/- 0.11, N = 3SE +/- 0.07, N = 3SE +/- 0.28, N = 324.5530.4723.6634.2521.3434.3116.5531.791. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: ArcFace ResNet-100 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1428425670SE +/- 0.28, N = 3SE +/- 0.12, N = 3SE +/- 0.21, N = 3SE +/- 0.11, N = 3SE +/- 0.10, N = 3SE +/- 0.09, N = 3SE +/- 0.27, N = 3SE +/- 0.27, N = 340.7332.8142.2729.2046.8529.1560.4131.461. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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: Handwritten English Recognition FP16-INT8 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P816243240SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 323.1224.0024.9425.7236.2025.1221.8722.96MIN: 18.74 / MAX: 31.08MIN: 19.19 / MAX: 45.88MIN: 19.75 / MAX: 33.18MIN: 20.7 / MAX: 44.98MIN: 20.73 / MAX: 45.28MIN: 19.86 / MAX: 59.96MIN: 20.16 / MAX: 32.36MIN: 20.8 / MAX: 34.861. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Handwritten English Recognition FP16-INT8 - Device: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P12002400360048006000SE +/- 3.41, N = 3SE +/- 1.09, N = 3SE +/- 4.55, N = 3SE +/- 2.81, N = 3SE +/- 1.46, N = 3SE +/- 2.46, N = 3SE +/- 0.73, N = 3SE +/- 3.43, N = 32074.303998.022243.844351.591643.434744.052924.665570.631. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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: SharpenXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P60120180240300SE +/- 0.00, N = 3SE +/- 1.20, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 1.67, N = 3SE +/- 0.00, N = 3SE +/- 1.67, N = 31141991302271572651672841. (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-GaussianXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 1.15, N = 3SE +/- 0.00, N = 3SE +/- 0.88, N = 31501501581641581821711941. (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: ResizingXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50100150200250SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 32451082139722995230961. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: ResNet50 v1-12-int8 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.01698, N = 3SE +/- 0.01680, N = 3SE +/- 0.03023, N = 3SE +/- 0.03080, N = 3SE +/- 0.00544, N = 3SE +/- 0.02968, N = 3SE +/- 0.02630, N = 3SE +/- 0.06875, N = 33.133666.005253.320306.517443.192986.390313.237806.496931. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: ResNet50 v1-12-int8 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P70140210280350SE +/- 1.74, N = 3SE +/- 0.47, N = 3SE +/- 2.72, N = 3SE +/- 0.73, N = 3SE +/- 0.53, N = 3SE +/- 0.73, N = 3SE +/- 2.50, N = 3SE +/- 1.63, N = 3319.04166.49301.14153.41313.10156.46308.80153.921. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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: HWB Color SpaceXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P60120180240300SE +/- 1.00, N = 3SE +/- 1.20, N = 3SE +/- 0.58, N = 3SE +/- 2.40, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 32861932772072652362852561. (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: EnhancedXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P80160240320400SE +/- 0.00, N = 3SE +/- 2.60, N = 3SE +/- 0.00, N = 3SE +/- 1.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 1.15, N = 31602691813012163422283741. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

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 Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.17Model: super-resolution-10 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1.32442.64883.97325.29766.622SE +/- 0.02534, N = 3SE +/- 0.03854, N = 3SE +/- 0.01694, N = 3SE +/- 0.02155, N = 3SE +/- 0.05586, N = 3SE +/- 0.02062, N = 3SE +/- 0.03569, N = 3SE +/- 0.02988, N = 35.109145.749215.260755.886284.866045.292874.263524.654311. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: super-resolution-10 - Device: CPU - Executor: ParallelXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50100150200250SE +/- 0.98, N = 3SE +/- 1.17, N = 3SE +/- 0.61, N = 3SE +/- 0.62, N = 3SE +/- 2.34, N = 3SE +/- 0.74, N = 3SE +/- 1.98, N = 3SE +/- 1.38, N = 3195.71173.91190.06169.85205.53188.89234.54214.811. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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: SwirlXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P150300450600750SE +/- 0.58, N = 3SE +/- 1.33, N = 3SE +/- 0.67, N = 3SE +/- 0.67, N = 3SE +/- 1.15, N = 3SE +/- 1.76, N = 3SE +/- 0.88, N = 3SE +/- 0.58, N = 33945094325585026175296891. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

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: Random ReadXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P120M240M360M480M600MSE +/- 81327.78, N = 3SE +/- 1120857.00, N = 3SE +/- 86315.43, N = 3SE +/- 785893.13, N = 3SE +/- 3338569.32, N = 3SE +/- 1411218.60, N = 3SE +/- 1119855.47, N = 3SE +/- 82996.27, N = 32053461544055331732344699394737559462653661705250499982975179465783928001. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

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: Classroom - Compute: CPU-OnlyXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.20, N = 3SE +/- 0.38, N = 3SE +/- 0.40, N = 3SE +/- 0.25, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.11, N = 3SE +/- 0.22, N = 386.6744.7176.8740.3868.6135.9163.6133.61

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: 256 - Model: ResNet-50Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1428425670SE +/- 0.76, N = 3SE +/- 0.52, N = 3SE +/- 0.62, N = 3SE +/- 0.45, N = 3SE +/- 0.52, N = 3SE +/- 0.49, N = 3SE +/- 0.64, N = 3SE +/- 0.29, N = 352.9741.6851.5637.9154.8437.4463.6041.45MIN: 33.28 / MAX: 55.78MIN: 37.38 / MAX: 44.02MIN: 30.49 / MAX: 54.7MIN: 16.15 / MAX: 40.53MIN: 46.5 / MAX: 58.5MIN: 26.88 / MAX: 41.26MIN: 37.79 / MAX: 65.67MIN: 38.42 / MAX: 42.93

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 64 - Model: ResNet-50Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1428425670SE +/- 0.71, N = 3SE +/- 0.51, N = 3SE +/- 0.62, N = 3SE +/- 0.46, N = 4SE +/- 0.24, N = 3SE +/- 0.50, N = 3SE +/- 0.12, N = 3SE +/- 0.12, N = 351.5542.3551.3538.3056.0137.0863.0640.77MIN: 37.49 / MAX: 54.33MIN: 36.76 / MAX: 44.4MIN: 41.6 / MAX: 53.81MIN: 27.11 / MAX: 40.39MIN: 51.84 / MAX: 59.12MIN: 26.46 / MAX: 40.15MIN: 33.2 / MAX: 65.54MIN: 16.88 / MAX: 42.06

CloverLeaf

CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeaf 1.3Input: clover_bm64_shortXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1020304050SE +/- 0.15, N = 3SE +/- 0.31, N = 15SE +/- 0.17, N = 3SE +/- 0.40, N = 4SE +/- 0.02, N = 3SE +/- 0.24, N = 3SE +/- 0.03, N = 3SE +/- 0.45, N = 341.4332.2542.2232.4439.4826.5839.3831.901. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

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: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P510152025SE +/- 0.10, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.15, N = 4SE +/- 0.09, N = 15SE +/- 0.17, N = 4SE +/- 1.08, N = 127.677.778.137.2620.9510.4416.617.49

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 TotalXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P7K14K21K28K35KSE +/- 938.95, N = 15SE +/- 175.78, N = 4SE +/- 281.72, N = 15SE +/- 490.54, N = 15SE +/- 236.27, N = 4SE +/- 312.17, N = 3SE +/- 298.34, N = 15SE +/- 356.81, N = 1521286.630236.015081.011866.724688.530108.726378.130547.61. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -mavx512f -mavx512cd -mavx512bw -mavx512dq -fno-trapping-math -fno-math-errno -ldl

RawTherapee

RawTherapee is a cross-platform, open-source multi-threaded RAW image processing program. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRawTherapeeTotal Benchmark TimeXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1326395265SE +/- 0.07, N = 3SE +/- 0.10, N = 3SE +/- 0.08, N = 3SE +/- 0.19, N = 3SE +/- 0.17, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 346.9655.4447.0956.7846.7653.7642.7852.691. RawTherapee, version 5.10, command line.

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 23.6Input: Carbon NanotubeXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1224364860SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.08, N = 3SE +/- 0.27, N = 3SE +/- 0.24, N = 3SE +/- 0.55, N = 3SE +/- 0.12, N = 3SE +/- 0.10, N = 354.3842.8748.4943.9147.7841.5643.0538.051. (CC) gcc options: -shared -lxc -lblas -lmpi

VVenC

VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.11Video Input: Bosphorus 4K - Video Preset: FasterXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P48121620SE +/- 0.05, N = 3SE +/- 0.11, N = 3SE +/- 0.13, N = 3SE +/- 0.13, N = 3SE +/- 0.02, N = 3SE +/- 0.16, N = 3SE +/- 0.08, N = 3SE +/- 0.12, N = 314.1011.9014.3611.9715.0612.5415.9713.261. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects

Coremark

This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1.1M2.2M3.3M4.4M5.5MSE +/- 1747.99, N = 3SE +/- 7219.73, N = 3SE +/- 5202.86, N = 3SE +/- 20649.43, N = 3SE +/- 9256.47, N = 3SE +/- 12743.41, N = 3SE +/- 4635.95, N = 3SE +/- 9700.21, N = 31614763.702993136.601855135.333279262.532194630.603902076.242369998.364966250.781. (CC) gcc options: -O2 -lrt" -lrt

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: SmallXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P10K20K30K40K50KSE +/- 601.03, N = 12SE +/- 1097.75, N = 15SE +/- 572.00, N = 12SE +/- 283.44, N = 3SE +/- 11.14, N = 4SE +/- 467.78, N = 4SE +/- 135.77, N = 4SE +/- 1633.92, N = 1520091.119876.518621.221583.235230.146877.531830.242114.61. (CXX) g++ options: -O3 -fopenmp -lmpi_cxx -lmpi

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 CompileXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1122334455SE +/- 0.04, N = 3SE +/- 0.06, N = 3SE +/- 0.16, N = 3SE +/- 0.07, N = 3SE +/- 0.16, N = 3SE +/- 0.07, N = 3SE +/- 0.31, N = 3SE +/- 0.09, N = 348.7146.8747.9546.5748.0146.6441.8540.38

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: WPA PSKXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P200K400K600K800K1000KSE +/- 987.33, N = 3SE +/- 5122.51, N = 3SE +/- 1274.00, N = 3SE +/- 788.50, N = 3SE +/- 1599.31, N = 3SE +/- 2519.00, N = 3SE +/- 864.17, N = 3SE +/- 15579.94, N = 152865525628753340766578894012787932534281167970931. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

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: Decompression RatingXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P120K240K360K480K600KSE +/- 916.22, N = 3SE +/- 2598.93, N = 3SE +/- 374.59, N = 3SE +/- 6276.88, N = 3SE +/- 61.74, N = 3SE +/- 185.14, N = 3SE +/- 119.21, N = 3SE +/- 28320.74, N = 32077384017072433474691312878625570543381465734101. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Compression RatingXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P140K280K420K560K700KSE +/- 1836.55, N = 3SE +/- 5127.15, N = 3SE +/- 2258.81, N = 3SE +/- 3924.87, N = 3SE +/- 1535.53, N = 3SE +/- 4200.62, N = 3SE +/- 1982.97, N = 3SE +/- 709.87, N = 33147164917293423635255143881905927294416366742391. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

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 2.6Scene: Rainbow Colors and Prism - Acceleration: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P48121620SE +/- 0.12, N = 7SE +/- 0.07, N = 15SE +/- 0.06, N = 4SE +/- 0.03, N = 3SE +/- 0.05, N = 4SE +/- 0.10, N = 15SE +/- 0.15, N = 4SE +/- 0.04, N = 312.809.6412.976.5113.545.9614.287.22MIN: 11.39 / MAX: 13.21MIN: 8.98 / MAX: 10.13MIN: 12.11 / MAX: 13.11MIN: 6.27 / MAX: 6.67MIN: 13.42 / MAX: 13.68MIN: 5.5 / MAX: 7.04MIN: 12.85 / MAX: 14.54MIN: 7.13 / MAX: 7.4

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 1 - Buffer Length: 256 - Filter Length: 512Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4M8M12M16M20MSE +/- 5238.74, N = 3SE +/- 17457.89, N = 3SE +/- 30179.10, N = 3SE +/- 7571.88, N = 3SE +/- 8685.88, N = 3SE +/- 38670.98, N = 3SE +/- 9492.69, N = 3SE +/- 162096.41, N = 1516906333169046671688533316884000168823331684333318774333182005331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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: defconfigXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1020304050SE +/- 0.59, N = 3SE +/- 0.40, N = 4SE +/- 0.57, N = 3SE +/- 0.31, N = 5SE +/- 0.55, N = 3SE +/- 0.43, N = 3SE +/- 0.39, N = 4SE +/- 0.23, N = 743.4232.1440.5030.5339.4529.9834.5925.87

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100, Lossless, Highest CompressionXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.1350.270.4050.540.675SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.550.540.550.540.520.510.600.591. (CC) gcc options: -fvisibility=hidden -O2 -lm

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: 512 - Model: AlexNetXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P400800120016002000SE +/- 3.89, N = 3SE +/- 17.88, N = 4SE +/- 5.13, N = 3SE +/- 8.55, N = 3SE +/- 0.51, N = 3SE +/- 18.52, N = 3SE +/- 1.84, N = 3SE +/- 19.23, N = 31047.961604.111108.951677.621278.451899.641486.122039.63

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: 90Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1122334455SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.32, N = 3SE +/- 0.17, N = 3SE +/- 0.12, N = 3SE +/- 0.49, N = 14SE +/- 0.50, N = 343.3634.8340.3533.2742.9140.0846.3239.521. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

Primesieve

Primesieve generates prime numbers using a highly optimized sieve of Eratosthenes implementation. Primesieve primarily benchmarks the CPU's L1/L2 cache performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 12.1Length: 1e13Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1530456075SE +/- 0.07, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 367.3233.9258.1630.2051.8227.1148.4125.021. (CXX) g++ options: -O3

Graph500

This is a benchmark of the reference implementation of Graph500, an HPC benchmark focused on data intensive loads and commonly tested on supercomputers for complex data problems. Graph500 primarily stresses the communication subsystem of the hardware under test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsssp median_TEPS, More Is BetterGraph500 3.0Scale: 26Xeon Platinum 8592+Xeon Platinum 8592+ 2P100M200M300M400M500M4165230004849770001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

Scale: 26

Xeon Platinum 8490H: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Platinum 8490H 2P: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9468: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9468 2P: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9480: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9480 2P: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

OpenBenchmarking.orgsssp max_TEPS, More Is BetterGraph500 3.0Scale: 26Xeon Platinum 8592+Xeon Platinum 8592+ 2P140M280M420M560M700M5396130006673650001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

Scale: 26

Xeon Platinum 8490H: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Platinum 8490H 2P: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9468: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9468 2P: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9480: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9480 2P: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

OpenBenchmarking.orgbfs median_TEPS, More Is BetterGraph500 3.0Scale: 26Xeon Platinum 8592+Xeon Platinum 8592+ 2P300M600M900M1200M1500M124430000012045200001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

Scale: 26

Xeon Platinum 8490H: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Platinum 8490H 2P: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9468: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9468 2P: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9480: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9480 2P: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

OpenBenchmarking.orgbfs max_TEPS, More Is BetterGraph500 3.0Scale: 26Xeon Platinum 8592+Xeon Platinum 8592+ 2P300M600M900M1200M1500M129917000012674800001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

Scale: 26

Xeon Platinum 8490H: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Platinum 8490H 2P: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9468: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9468 2P: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9480: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Xeon Max 9480 2P: The test quit with a non-zero exit status. E: AML: Fatal: non power2 groupsize unsupported. Define macro PROCS_PER_NODE_NOT_POWER_OF_TWO to override

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 1 - Buffer Length: 256 - Filter Length: 57Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P16M32M48M64M80MSE +/- 4910.31, N = 3SE +/- 4163.33, N = 3SE +/- 5364.49, N = 3SE +/- 822374.71, N = 4SE +/- 666.67, N = 3SE +/- 3844.19, N = 3SE +/- 599433.09, N = 9SE +/- 68119.99, N = 367130333670640006712766766266000670746676699133373140667704990001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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: PUSCH Processor Benchmark, Throughput TotalXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P12002400360048006000SE +/- 0.09, N = 3SE +/- 26.47, N = 3SE +/- 18.30, N = 3SE +/- 0.15, N = 3SE +/- 18.43, N = 3SE +/- 47.68, N = 6SE +/- 18.09, N = 3SE +/- 0.12, N = 32170.73787.02514.04386.62616.94853.63088.75493.4MIN: 1388.2 / MAX: 2170.8MIN: 2545.7 / MAX: 3839.9MIN: 1550.4 / MAX: 2532.5MIN: 2970.2 / MAX: 4386.9MIN: 1591.7 / MAX: 2653.8MIN: 3181.8 / MAX: 4929.4MIN: 1852.6 / MAX: 3124.9MIN: 3703.4 / MAX: 5493.61. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -mavx512f -mavx512cd -mavx512bw -mavx512dq -fno-trapping-math -fno-math-errno -ldl

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: 2Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P918273645SE +/- 0.21, N = 3SE +/- 0.22, N = 3SE +/- 0.31, N = 3SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.16, N = 3SE +/- 0.05, N = 3SE +/- 0.09, N = 341.1338.6539.8838.1437.7537.3134.7933.651. (CXX) g++ options: -O3 -fPIC -lm

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - GriddingXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P15K30K45K60K75KSE +/- 99.83, N = 3SE +/- 194.30, N = 3SE +/- 0.00, N = 3SE +/- 802.88, N = 3SE +/- 76.67, N = 3SE +/- 412.68, N = 4SE +/- 219.20, N = 3SE +/- 1080.03, N = 321766.542939.223250.245195.121314.241773.037265.972004.01. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - DegriddingXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P13K26K39K52K65KSE +/- 0.00, N = 3SE +/- 143.60, N = 3SE +/- 76.47, N = 3SE +/- 0.00, N = 3SE +/- 93.91, N = 3SE +/- 378.22, N = 4SE +/- 153.57, N = 3SE +/- 480.95, N = 318522.036900.520561.440368.617891.634614.131177.259139.21. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

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: Cell Phone Drop TestXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P816243240SE +/- 0.08, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.08, N = 3SE +/- 0.10, N = 3SE +/- 0.18, N = 3SE +/- 0.32, N = 333.7527.5131.9227.6331.4928.2227.9026.88

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 64 - Buffer Length: 256 - Filter Length: 512Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P200M400M600M800M1000MSE +/- 1611276.24, N = 3SE +/- 4432294.36, N = 3SE +/- 1289603.73, N = 3SE +/- 6141471.96, N = 3SE +/- 940265.92, N = 3SE +/- 3717554.45, N = 3SE +/- 1515552.70, N = 3SE +/- 7562646.32, N = 86323966677325600006890033337571766677444100007994966677837700008837012501. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 1 - Buffer Length: 256 - Filter Length: 32Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P9M18M27M36M45MSE +/- 1666.67, N = 3SE +/- 1201.85, N = 3SE +/- 3000.00, N = 3SE +/- 1452.97, N = 3SE +/- 1527.53, N = 3SE +/- 1154.70, N = 3SE +/- 417042.80, N = 5SE +/- 419600.12, N = 536537667364976673653900036521333365060003646100040244000389506001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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 2.3Time To CompileXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P816243240SE +/- 0.16, N = 3SE +/- 0.12, N = 3SE +/- 0.16, N = 3SE +/- 0.17, N = 3SE +/- 0.03, N = 3SE +/- 0.11, N = 3SE +/- 0.05, N = 3SE +/- 0.07, N = 335.6636.2035.2035.7034.9335.4630.1730.471. (CC) gcc options: -m64 -ldl -lgcc_s -lutil -lrt -lpthread -lm -lc -pie -nodefaultlibs

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-50Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.28, N = 3SE +/- 0.38, N = 8SE +/- 0.36, N = 3SE +/- 0.44, N = 3SE +/- 0.03, N = 3SE +/- 0.49, N = 3SE +/- 0.16, N = 3SE +/- 0.52, N = 350.9946.4948.5935.4651.4635.5377.5245.69MIN: 11.73 / MAX: 59.01MIN: 15.6 / MAX: 50.65MIN: 4.58 / MAX: 55.2MIN: 0.7 / MAX: 41MIN: 35.78 / MAX: 55.29MIN: 16.36 / MAX: 38.41MIN: 36.51 / MAX: 79.65MIN: 13.24 / MAX: 48.92

Appleseed

Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: Disney MaterialXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P132639526558.3851.2556.4049.1453.4948.5648.4343.70

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.orgMT/s, More Is BetterASTC Encoder 4.7Preset: Very ThoroughXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P48121620SE +/- 0.0016, N = 3SE +/- 0.0055, N = 3SE +/- 0.0016, N = 3SE +/- 0.0022, N = 3SE +/- 0.0054, N = 3SE +/- 0.0379, N = 3SE +/- 0.0294, N = 3SE +/- 0.0007, N = 35.134210.20515.970211.88757.088513.75187.624014.63341. (CXX) g++ options: -O3 -flto -pthread

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 64 - Buffer Length: 256 - Filter Length: 57Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P600M1200M1800M2400M3000MSE +/- 2629533.12, N = 3SE +/- 28812627.44, N = 4SE +/- 5493127.02, N = 3SE +/- 27382304.50, N = 5SE +/- 7076957.92, N = 3SE +/- 28500058.48, N = 3SE +/- 11426868.92, N = 3SE +/- 32405829.38, N = 3213876666724940500002308966667249134000025392000002655500000247550000026200666671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: SlowXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P612182430SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 313.3619.7513.8922.1415.7124.7116.0627.11

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.orgMT/s, More Is BetterASTC Encoder 4.7Preset: ExhaustiveXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.0003, N = 3SE +/- 0.0033, N = 3SE +/- 0.0011, N = 3SE +/- 0.0016, N = 3SE +/- 0.0024, N = 3SE +/- 0.0164, N = 3SE +/- 0.0072, N = 3SE +/- 0.0157, N = 33.12046.21263.62767.20754.26898.37014.63038.90911. (CXX) g++ options: -O3 -flto -pthread

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 128 - Buffer Length: 256 - Filter Length: 512Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P300M600M900M1200M1500MSE +/- 681281.47, N = 3SE +/- 11560043.25, N = 5SE +/- 1345523.44, N = 3SE +/- 2123152.79, N = 3SE +/- 1740596.96, N = 3SE +/- 6842108.19, N = 3SE +/- 2313246.88, N = 3SE +/- 4790383.89, N = 3800186667109146000088220000012060666679185666671126466667115293333314938333331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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 13 - Input: Bosphorus 4KXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 1.44, N = 15SE +/- 1.63, N = 15SE +/- 1.62, N = 15SE +/- 1.67, N = 15SE +/- 2.31, N = 15SE +/- 2.18, N = 15SE +/- 2.83, N = 15SE +/- 2.84, N = 15144.70122.62150.00124.26162.47154.50174.83162.981. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

QuantLib

QuantLib is an open-source library/framework around quantitative finance for modeling, trading and risk management scenarios. QuantLib is written in C++ with Boost and its built-in benchmark used reports the QuantLib Benchmark Index benchmark score. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.32Configuration: Single-ThreadedXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P8001600240032004000SE +/- 22.17, N = 3SE +/- 28.11, N = 3SE +/- 25.35, N = 3SE +/- 28.57, N = 3SE +/- 28.42, N = 3SE +/- 25.40, N = 3SE +/- 33.58, N = 3SE +/- 30.36, N = 123225.73207.03229.13222.83209.63210.13600.53586.31. (CXX) g++ options: -O3 -march=native -fPIE -pie

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 128 - Buffer Length: 256 - Filter Length: 57Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1000M2000M3000M4000M5000MSE +/- 2030052.00, N = 3SE +/- 25888414.40, N = 3SE +/- 3119294.79, N = 3SE +/- 14772609.79, N = 3SE +/- 3116800.35, N = 3SE +/- 13655808.69, N = 3SE +/- 5260228.13, N = 3SE +/- 51282832.98, N = 4267003333335964000002972400000389650000031812666673723666667350320000044505250001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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: 1000Xeon Max 9480Xeon Platinum 8592+Xeon Platinum 8592+ 2P90K180K270K360K450KSE +/- 125.38, N = 3SE +/- 142.71, N = 3SE +/- 576.43, N = 2367199.06399698.78381352.691. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Connections: 1000

Xeon Platinum 8490H: The test quit with a non-zero exit status.

Xeon Platinum 8490H 2P: The test quit with a non-zero exit status.

Xeon Max 9468: The test quit with a non-zero exit status.

Xeon Max 9468 2P: The test quit with a non-zero exit status.

Xeon Max 9480 2P: The test quit with a non-zero exit status.

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 256 - Buffer Length: 256 - Filter Length: 512Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P400M800M1200M1600M2000MSE +/- 569766.03, N = 3SE +/- 4238841.56, N = 3SE +/- 740202.67, N = 3SE +/- 6728628.72, N = 3SE +/- 1722723.94, N = 3SE +/- 5630373.98, N = 3SE +/- 240370.09, N = 3SE +/- 14465476.14, N = 38385300001444066667944550000151893333310208333331616966667117876666720270000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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-OnlyXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1020304050SE +/- 0.04, N = 3SE +/- 0.06, N = 3SE +/- 0.10, N = 3SE +/- 0.19, N = 3SE +/- 0.24, N = 3SE +/- 0.04, N = 3SE +/- 0.10, N = 3SE +/- 0.09, N = 342.8524.9939.1723.4535.9621.1532.0818.31

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.DXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3K6K9K12K15KSE +/- 155.10, N = 12SE +/- 4.83, N = 4SE +/- 6.53, N = 3SE +/- 10.93, N = 4SE +/- 6.77, N = 3SE +/- 100.23, N = 4SE +/- 22.33, N = 3SE +/- 30.19, N = 45366.6910970.456445.0912793.017562.8014946.768089.1416004.071. (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

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: bcryptXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P40K80K120K160K200KSE +/- 8.84, N = 3SE +/- 866.68, N = 3SE +/- 8.67, N = 3SE +/- 576.80, N = 3SE +/- 10.67, N = 3SE +/- 292.24, N = 3SE +/- 46.28, N = 3SE +/- 288.38, N = 36787513386679144155959945511836031057812067781. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: BlowfishXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P40K80K120K160K200KSE +/- 17.33, N = 3SE +/- 116.69, N = 3SE +/- 8.67, N = 3SE +/- 169.28, N = 3SE +/- 11.39, N = 3SE +/- 348.45, N = 3SE +/- 79.75, N = 3SE +/- 649.08, N = 36789313456279152156658945131835831058312054991. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 256 - Buffer Length: 256 - Filter Length: 57Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1300M2600M3900M5200M6500MSE +/- 4088330.28, N = 3SE +/- 17558125.69, N = 3SE +/- 1976810.00, N = 3SE +/- 28453880.81, N = 3SE +/- 5292237.50, N = 3SE +/- 2331189.49, N = 3SE +/- 3555434.03, N = 3SE +/- 61971606.40, N = 3284893333348251333333222533333503600000035008333335361266667380683333362843000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 256 - Buffer Length: 256 - Filter Length: 32Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1300M2600M3900M5200M6500MSE +/- 1320353.49, N = 3SE +/- 2447674.63, N = 3SE +/- 1338323.99, N = 3SE +/- 895048.11, N = 3SE +/- 4021608.30, N = 3SE +/- 10496401.50, N = 3SE +/- 2516611.48, N = 3SE +/- 6711433.03, N = 3238180000046709666672769066667536703333332438000006116566667316720000060773000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: MediumXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P714212835SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.16, N = 3SE +/- 0.05, N = 315.1722.2515.6824.6517.8527.6516.6430.37

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 128 - Buffer Length: 256 - Filter Length: 32Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P800M1600M2400M3200M4000MSE +/- 400000.00, N = 3SE +/- 4598671.31, N = 3SE +/- 1342054.81, N = 3SE +/- 2403700.85, N = 3SE +/- 1858314.65, N = 3SE +/- 26633124.74, N = 3SE +/- 2236316.42, N = 3SE +/- 13813198.20, N = 3235190000032540333332739633333333706666732006000003558600000314823333335207666671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Google SynthMark

SynthMark is a cross platform tool for benchmarking CPU performance under a variety of real-time audio workloads. It uses a polyphonic synthesizer model to provide standardized tests for latency, jitter and computational throughput. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVoices, More Is BetterGoogle SynthMark 20201109Test: VoiceMark_100Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2004006008001000SE +/- 0.26, N = 3SE +/- 0.30, N = 3SE +/- 0.24, N = 3SE +/- 0.48, N = 3SE +/- 0.22, N = 3SE +/- 1.51, N = 3SE +/- 0.13, N = 3SE +/- 0.38, N = 3711.10710.92712.01712.09711.38712.70799.72799.491. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 64 - Buffer Length: 256 - Filter Length: 32Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P500M1000M1500M2000M2500MSE +/- 688799.28, N = 3SE +/- 2206304.10, N = 3SE +/- 692820.32, N = 3SE +/- 3017909.50, N = 3SE +/- 3167192.94, N = 3SE +/- 3300505.01, N = 3SE +/- 27191440.65, N = 3SE +/- 17236814.61, N = 3164026666719632333331679700000209046666718905666672025000000192776666721546333331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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: Fishy Cat - Compute: CPU-OnlyXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1020304050SE +/- 0.17, N = 3SE +/- 0.13, N = 3SE +/- 0.04, N = 3SE +/- 0.18, N = 3SE +/- 0.11, N = 3SE +/- 0.08, N = 3SE +/- 0.06, N = 3SE +/- 0.04, N = 342.7823.1738.4821.4534.8319.1331.9117.29

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: GoogLeNetXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P816243240SE +/- 0.27, N = 7SE +/- 0.20, N = 15SE +/- 0.27, N = 6SE +/- 0.23, N = 12SE +/- 0.19, N = 6SE +/- 0.23, N = 15SE +/- 0.21, N = 6SE +/- 0.12, N = 636.1318.2431.4714.6733.5613.5021.6712.69

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance of various popular real-world Java workloads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: Apache Lucene Search IndexXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P8001600240032004000SE +/- 33.41, N = 3SE +/- 10.84, N = 3SE +/- 42.24, N = 4SE +/- 11.86, N = 3SE +/- 11.79, N = 3SE +/- 41.09, N = 5SE +/- 18.19, N = 3SE +/- 12.44, N = 333783643339635373581368932423345

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 Dragon ObjXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 0.08, N = 3SE +/- 0.76, N = 3SE +/- 0.09, N = 3SE +/- 0.89, N = 15SE +/- 0.09, N = 3SE +/- 0.22, N = 4SE +/- 0.35, N = 3SE +/- 0.29, N = 464.96104.8471.83112.7279.38132.2897.19173.22MIN: 63.12 / MAX: 67.81MIN: 101.05 / MAX: 112.02MIN: 69.55 / MAX: 76.73MIN: 102 / MAX: 125.96MIN: 76.07 / MAX: 85.98MIN: 125 / MAX: 146.52MIN: 92.96 / MAX: 102.04MIN: 164.07 / MAX: 187.8

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_bareXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P510152025SE +/- 0.019, N = 3SE +/- 0.025, N = 3SE +/- 0.007, N = 3SE +/- 0.032, N = 3SE +/- 0.010, N = 3SE +/- 0.019, N = 3SE +/- 0.009, N = 3SE +/- 0.046, N = 37.54513.3928.28214.1278.76215.45010.47718.5401. (CXX) g++ options: -O3 -lm

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 4 - Input: Bosphorus 4KXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.027, N = 3SE +/- 0.015, N = 3SE +/- 0.047, N = 3SE +/- 0.042, N = 3SE +/- 0.030, N = 3SE +/- 0.049, N = 3SE +/- 0.005, N = 3SE +/- 0.044, N = 36.4905.9916.5726.0527.0616.6987.8767.3601. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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: 64 - Model: GoogLeNetXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P80160240320400SE +/- 1.94, N = 3SE +/- 3.52, N = 5SE +/- 0.83, N = 3SE +/- 2.38, N = 3SE +/- 1.62, N = 3SE +/- 2.37, N = 3SE +/- 1.12, N = 3SE +/- 0.21, N = 3333.73341.51335.58308.95380.14331.76319.76291.37

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - DegriddingXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P11K22K33K44K55KSE +/- 41.81, N = 5SE +/- 0.00, N = 4SE +/- 69.31, N = 5SE +/- 955.08, N = 5SE +/- 677.85, N = 15SE +/- 948.80, N = 15SE +/- 131.12, N = 15SE +/- 1365.55, N = 156758.814437.607046.567304.5830471.5051476.2016936.8048686.801. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - GriddingXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4K8K12K16K20KSE +/- 13.21, N = 5SE +/- 40.14, N = 4SE +/- 33.09, N = 5SE +/- 13.50, N = 5SE +/- 195.05, N = 15SE +/- 139.77, N = 15SE +/- 211.64, N = 15SE +/- 221.77, N = 154173.463916.784438.583836.7319798.5017010.8019141.8020204.901. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: SlowXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P918273645SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.07, N = 3SE +/- 0.04, N = 417.5226.0018.2428.9620.2331.2424.2340.921. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: MediumXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P918273645SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.27, N = 418.1326.6918.8429.6220.9732.0424.8041.581. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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 8 - Input: Bosphorus 4KXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.26, N = 4SE +/- 0.58, N = 5SE +/- 0.29, N = 4SE +/- 0.44, N = 15SE +/- 0.20, N = 5SE +/- 0.46, N = 4SE +/- 0.22, N = 5SE +/- 0.79, N = 562.7554.8265.3655.9870.9165.2279.4470.431. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Ultra FastXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1428425670SE +/- 0.06, N = 4SE +/- 0.50, N = 5SE +/- 0.13, N = 4SE +/- 0.36, N = 15SE +/- 0.07, N = 5SE +/- 0.58, N = 5SE +/- 0.04, N = 4SE +/- 0.17, N = 548.9146.1350.0945.7960.8755.2639.3157.13

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: 256 - Model: AlexNetXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P400800120016002000SE +/- 1.21, N = 3SE +/- 1.15, N = 3SE +/- 6.82, N = 3SE +/- 5.81, N = 3SE +/- 0.54, N = 3SE +/- 5.06, N = 3SE +/- 4.27, N = 3SE +/- 1.86, N = 3986.041479.531058.991528.411205.301730.251277.461681.44

miniBUDE

MiniBUDE is a mini application for the the core computation of the Bristol University Docking Engine (BUDE). This test profile currently makes use of the OpenMP implementation of miniBUDE for CPU benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 0.12, N = 4SE +/- 1.28, N = 12SE +/- 0.84, N = 15SE +/- 1.85, N = 12SE +/- 0.61, N = 5SE +/- 0.81, N = 7SE +/- 0.20, N = 6SE +/- 1.02, N = 863.62108.4070.84114.9082.85133.63106.40199.731. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P11002200330044005500SE +/- 2.91, N = 4SE +/- 32.01, N = 12SE +/- 21.10, N = 15SE +/- 46.19, N = 12SE +/- 15.23, N = 5SE +/- 20.15, N = 7SE +/- 5.07, N = 6SE +/- 25.44, N = 81590.502709.931771.072872.562071.193340.692659.884993.341. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

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: AlexNetXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1326395265SE +/- 0.76, N = 15SE +/- 0.57, N = 15SE +/- 1.22, N = 15SE +/- 0.52, N = 15SE +/- 0.08, N = 8SE +/- 0.44, N = 15SE +/- 0.66, N = 15SE +/- 0.45, N = 1548.2837.7643.6733.2156.5337.5735.5030.76

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: BMW27 - Compute: CPU-OnlyXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P714212835SE +/- 0.11, N = 3SE +/- 0.03, N = 3SE +/- 0.24, N = 3SE +/- 0.08, N = 4SE +/- 0.15, N = 3SE +/- 0.08, N = 4SE +/- 0.04, N = 3SE +/- 0.03, N = 431.6616.5528.4915.2925.8913.7423.7212.76

Y-Cruncher

Y-Cruncher is a multi-threaded Pi benchmark capable of computing Pi to trillions of digits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.8.3Pi Digits To Calculate: 1BXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.045, N = 5SE +/- 0.122, N = 15SE +/- 0.026, N = 5SE +/- 0.142, N = 12SE +/- 0.005, N = 5SE +/- 0.007, N = 6SE +/- 0.030, N = 5SE +/- 0.010, N = 68.3006.3347.9596.2667.0475.0086.7015.110

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 - Execution TimeXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P102030405045.9528.9941.7527.5842.2228.5633.0723.561. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Mesh TimeXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P81624324031.7832.4832.0028.8431.3835.6031.7029.331. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -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: IS.DXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P16003200480064008000SE +/- 15.55, N = 4SE +/- 12.99, N = 4SE +/- 18.08, N = 4SE +/- 4.96, N = 4SE +/- 6.56, N = 3SE +/- 23.92, N = 4SE +/- 8.92, N = 4SE +/- 155.00, N = 153050.163816.703076.663819.832865.723960.284059.707437.811. (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

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Super FastXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1530456075SE +/- 0.12, N = 5SE +/- 0.49, N = 5SE +/- 0.20, N = 5SE +/- 0.49, N = 15SE +/- 0.07, N = 5SE +/- 0.50, N = 5SE +/- 0.09, N = 4SE +/- 0.21, N = 553.1352.6354.4754.1966.7469.0539.2066.781. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

Algebraic Multi-Grid Benchmark

AMG is a parallel algebraic multigrid solver for linear systems arising from problems on unstructured grids. The driver provided with AMG builds linear systems for various 3-dimensional problems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid Benchmark 1.2Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P800M1600M2400M3200M4000MSE +/- 3359086.30, N = 3SE +/- 7233468.75, N = 3SE +/- 1778769.08, N = 3SE +/- 9164253.67, N = 3SE +/- 468935.32, N = 3SE +/- 2438223.49, N = 3SE +/- 4000850.45, N = 3SE +/- 9014927.20, N = 3158849666731562616671597993333320516100016050970003124402667181476000036096463331. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi

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 DirectionXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P48121620SE +/- 0.04445674, N = 4SE +/- 0.01741826, N = 5SE +/- 0.01708632, N = 4SE +/- 0.05695382, N = 15SE +/- 0.01004100, N = 4SE +/- 0.04312801, N = 5SE +/- 0.02923582, N = 4SE +/- 0.03329814, N = 513.969237607.2934911712.799758206.6638751012.929341306.3916981710.196706605.391259671. (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

Timed FFmpeg Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 7.0Time To CompileXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P612182430SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.05, N = 326.4620.7924.8420.0124.0619.5121.3617.39

Intel Open Image Denoise

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

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 2.2Run: RTLightmap.hdr.4096x4096 - Device: CPU-OnlyXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.541.081.622.162.7SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 4SE +/- 0.00, N = 3SE +/- 0.01, N = 41.081.721.141.741.292.181.582.40

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: Deconvolution Batch shapes_1d - Engine: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P918273645SE +/- 0.00876, N = 3SE +/- 0.10586, N = 3SE +/- 0.02700, N = 3SE +/- 0.09318, N = 3SE +/- 0.00627, N = 3SE +/- 0.10023, N = 3SE +/- 0.08324, N = 3SE +/- 0.32768, N = 35.3095313.450006.0974716.090106.1585616.550706.6969238.08130MIN: 11.97MIN: 9.89MIN: 14.58MIN: 3.74MIN: 9.761. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

x265

This is a simple test of the x265 encoder run on the CPU with 1080p and 4K options for H.265 video encode performance with x265. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.6Video Input: Bosphorus 4KXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P918273645SE +/- 0.08, N = 3SE +/- 0.12, N = 3SE +/- 0.39, N = 3SE +/- 0.10, N = 3SE +/- 0.11, N = 4SE +/- 0.20, N = 4SE +/- 0.06, N = 4SE +/- 0.09, N = 429.0527.2631.4027.9037.8637.0438.9339.311. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Very FastXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1224364860SE +/- 0.12, N = 4SE +/- 0.39, N = 4SE +/- 0.02, N = 4SE +/- 0.38, N = 9SE +/- 0.01, N = 4SE +/- 0.50, N = 5SE +/- 0.08, N = 3SE +/- 0.16, N = 539.5644.7540.8246.2549.4254.8835.1053.66

LULESH

LULESH is the Livermore Unstructured Lagrangian Explicit Shock Hydrodynamics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.3Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P15K30K45K60K75KSE +/- 211.12, N = 6SE +/- 206.20, N = 5SE +/- 217.67, N = 6SE +/- 207.54, N = 5SE +/- 182.38, N = 15SE +/- 272.44, N = 4SE +/- 437.51, N = 4SE +/- 359.29, N = 323848.5452102.6725831.8453269.4723072.3043724.0638433.9770826.731. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi

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 1D - Engine: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.17620.35240.52860.70480.881SE +/- 0.001869, N = 4SE +/- 0.004822, N = 4SE +/- 0.000666, N = 4SE +/- 0.003500, N = 4SE +/- 0.000915, N = 4SE +/- 0.004208, N = 4SE +/- 0.000767, N = 4SE +/- 0.004443, N = 40.7259740.6623870.6171900.7594340.5394670.7831800.3984460.580664MIN: 0.67MIN: 0.58MIN: 0.55MIN: 0.67MIN: 0.5MIN: 0.69MIN: 0.37MIN: 0.511. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

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: 64 - Model: AlexNetXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P2004006008001000SE +/- 1.95, N = 5SE +/- 6.39, N = 13SE +/- 2.31, N = 5SE +/- 7.35, N = 5SE +/- 0.51, N = 5SE +/- 7.98, N = 5SE +/- 6.22, N = 5SE +/- 2.22, N = 5841.68932.59852.36940.40969.62996.47819.89858.84

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.CXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50K100K150K200K250KSE +/- 67.02, N = 3SE +/- 679.14, N = 4SE +/- 9.58, N = 3SE +/- 420.81, N = 4SE +/- 150.79, N = 3SE +/- 560.20, N = 4SE +/- 31.76, N = 4SE +/- 1528.09, N = 558815.36135939.2569776.41141954.1571245.26147311.14109416.42228184.681. (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

Pennant

Pennant is an application focused on hydrodynamics on general unstructured meshes in 2D. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: sedovbigXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P510152025SE +/- 0.001847, N = 3SE +/- 0.003237, N = 4SE +/- 0.013353, N = 3SE +/- 0.003253, N = 4SE +/- 0.002987, N = 3SE +/- 0.016400, N = 4SE +/- 0.002400, N = 3SE +/- 0.018471, N = 421.02112013.21125019.88246012.10690016.03080011.28438016.9691009.9645261. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Ultra FastXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.13, N = 5SE +/- 0.55, N = 5SE +/- 0.32, N = 5SE +/- 0.56, N = 5SE +/- 0.75, N = 15SE +/- 0.54, N = 6SE +/- 0.17, N = 4SE +/- 0.39, N = 663.1054.9863.9958.1474.5772.0147.0773.811. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Super FastXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1224364860SE +/- 0.15, N = 4SE +/- 0.43, N = 4SE +/- 0.15, N = 4SE +/- 0.41, N = 7SE +/- 0.05, N = 5SE +/- 0.16, N = 5SE +/- 0.07, N = 3SE +/- 0.21, N = 544.0545.0945.3244.4055.5455.4035.5654.95

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance of various popular real-world Java workloads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: JythonXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P8001600240032004000SE +/- 20.55, N = 4SE +/- 28.30, N = 4SE +/- 38.86, N = 5SE +/- 24.34, N = 4SE +/- 7.09, N = 4SE +/- 24.09, N = 4SE +/- 21.99, N = 4SE +/- 9.71, N = 437853926370739213801389234243466

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: 100Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P714212835SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 4SE +/- 0.08, N = 426.2526.8427.0926.9928.1027.6631.3530.971. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

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 CompileXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P510152025SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.08, N = 3SE +/- 0.17, N = 3SE +/- 0.07, N = 3SE +/- 0.12, N = 3SE +/- 0.07, N = 3SE +/- 0.05, N = 420.1818.4719.6518.1419.2318.2917.2616.18

PyBench

This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P170340510680850SE +/- 0.33, N = 3SE +/- 0.67, N = 3SE +/- 0.58, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 3SE +/- 1.15, N = 3SE +/- 0.41, N = 4SE +/- 0.48, N = 4763763762763765767691692

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance of various popular real-world Java workloads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: Apache KafkaXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P12002400360048006000SE +/- 6.44, N = 3SE +/- 2.03, N = 3SE +/- 3.51, N = 3SE +/- 4.06, N = 3SE +/- 1.20, N = 3SE +/- 1.76, N = 3SE +/- 31.50, N = 3SE +/- 10.11, N = 351085101510250885106511355605611

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100, LosslessXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.33750.6751.01251.351.6875SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 4SE +/- 0.00, N = 41.401.391.391.391.311.301.501.501. (CC) gcc options: -fvisibility=hidden -O2 -lm

Timed ImageMagick Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed ImageMagick Compilation 6.9.0Time To CompileXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P48121620SE +/- 0.15, N = 4SE +/- 0.11, N = 4SE +/- 0.06, N = 4SE +/- 0.07, N = 4SE +/- 0.10, N = 4SE +/- 0.13, N = 4SE +/- 0.10, N = 4SE +/- 0.03, N = 416.0814.4015.3814.0914.7213.2612.6011.49

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: JPEG - Quality: 100Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P714212835SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 4SE +/- 0.07, N = 426.8627.3327.7527.2628.7528.1632.2431.611. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -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.orgMT/s, More Is BetterASTC Encoder 4.7Preset: ThoroughXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20406080100SE +/- 0.01, N = 3SE +/- 0.05, N = 4SE +/- 0.03, N = 4SE +/- 0.03, N = 5SE +/- 0.01, N = 4SE +/- 0.19, N = 5SE +/- 0.05, N = 4SE +/- 0.55, N = 537.0172.3342.9883.3451.3086.9854.4597.941. (CXX) g++ options: -O3 -flto -pthread

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.CXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P100K200K300K400K500KSE +/- 71.37, N = 3SE +/- 441.66, N = 5SE +/- 37.04, N = 4SE +/- 411.20, N = 5SE +/- 130.20, N = 4SE +/- 398.40, N = 5SE +/- 213.44, N = 5SE +/- 726.00, N = 5126171.52246822.52131606.86257384.55135125.15261127.09232012.77455312.091. (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

m-queens

A solver for the N-queens problem with multi-threading support via the OpenMP library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterm-queens 1.2Time To SolveXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P510152025SE +/- 0.031, N = 3SE +/- 0.021, N = 5SE +/- 0.035, N = 3SE +/- 0.028, N = 5SE +/- 0.025, N = 4SE +/- 0.021, N = 6SE +/- 0.030, N = 4SE +/- 0.028, N = 621.03010.84018.1029.28515.2117.88114.2577.4021. (CXX) g++ options: -fopenmp -O2 -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: PUSCH Processor Benchmark, Throughput ThreadXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3129.7129.7129.7129.7129.7129.7129.5129.5MIN: 105.9MIN: 105.9MIN: 105.9MIN: 105.9MIN: 105.9MIN: 105.9MIN: 105.7MIN: 105.71. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -mavx512f -mavx512cd -mavx512bw -mavx512dq -fno-trapping-math -fno-math-errno -ldl

Y-Cruncher

Y-Cruncher is a multi-threaded Pi benchmark capable of computing Pi to trillions of digits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.8.3Pi Digits To Calculate: 500MXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.85071.70142.55213.40284.2535SE +/- 0.010, N = 7SE +/- 0.064, N = 15SE +/- 0.023, N = 15SE +/- 0.074, N = 15SE +/- 0.003, N = 7SE +/- 0.002, N = 8SE +/- 0.008, N = 7SE +/- 0.004, N = 73.7813.1103.7042.9613.2672.3643.3912.659

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.CXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P20K40K60K80K100KSE +/- 313.88, N = 7SE +/- 545.92, N = 15SE +/- 260.16, N = 15SE +/- 560.41, N = 15SE +/- 243.41, N = 12SE +/- 178.23, N = 8SE +/- 328.18, N = 8SE +/- 271.54, N = 635116.5565396.2235234.0666492.4433811.6358128.2463760.4598764.931. (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

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Very FastXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1530456075SE +/- 0.06, N = 4SE +/- 0.33, N = 4SE +/- 0.08, N = 4SE +/- 0.29, N = 4SE +/- 0.08, N = 4SE +/- 0.45, N = 5SE +/- 0.04, N = 4SE +/- 0.36, N = 540.2351.9141.3452.2350.0367.6438.0265.381. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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 DragonXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 0.04, N = 5SE +/- 0.63, N = 6SE +/- 0.11, N = 5SE +/- 1.20, N = 15SE +/- 0.05, N = 5SE +/- 0.13, N = 7SE +/- 0.13, N = 6SE +/- 0.26, N = 875.47125.7583.49129.7591.99152.75112.66199.61MIN: 72.93 / MAX: 78.77MIN: 120.06 / MAX: 135.13MIN: 80.58 / MAX: 88.97MIN: 110.67 / MAX: 151.91MIN: 88.57 / MAX: 98.54MIN: 143.8 / MAX: 168.98MIN: 107.88 / MAX: 117.19MIN: 185.98 / MAX: 216.64

GNU Octave Benchmark

This test profile measures how long it takes to complete several reference GNU Octave files via octave-benchmark. GNU Octave is used for numerical computations and is an open-source alternative to MATLAB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGNU Octave Benchmark 8.4.0Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3691215SE +/- 0.015, N = 6SE +/- 0.096, N = 5SE +/- 0.024, N = 6SE +/- 0.082, N = 5SE +/- 0.015, N = 6SE +/- 0.019, N = 6SE +/- 0.020, N = 6SE +/- 0.019, N = 67.62212.6177.74812.5787.5328.0366.4586.856

Intel Open Image Denoise

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

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 2.2Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-OnlyXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1.13182.26363.39544.52725.659SE +/- 0.00, N = 4SE +/- 0.01, N = 5SE +/- 0.00, N = 4SE +/- 0.04, N = 5SE +/- 0.00, N = 4SE +/- 0.03, N = 6SE +/- 0.00, N = 5SE +/- 0.03, N = 62.263.512.383.652.714.413.315.03

Pennant

Pennant is an application focused on hydrodynamics on general unstructured meshes in 2D. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: leblancbigXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1.33452.6694.00355.3386.6725SE +/- 0.025580, N = 6SE +/- 0.010204, N = 7SE +/- 0.023904, N = 7SE +/- 0.008354, N = 7SE +/- 0.022393, N = 7SE +/- 0.019069, N = 15SE +/- 0.022698, N = 7SE +/- 0.013568, N = 65.9309572.8810995.1141612.5273634.9558652.5298074.1827072.1594591. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi

Intel Open Image Denoise

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

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 2.2Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-OnlyXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1.14532.29063.43594.58125.7265SE +/- 0.00, N = 4SE +/- 0.02, N = 5SE +/- 0.00, N = 4SE +/- 0.02, N = 5SE +/- 0.00, N = 4SE +/- 0.01, N = 6SE +/- 0.00, N = 5SE +/- 0.03, N = 62.273.492.393.662.714.483.325.09

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: CrownXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P306090120150SE +/- 0.05, N = 4SE +/- 0.14, N = 6SE +/- 0.02, N = 4SE +/- 0.48, N = 6SE +/- 0.07, N = 5SE +/- 0.12, N = 6SE +/- 0.05, N = 5SE +/- 0.07, N = 758.42107.6365.33116.6170.05124.7182.97153.16MIN: 55.81 / MAX: 61.43MIN: 102.22 / MAX: 114.8MIN: 60.89 / MAX: 69.91MIN: 107.47 / MAX: 129.72MIN: 63.26 / MAX: 79.13MIN: 115.3 / MAX: 143.82MIN: 79.09 / MAX: 87.11MIN: 143.78 / MAX: 166.08

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100, Highest CompressionXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.78081.56162.34243.12323.904SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.00, N = 63.103.093.103.093.083.073.473.461. (CC) gcc options: -fvisibility=hidden -O2 -lm

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: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.9731.9462.9193.8924.865SE +/- 0.003490, N = 5SE +/- 0.004988, N = 5SE +/- 0.010419, N = 5SE +/- 0.005278, N = 5SE +/- 0.003025, N = 5SE +/- 0.005935, N = 5SE +/- 0.001432, N = 5SE +/- 0.004380, N = 54.3246504.2839904.2748104.1755300.7715060.7146990.8416080.738442MIN: 4.21MIN: 4.18MIN: 4.18MIN: 4.06MIN: 0.811. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

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: Church FacadeXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P15003000450060007500SE +/- 7.18, N = 6SE +/- 7.99, N = 6SE +/- 9.53, N = 6SE +/- 22.13, N = 6SE +/- 14.74, N = 6SE +/- 18.12, N = 6SE +/- 30.70, N = 7SE +/- 20.74, N = 7670768216693677756705672456546931. (CXX) g++ options: -O3

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: Convolution Batch Shapes Auto - Engine: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3691215SE +/- 0.005800, N = 7SE +/- 0.039215, N = 7SE +/- 0.017741, N = 7SE +/- 0.015670, N = 7SE +/- 0.001388, N = 7SE +/- 0.001522, N = 7SE +/- 0.002339, N = 7SE +/- 0.001169, N = 79.1943407.7543108.7510507.0604300.6506180.4381540.5858040.306795MIN: 6.88MIN: 6.74MIN: 7.59MIN: 6.831. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

ACES DGEMM

This is a multi-threaded DGEMM benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1326395265SE +/- 0.04, N = 4SE +/- 0.06, N = 7SE +/- 0.09, N = 5SE +/- 0.04, N = 7SE +/- 0.09, N = 5SE +/- 0.28, N = 7SE +/- 0.08, N = 6SE +/- 0.11, N = 819.7739.3522.5444.6723.5445.9729.1458.251. (CC) gcc options: -O3 -march=native -fopenmp

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.orgMT/s, More Is BetterASTC Encoder 4.7Preset: MediumXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P120240360480600SE +/- 0.03, N = 6SE +/- 0.52, N = 6SE +/- 0.08, N = 6SE +/- 0.77, N = 6SE +/- 0.16, N = 6SE +/- 2.79, N = 6SE +/- 1.17, N = 6SE +/- 1.56, N = 7247.30441.65284.33483.10333.99484.55327.78550.021. (CXX) g++ options: -O3 -flto -pthread

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: 6, LosslessXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.045, N = 6SE +/- 0.027, N = 6SE +/- 0.033, N = 6SE +/- 0.035, N = 6SE +/- 0.013, N = 7SE +/- 0.022, N = 7SE +/- 0.025, N = 7SE +/- 0.031, N = 76.8367.1216.6057.0386.2786.1725.7845.6131. (CXX) g++ options: -O3 -fPIC -lm

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 4KXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P4080120160200SE +/- 0.61, N = 7SE +/- 0.83, N = 5SE +/- 0.40, N = 7SE +/- 0.97, N = 6SE +/- 0.41, N = 7SE +/- 0.25, N = 7SE +/- 0.20, N = 7SE +/- 0.42, N = 7146.10122.09151.72127.55164.25156.52177.86165.501. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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: LionXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P11002200330044005500SE +/- 21.44, N = 6SE +/- 18.24, N = 6SE +/- 15.47, N = 6SE +/- 25.38, N = 6SE +/- 23.47, N = 7SE +/- 25.25, N = 7SE +/- 22.92, N = 8SE +/- 21.09, N = 7521753425188532247314733386138941. (CXX) g++ options: -O3

LAMMPS Molecular Dynamics Simulator

LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: Rhodopsin ProteinXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1224364860SE +/- 0.05, N = 11SE +/- 0.11, N = 9SE +/- 0.04, N = 11SE +/- 0.09, N = 8SE +/- 0.36, N = 15SE +/- 0.33, N = 8SE +/- 0.29, N = 15SE +/- 0.97, N = 1527.6347.5130.8452.2428.7242.2834.3055.501. (CXX) g++ options: -O3 -lm -ldl

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: 10, LosslessXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P1.19632.39263.58894.78525.9815SE +/- 0.005, N = 7SE +/- 0.019, N = 7SE +/- 0.015, N = 7SE +/- 0.028, N = 7SE +/- 0.006, N = 7SE +/- 0.014, N = 7SE +/- 0.008, N = 8SE +/- 0.015, N = 84.9165.3174.8675.3114.9235.0744.4564.5791. (CXX) g++ options: -O3 -fPIC -lm

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 ThreadXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P160320480640800SE +/- 1.64, N = 8SE +/- 3.86, N = 8SE +/- 1.27, N = 8SE +/- 5.02, N = 8SE +/- 1.17, N = 8SE +/- 2.57, N = 8SE +/- 1.34, N = 8SE +/- 1.89, N = 8702.6651.9705.5663.8680.2675.5742.9755.51. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -mavx512f -mavx512cd -mavx512bw -mavx512dq -fno-trapping-math -fno-math-errno -ldl

Primesieve

Primesieve generates prime numbers using a highly optimized sieve of Eratosthenes implementation. Primesieve primarily benchmarks the CPU's L1/L2 cache performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 12.1Length: 1e12Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P246810SE +/- 0.006, N = 7SE +/- 0.004, N = 9SE +/- 0.010, N = 7SE +/- 0.005, N = 10SE +/- 0.019, N = 8SE +/- 0.006, N = 10SE +/- 0.007, N = 8SE +/- 0.008, N = 116.1603.1215.3322.6914.5762.3254.2502.1871. (CXX) g++ options: -O3

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: 6Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.84961.69922.54883.39844.248SE +/- 0.017, N = 9SE +/- 0.029, N = 8SE +/- 0.011, N = 9SE +/- 0.022, N = 9SE +/- 0.022, N = 9SE +/- 0.008, N = 9SE +/- 0.016, N = 9SE +/- 0.009, N = 103.7303.7763.5503.7383.1693.0442.9712.8051. (CXX) g++ options: -O3 -fPIC -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: MG.CXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P50K100K150K200K250KSE +/- 436.97, N = 10SE +/- 1415.96, N = 9SE +/- 382.59, N = 10SE +/- 1298.67, N = 9SE +/- 152.92, N = 9SE +/- 1262.04, N = 9SE +/- 406.29, N = 9SE +/- 1422.07, N = 785910.95172614.1086407.46171436.4783448.70163333.21107321.89220893.781. (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

Parallel BZIP2 Compression

This test measures the time needed to compress a file (FreeBSD-13.0-RELEASE-amd64-memstick.img) using Parallel BZIP2 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterParallel BZIP2 Compression 1.1.13FreeBSD-13.0-RELEASE-amd64-memstick.img CompressionXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.59221.18441.77662.36882.961SE +/- 0.023295, N = 15SE +/- 0.030907, N = 15SE +/- 0.020941, N = 15SE +/- 0.028863, N = 15SE +/- 0.007749, N = 10SE +/- 0.017240, N = 15SE +/- 0.023842, N = 15SE +/- 0.024402, N = 152.6320771.9829322.2877661.7112962.1448571.4692101.8475371.4323621. (CXX) g++ options: -O2 -pthread -lbz2 -lpthread

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: Deconvolution Batch shapes_3d - Engine: CPUXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P0.28440.56880.85321.13761.422SE +/- 0.000508, N = 9SE +/- 0.000806, N = 9SE +/- 0.000927, N = 9SE +/- 0.001240, N = 9SE +/- 0.000805, N = 9SE +/- 0.001274, N = 9SE +/- 0.000392, N = 9SE +/- 0.000902, N = 91.2640700.9450751.1964300.8643581.0774300.7361700.9071610.624250MIN: 1.21MIN: 0.9MIN: 1.16MIN: 0.81MIN: 1.06MIN: 0.66MIN: 0.87MIN: 0.581. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100Xeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P3691215SE +/- 0.00, N = 10SE +/- 0.00, N = 10SE +/- 0.00, N = 10SE +/- 0.00, N = 10SE +/- 0.01, N = 10SE +/- 0.01, N = 10SE +/- 0.00, N = 10SE +/- 0.01, N = 1010.1310.0110.1310.0310.1110.0511.3611.311. (CC) gcc options: -fvisibility=hidden -O2 -lm

Timed CPython Compilation

This test times how long it takes to build the reference Python implementation, CPython, with optimizations and LTO enabled for a release build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed CPython Compilation 3.10.6Build Configuration: DefaultXeon Max 9468Xeon Max 9468 2PXeon Max 9480Xeon Max 9480 2PXeon Platinum 8490HXeon Platinum 8490H 2PXeon Platinum 8592+Xeon Platinum 8592+ 2P51015202519.1218.8118.8518.8219.2018.9317.0816.93

320 Results Shown

WRF
OpenVKL
ASKAP
NWChem
RELION
easyWave
ASKAP
CockroachDB
BRL-CAD
CockroachDB
PostgreSQL:
  100 - 1000 - Read Only - Average Latency
  100 - 1000 - Read Only
TensorFlow
CloverLeaf
NAMD
High Performance Conjugate Gradient
Apache IoTDB:
  500 - 100 - 800 - 400:
    Average Latency
    point/sec
ClickHouse:
  100M Rows Hits Dataset, Third Run
  100M Rows Hits Dataset, Second Run
  100M Rows Hits Dataset, First Run / Cold Cache
OpenSSL:
  RSA4096:
    verify/s
    sign/s
TensorFlow
Timed Linux Kernel Compilation
Xcompact3d Incompact3d
Stockfish
SecureMark
Apache IoTDB:
  800 - 100 - 800 - 100:
    Average Latency
    point/sec
CockroachDB
PyTorch
LuxCoreRender
Timed Gem5 Compilation
RocksDB
LAMMPS Molecular Dynamics Simulator
Blender
Llamafile
Timed Node.js Compilation
ONNX Runtime:
  fcn-resnet101-11 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
DuckDB
easyWave
OpenRadioss
LuxCoreRender
FFmpeg
ONNX Runtime:
  yolov4 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
OpenSSL:
  AES-128-GCM
  AES-256-GCM
  ChaCha20-Poly1305
  SHA512
  ChaCha20
  SHA256
Speedb
FFmpeg:
  libx265 - Video On Demand
  libx265 - Platform
PyTorch
Zstd Compression:
  19 - Decompression Speed
  19 - Compression Speed
oneDNN
John The Ripper
PyTorch
OSPRay
DuckDB
OSPRay:
  particle_volume/pathtracer/real_time
  gravity_spheres_volume/dim_512/scivis/real_time
LuxCoreRender
Timed LLVM Compilation
NAMD
VVenC
Numpy Benchmark
PyTorch
LuxCoreRender
OpenFOAM:
  drivaerFastback, Medium Mesh Size - Execution Time
  drivaerFastback, Medium Mesh Size - Mesh Time
PostgreSQL:
  100 - 1000 - Read Write - Average Latency
  100 - 1000 - Read Write
OpenRadioss:
  Rubber O-Ring Seal Installation
  INIVOL and Fluid Structure Interaction Drop Container
Zstd Compression:
  19, Long Mode - Decompression Speed
  19, Long Mode - Compression Speed
Timed Godot Game Engine Compilation
Xmrig
TensorFlow
ONNX Runtime:
  CaffeNet 12-int8 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  GPT-2 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  T5 Encoder - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
OpenRadioss
QMCPACK
Appleseed
ONNX Runtime:
  ArcFace ResNet-100 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
OSPRay:
  gravity_spheres_volume/dim_512/ao/real_time
  particle_volume/ao/real_time
oneDNN
Speedb
Memcached
DaCapo Benchmark
ONNX Runtime:
  ResNet50 v1-12-int8 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  super-resolution-10 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
PyTorch
ONNX Runtime:
  bertsquad-12 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
OpenVINO:
  Age Gender Recognition Retail 0013 FP16-INT8 - CPU:
    ms
    FPS
ONNX Runtime:
  T5 Encoder - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
Timed CPython Compilation
Appleseed
ONNX Runtime:
  fcn-resnet101-11 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
OpenRadioss
Aircrack-ng
Helsing
DaCapo Benchmark
TensorFlow:
  CPU - 1 - ResNet-50
  CPU - 64 - ResNet-50
Blender
libavif avifenc
OSPRay Studio
TensorFlow
GraphicsMagick
OSPRay Studio
OpenVINO:
  Face Detection FP16-INT8 - CPU:
    ms
    FPS
OSPRay Studio
miniBUDE:
  OpenMP - BM2:
    Billion Interactions/s
    GFInst/s
OSPRay
OSPRay Studio:
  1 - 4K - 32 - Path Tracer - CPU
  1 - 4K - 16 - Path Tracer - CPU
OpenVINO:
  Weld Porosity Detection FP16-INT8 - CPU:
    ms
    FPS
  Noise Suppression Poconet-Like FP16 - CPU:
    ms
    FPS
OSPRay Studio
ONNX Runtime:
  CaffeNet 12-int8 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
OpenVINO:
  Person Vehicle Bike Detection FP16 - CPU:
    ms
    FPS
  Person Re-Identification Retail FP16 - CPU:
    ms
    FPS
  Road Segmentation ADAS FP16-INT8 - CPU:
    ms
    FPS
QuantLib
ONNX Runtime:
  Faster R-CNN R-50-FPN-int8 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  Faster R-CNN R-50-FPN-int8 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
JPEG-XL libjxl
OpenVINO:
  Person Detection FP16 - CPU:
    ms
    FPS
ONNX Runtime:
  GPT-2 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
OpenVINO:
  Machine Translation EN To DE FP16 - CPU:
    ms
    FPS
Node.js V8 Web Tooling Benchmark
OpenVINO:
  Vehicle Detection FP16-INT8 - CPU:
    ms
    FPS
ONNX Runtime:
  yolov4 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
  bertsquad-12 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
OpenVINO:
  Face Detection Retail FP16-INT8 - CPU:
    ms
    FPS
ONNX Runtime:
  ArcFace ResNet-100 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
OpenVINO:
  Handwritten English Recognition FP16-INT8 - CPU:
    ms
    FPS
GraphicsMagick:
  Sharpen
  Noise-Gaussian
  Resizing
ONNX Runtime:
  ResNet50 v1-12-int8 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
GraphicsMagick:
  HWB Color Space
  Enhanced
ONNX Runtime:
  super-resolution-10 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
GraphicsMagick
RocksDB
Blender
PyTorch:
  CPU - 256 - ResNet-50
  CPU - 64 - ResNet-50
CloverLeaf
Llamafile
srsRAN Project
RawTherapee
GPAW
VVenC
Coremark
miniFE
Timed PHP Compilation
John The Ripper
7-Zip Compression:
  Decompression Rating
  Compression Rating
LuxCoreRender
Liquid-DSP
Timed Linux Kernel Compilation
WebP Image Encode
TensorFlow
JPEG-XL libjxl
Primesieve
Graph500:
  26:
    sssp median_TEPS
    sssp max_TEPS
    bfs median_TEPS
    bfs max_TEPS
Liquid-DSP
srsRAN Project
libavif avifenc
ASKAP:
  tConvolve MPI - Gridding
  tConvolve MPI - Degridding
OpenRadioss
Liquid-DSP:
  64 - 256 - 512
  1 - 256 - 32
Timed Wasmer Compilation
PyTorch
Appleseed
ASTC Encoder
Liquid-DSP
uvg266
ASTC Encoder
Liquid-DSP
SVT-AV1
QuantLib
Liquid-DSP
nginx
Liquid-DSP
Blender
NAS Parallel Benchmarks
John The Ripper:
  bcrypt
  Blowfish
Liquid-DSP:
  256 - 256 - 57
  256 - 256 - 32
uvg266
Liquid-DSP
Google SynthMark
Liquid-DSP
Blender
TensorFlow
DaCapo Benchmark
Embree
GROMACS
SVT-AV1
TensorFlow
ASKAP:
  tConvolve OpenMP - Degridding
  tConvolve OpenMP - Gridding
Kvazaar:
  Bosphorus 4K - Slow
  Bosphorus 4K - Medium
SVT-AV1
uvg266
TensorFlow
miniBUDE:
  OpenMP - BM1:
    Billion Interactions/s
    GFInst/s
TensorFlow
Blender
Y-Cruncher
OpenFOAM:
  drivaerFastback, Small Mesh Size - Execution Time
  drivaerFastback, Small Mesh Size - Mesh Time
NAS Parallel Benchmarks
Kvazaar
Algebraic Multi-Grid Benchmark
Xcompact3d Incompact3d
Timed FFmpeg Compilation
Intel Open Image Denoise
oneDNN
x265
uvg266
LULESH
oneDNN
TensorFlow
NAS Parallel Benchmarks
Pennant
Kvazaar
uvg266
DaCapo Benchmark
JPEG-XL libjxl
Timed Mesa Compilation
PyBench
DaCapo Benchmark
WebP Image Encode
Timed ImageMagick Compilation
JPEG-XL libjxl
ASTC Encoder
NAS Parallel Benchmarks
m-queens
srsRAN Project
Y-Cruncher
NAS Parallel Benchmarks
Kvazaar
Embree
GNU Octave Benchmark
Intel Open Image Denoise
Pennant
Intel Open Image Denoise
Embree
WebP Image Encode
oneDNN
Google Draco
oneDNN
ACES DGEMM
ASTC Encoder
libavif avifenc
SVT-AV1
Google Draco
LAMMPS Molecular Dynamics Simulator
libavif avifenc
srsRAN Project
Primesieve
libavif avifenc
NAS Parallel Benchmarks
Parallel BZIP2 Compression
oneDNN
WebP Image Encode
Timed CPython Compilation