AMD EPYC 4004 CPU Review Benchmarks

Benchmarks preparing for a future article on Phoronix.com on launch day by Michael Larabel.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2405204-PTS-EPYC400480
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
C++ Boost Tests 4 Tests
Chess Test Suite 2 Tests
Timed Code Compilation 10 Tests
C/C++ Compiler Tests 18 Tests
Compression Tests 3 Tests
CPU Massive 34 Tests
Creator Workloads 26 Tests
Cryptography 4 Tests
Database Test Suite 7 Tests
Encoding 9 Tests
Fortran Tests 6 Tests
Game Development 6 Tests
HPC - High Performance Computing 20 Tests
Imaging 3 Tests
Java 3 Tests
Common Kernel Benchmarks 4 Tests
Machine Learning 6 Tests
Molecular Dynamics 7 Tests
MPI Benchmarks 7 Tests
Multi-Core 50 Tests
NVIDIA GPU Compute 4 Tests
Intel oneAPI 7 Tests
OpenMPI Tests 12 Tests
Programmer / Developer System Benchmarks 13 Tests
Python 2 Tests
Raytracing 2 Tests
Renderers 6 Tests
Scientific Computing 10 Tests
Software Defined Radio 2 Tests
Server 13 Tests
Server CPU Tests 25 Tests
Single-Threaded 3 Tests
Texture Compression 2 Tests
Video Encoding 9 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
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

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
Condense Test Profiles With Multiple Version Results Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Xeon E-2488
April 01
  1 Day, 22 Hours, 13 Minutes
EPYC 4244P
May 12
  1 Day, 16 Hours, 20 Minutes
EPYC 4344P
May 08
  1 Day, 11 Hours, 13 Minutes
EPYC 4364P
April 28
  1 Day, 17 Hours, 57 Minutes
EPYC 4464P
May 17
  1 Day, 11 Hours, 26 Minutes
EPYC 4484PX
May 10
  1 Day, 14 Hours, 16 Minutes
EPYC 4564P
April 27
  1 Day, 8 Hours, 35 Minutes
EPYC 4584PX
May 15
  1 Day, 9 Hours, 50 Minutes
Invert Hiding All Results Option
  1 Day, 13 Hours, 59 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):


AMD EPYC 4004 CPU Review BenchmarksProcessorMotherboardChipsetMemoryDiskGraphicsNetworkAudioOSKernelDesktopDisplay ServerCompilerFile-SystemScreen ResolutionXeon E-2488EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXIntel Xeon E E-2488 @ 3.20GHz (8 Cores / 16 Threads)Supermicro Super Server X13SCL-F v0123456789 (1.1 BIOS)Intel Raptor Lake-S PCH2 x 32GB DRAM-4800MT/s Micron MTC20C2085S1EC48BA13201GB Micron_7450_MTFDKCC3T2TFS + 960GB Micron_7450_MTFDKBA960TFRASPEED2 x Intel I210Ubuntu 24.046.8.0-11-generic (x86_64)GNOME Shell 45.3X Server 1.21.1.11GCC 13.2.0ext41920x1080AMD EPYC 4244P 6-Core @ 5.17GHz (6 Cores / 12 Threads)Supermicro AS-3015A-I H13SAE-MF v1.00 (1.1 BIOS)AMD Device 14d82 x 32GB DRAM-5200MT/sAMD Rembrandt Radeon HD AudioAMD EPYC 4344P 8-Core @ 5.39GHz (8 Cores / 16 Threads)AMD EPYC 4364P 8-Core @ 5.57GHz (8 Cores / 16 Threads)AMD EPYC 4464P 12-Core @ 5.48GHz (12 Cores / 24 Threads)AMD EPYC 4484PX 12-Core @ 5.66GHz (12 Cores / 24 Threads)AMD EPYC 4564P 16-Core @ 5.88GHz (16 Cores / 32 Threads)AMD EPYC 4584PX 16-Core @ 5.76GHz (16 Cores / 32 Threads)OpenBenchmarking.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-fxIygj/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-fxIygj/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 Disk Details- Xeon E-2488: NONE / errors=remount-ro,relatime,rw,stripe=64 / Block Size: 4096Processor Details- Xeon E-2488: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x122 - Thermald 2.5.6- EPYC 4244P: Scaling Governor: amd-pstate-epp performance (EPP: performance) - CPU Microcode: 0xa601203- EPYC 4344P: Scaling Governor: amd-pstate-epp performance (EPP: performance) - CPU Microcode: 0xa601203- EPYC 4364P: Scaling Governor: amd-pstate-epp performance (EPP: performance) - CPU Microcode: 0xa601203- EPYC 4464P: Scaling Governor: amd-pstate-epp performance (EPP: performance) - CPU Microcode: 0xa601203- EPYC 4484PX: Scaling Governor: amd-pstate-epp performance (EPP: performance) - CPU Microcode: 0xa601203- EPYC 4564P: Scaling Governor: amd-pstate-epp performance (EPP: performance) - CPU Microcode: 0xa601203- EPYC 4584PX: Scaling Governor: amd-pstate-epp performance (EPP: performance) - CPU Microcode: 0xa601203Java Details- OpenJDK Runtime Environment (build 21.0.2+13-Ubuntu-2)Python Details- Python 3.12.2Security Details- Xeon E-2488: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of 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 - EPYC 4244P: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Vulnerable: Safe RET no microcode + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - EPYC 4344P: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Vulnerable: Safe RET no microcode + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - EPYC 4364P: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Vulnerable: Safe RET no microcode + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - EPYC 4464P: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Vulnerable: Safe RET no microcode + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - EPYC 4484PX: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Vulnerable: Safe RET no microcode + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - EPYC 4564P: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Vulnerable: Safe RET no microcode + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - EPYC 4584PX: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Vulnerable: Safe RET no microcode + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Xeon E-2488EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXResult OverviewPhoronix Test Suite100%176%252%328%404%NAMDminiBUDEJohn The RipperOpenSSLPrimesieveOSPRay StudioStockfishOpenVINOACES DGEMMPostgreSQLCpuminer-OptASTC Encoderm-queensAircrack-ngStress-NGJava JMHBlenderCoremarkQuantLibChaos Group V-RAYHelsingEmbreeLuxCoreRenderIntel Open Image DenoiseOpenVKLASKAPOSPRayMemcachedBRL-CADTimed Node.js CompilationTimed Linux Kernel CompilationParallel BZIP2 Compression7-Zip CompressionTimed LLVM CompilationTimed Godot Game Engine CompilationSVT-HEVCnginxAppleseedGROMACSTimed FFmpeg CompilationRocksDBuvg266oneDNNSPECFEM3DTimed Mesa CompilationSpeedbTimed Gem5 CompilationGraphicsMagickKvazaarHeFFTe - Highly Efficient FFT for ExascaleScyllaDBSVT-VP9LAMMPS Molecular Dynamics SimulatorNAS Parallel BenchmarksJPEG-XL libjxlCockroachDBCloverLeafY-Cruncherlibavif avifencPennantsrsRAN Projectx265DaCapo BenchmarkTensorFlowSVT-AV1Zstd CompressionTimed PHP CompilationOpenFOAMVVenCGPAWMariaDBOpenRadiossXcompact3d Incompact3dPyBenchONNX RuntimeTimed Wasmer CompilationDuckDBGoogle DracoClickHouseZ3 Theorem ProverLiquid-DSPApache IoTDBPyTorchRenaissanceTimed CPython CompilationNumpy BenchmarkGNU Octave BenchmarkMariaDB mariadb-slapFFmpeg

Xeon E-2488EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXPer Watt Result OverviewPhoronix Test Suite100%163%225%288%350%NAMDminiBUDEJohn The Ripper7-Zip CompressionHeFFTe - Highly Efficient FFT for ExascaleACES DGEMMASKAPOpenSSLJava JMHEmbreeIntel Open Image DenoiseAircrack-ngChaos Group V-RAYNumpy BenchmarkCoremarkOpenVKLCpuminer-OptStockfishClickHouseOSPRayStress-NGGROMACSMemcachednginxASTC EncodersrsRAN ProjectKvazaaruvg266LAMMPS Molecular Dynamics SimulatorSVT-HEVCx265SVT-VP9TensorFlowNAS Parallel BenchmarksPyTorchLuxCoreRenderLiquid-DSPSVT-AV1JPEG-XL libjxlBRL-CADQuantLibRocksDBScyllaDBMariaDB mariadb-slapSpeedbGraphicsMagickFFmpegCockroachDBVVenCMariaDBP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.M

AMD EPYC 4004 CPU Review Benchmarksjohn-the-ripper: WPA PSKopenvino: Weld Porosity Detection FP16 - CPUopenvino: Face Detection FP16 - CPUopenvino: Handwritten English Recognition FP16 - CPUopenssl: RSA4096srsran: PDSCH Processor Benchmark, Throughput Totalopenvino: Person Re-Identification Retail FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUopenvino: Handwritten English Recognition FP16-INT8 - CPUjohn-the-ripper: MD5john-the-ripper: HMAC-SHA512namd: ATPase with 327,506 Atomsopenvino: Person Re-Identification Retail FP16 - CPUopenvino: Face Detection Retail FP16 - CPUopenvino: Road Segmentation ADAS FP16 - CPUopenvino: Machine Translation EN To DE FP16 - CPUopenvino: Vehicle Detection FP16 - CPUpgbench: 100 - 800 - Read Only - Average Latencypgbench: 100 - 800 - Read Onlyminibude: OpenMP - BM1minibude: OpenMP - BM1openvino: Face Detection Retail FP16 - CPUminibude: OpenMP - BM2minibude: OpenMP - BM2openvino: Person Vehicle Bike Detection FP16 - CPUopenssl: AES-128-GCMopenssl: ChaCha20-Poly1305openssl: AES-256-GCMopenssl: ChaCha20namd: STMV with 1,066,628 Atomsopenvino: Noise Suppression Poconet-Like FP16 - CPUopenvino: Vehicle Detection FP16-INT8 - CPUpgbench: 100 - 500 - Read Only - Average Latencypgbench: 100 - 500 - Read Onlyopenvino: Vehicle Detection FP16-INT8 - CPUopenvino: Person Detection FP32 - CPUopenvino: Person Detection FP16 - CPUprimesieve: 1e12openvino: Face Detection Retail FP16-INT8 - CPUprimesieve: 1e13cpuminer-opt: Triple SHA-256, Onecoincpuminer-opt: Quad SHA-256, Pyriteospray-studio: 1 - 1080p - 16 - Path Tracer - CPUospray-studio: 2 - 1080p - 16 - Path Tracer - CPUluxcorerender: Danish Mood - CPUospray-studio: 3 - 1080p - 16 - Path Tracer - CPUstress-ng: Vector Floating Pointonednn: IP Shapes 3D - CPUospray: gravity_spheres_volume/dim_512/scivis/real_timecompress-7zip: Decompression Ratingluxcorerender: LuxCore Benchmark - CPUopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUopenvino: Road Segmentation ADAS FP16-INT8 - CPUopenssl: SHA256rocksdb: Rand Readstockfish: Chess Benchmarkospray: gravity_spheres_volume/dim_512/ao/real_timeopenvino: Weld Porosity Detection FP16 - CPUospray-studio: 2 - 1080p - 1 - Path Tracer - CPUospray-studio: 1 - 1080p - 1 - Path Tracer - CPUospray-studio: 3 - 1080p - 1 - Path Tracer - CPUopenvino: Face Detection FP16 - CPUcpuminer-opt: LBC, LBRY Creditsastcenc: Mediumblender: Barbershop - CPU-Onlyastcenc: Fastcpuminer-opt: Blake-2 Sastcenc: Thoroughopenvino: Weld Porosity Detection FP16-INT8 - CPUm-queens: Time To Solveembree: Pathtracer ISPC - Crownopenvino: Handwritten English Recognition FP16 - CPUospray: particle_volume/scivis/real_timeaircrack-ng: openvino: Face Detection FP16-INT8 - CPUospray: particle_volume/ao/real_timeastcenc: Very Thoroughjava-jmh: Throughputastcenc: Exhaustiveopenssl: RSA4096cpuminer-opt: Skeincoinaskap: tConvolve OpenMP - Griddingopenssl: SHA512stress-ng: Floating Pointaskap: tConvolve OpenMP - Degriddingblender: Classroom - CPU-Onlyonednn: IP Shapes 1D - CPUsvt-hevc: 1 - Bosphorus 4Kcpuminer-opt: Deepcoincpuminer-opt: scryptblender: BMW27 - CPU-Onlystress-ng: Vector Shufflehelsing: 12 digitospray: gravity_spheres_volume/dim_512/pathtracer/real_timecpuminer-opt: Ringcoinospray-studio: 3 - 1080p - 32 - Path Tracer - CPUdacapobench: Apache Tomcatblender: Junkshop - CPU-Onlycoremark: CoreMark Size 666 - Iterations Per Secondcpuminer-opt: Garlicoinquantlib: Multi-Threadedstress-ng: Fused Multiply-Addstress-ng: AVX-512 VNNIblender: Pabellon Barcelona - CPU-Onlyospray-studio: 2 - 1080p - 32 - Path Tracer - CPUblender: Fishy Cat - CPU-Onlycpuminer-opt: Magiv-ray: CPUspeedb: Rand Readsvt-hevc: 1 - Bosphorus 1080pospray-studio: 1 - 1080p - 32 - Path Tracer - CPUstress-ng: Wide Vector Mathcpuminer-opt: Myriad-Groestlgraphics-magick: Sharpenhelsing: 14 digitopenvino: Road Segmentation ADAS FP16-INT8 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUjohn-the-ripper: Blowfishjohn-the-ripper: bcryptgraphics-magick: Enhancedappleseed: Disney Materialopenvino: Machine Translation EN To DE FP16 - CPUmemcached: 1:100uvg266: Bosphorus 4K - Mediumbuild-linux-kernel: allmodconfigluxcorerender: DLSC - CPUoidn: RTLightmap.hdr.4096x4096 - CPU-Onlytensorflow: CPU - 512 - AlexNetopenvkl: vklBenchmarkCPU ISPCembree: Pathtracer ISPC - Asian Dragon Objopenvino: Handwritten English Recognition FP16-INT8 - CPUoidn: RT.ldr_alb_nrm.3840x2160 - CPU-Onlyoidn: RT.hdr_alb_nrm.3840x2160 - CPU-Onlyembree: Pathtracer ISPC - Asian Dragonkvazaar: Bosphorus 4K - Mediumbuild-llvm: Ninjagraphics-magick: Swirltensorflow: CPU - 256 - AlexNetdacapobench: Zxing 1D/2D Barcode Image Processingluxcorerender: Orange Juice - CPUonednn: Recurrent Neural Network Training - CPUbrl-cad: VGR Performance Metriconednn: Recurrent Neural Network Inference - CPUrocksdb: Read While Writingbuild-nodejs: Time To Compilecompress-pbzip2: FreeBSD-13.0-RELEASE-amd64-memstick.img Compressionuvg266: Bosphorus 4K - Super Fastsvt-hevc: 7 - Bosphorus 1080psvt-hevc: 7 - Bosphorus 4Konednn: Deconvolution Batch shapes_1d - CPUmemcached: 1:10npb: SP.Bonednn: Deconvolution Batch shapes_3d - CPUnginx: 1000build-godot: Time To Compileuvg266: Bosphorus 4K - Very Fastappleseed: Emilynginx: 500openvino: Road Segmentation ADAS FP16 - CPUuvg266: Bosphorus 4K - Ultra Fastbuild-llvm: Unix Makefilesspeedb: Read While Writingbuild-linux-kernel: defconfigkvazaar: Bosphorus 4K - Super Fastopenfoam: drivaerFastback, Small Mesh Size - Execution Timenginx: 200jpegxl: PNG - 100specfem3d: Tomographic Modeljpegxl: JPEG - 100npb: BT.Cgromacs: MPI CPU - water_GMX50_barecompress-7zip: Compression Ratingbuild-ffmpeg: Time To Compilespecfem3d: Homogeneous Halfspacesvt-vp9: Visual Quality Optimized - Bosphorus 1080pkvazaar: Bosphorus 4K - Very Fastopenvino: Person Vehicle Bike Detection FP16 - CPUonnx: super-resolution-10 - CPU - Parallelopenvino: Vehicle Detection FP16 - CPUx265: Bosphorus 4Kuvg266: Bosphorus 1080p - Super Fastkvazaar: Bosphorus 4K - Ultra Fastbuild-mesa: Time To Compilespecfem3d: Layered Halfspaceonednn: Convolution Batch Shapes Auto - CPUbuild-gem5: Time To Compilesvt-av1: Preset 8 - Bosphorus 4Ktensorflow: CPU - 64 - AlexNetspecfem3d: Mount St. Helenssvt-hevc: 10 - Bosphorus 1080ppgbench: 100 - 800 - Read Write - Average Latencypgbench: 100 - 800 - Read Writespecfem3d: Water-layered Halfspacecompress-zstd: 8 - Compression Speedcompress-zstd: 8, Long Mode - Compression Speedavifenc: 6graphics-magick: Noise-Gaussiansvt-vp9: Visual Quality Optimized - Bosphorus 4Kuvg266: Bosphorus 1080p - Very Fastsvt-vp9: VMAF Optimized - Bosphorus 1080pkvazaar: Bosphorus 1080p - Super Fastsvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080puvg266: Bosphorus 1080p - Mediumscylladb: Writesuvg266: Bosphorus 1080p - Ultra Fastopenvino: Person Detection FP32 - CPUopenvino: Noise Suppression Poconet-Like FP16 - CPUluxcorerender: Rainbow Colors and Prism - CPUy-cruncher: 1Bopenvino: Person Detection FP16 - CPUavifenc: 0cockroach: KV, 95% Reads - 128onnx: ArcFace ResNet-100 - CPU - Parallelsvt-hevc: 10 - Bosphorus 4Kkvazaar: Bosphorus 1080p - Mediumappleseed: Material Testerpennant: leblancbigliquid-dsp: 16 - 256 - 512cockroach: KV, 50% Reads - 128tensorflow: CPU - 32 - GoogLeNetcockroach: KV, 60% Reads - 128pgbench: 100 - 500 - Read Writepgbench: 100 - 500 - Read Write - Average Latencycloverleaf: clover_bmtensorflow: CPU - 64 - GoogLeNetonnx: fcn-resnet101-11 - CPU - Parallelkvazaar: Bosphorus 1080p - Very Fastlammps: Rhodopsin Proteinnpb: EP.Copenradioss: INIVOL and Fluid Structure Interaction Drop Containernpb: LU.Csvt-av1: Preset 4 - Bosphorus 4Ktensorflow: CPU - 512 - GoogLeNetsvt-vp9: PSNR/SSIM Optimized - Bosphorus 4Kkvazaar: Bosphorus 1080p - Ultra Fasttensorflow: CPU - 256 - GoogLeNetpennant: sedovbigtensorflow: CPU - 512 - ResNet-50y-cruncher: 500Monnx: ResNet50 v1-12-int8 - CPU - Parallelsvt-vp9: VMAF Optimized - Bosphorus 4Kavifenc: 2tensorflow: CPU - 256 - ResNet-50tensorflow: CPU - 64 - ResNet-50tensorflow: CPU - 32 - ResNet-50tensorflow: CPU - 32 - AlexNetopenradioss: Bird Strike on Windshieldsvt-av1: Preset 12 - Bosphorus 4Kvvenc: Bosphorus 4K - Fastavifenc: 6, Losslessonnx: bertsquad-12 - CPU - Parallelonnx: yolov4 - CPU - Parallelgraphics-magick: Resizingmariadb: oltp_write_only - 64ospray: particle_volume/pathtracer/real_timevvenc: Bosphorus 4K - Fastercompress-zstd: 12 - Compression Speedrocksdb: Read Rand Write Randjpegxl: PNG - 90jpegxl: JPEG - 90onnx: CaffeNet 12-int8 - CPU - Parallelopenvino: Face Detection Retail FP16-INT8 - CPUnpb: SP.Cincompact3d: input.i3d 129 Cells Per Directionsvt-av1: Preset 8 - Bosphorus 1080pbuild-php: Time To Compilespeedb: Read Rand Write Randgpaw: Carbon Nanotubesvt-av1: Preset 4 - Bosphorus 1080pmariadb: oltp_read_write - 64srsran: PUSCH Processor Benchmark, Throughput Totalvvenc: Bosphorus 1080p - Fastdacapobench: Apache Lucene Search Enginecompress-zstd: 19, Long Mode - Compression Speedcompress-zstd: 19 - Compression Speedsvt-av1: Preset 12 - Bosphorus 1080px265: Bosphorus 1080ponnx: T5 Encoder - CPU - Standardliquid-dsp: 16 - 256 - 32openfoam: drivaerFastback, Medium Mesh Size - Mesh Timepybench: Total For Average Test Timesopenfoam: drivaerFastback, Medium Mesh Size - Execution Timemariadb: oltp_update_non_index - 64pytorch: CPU - 64 - ResNet-50build-wasmer: Time To Compileopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUopenradioss: Bumper Beamdraco: Church Facadeduckdb: IMDBdacapobench: H2 Database Enginemariadb: oltp_update_index - 64clickhouse: 100M Rows Hits Dataset, Second Rungraphics-magick: HWB Color Spaceopenvino: Weld Porosity Detection FP16-INT8 - CPUincompact3d: input.i3d 193 Cells Per Directionopenvino: Face Detection FP16-INT8 - CPUopenradioss: Chrysler Neon 1Mapache-iotdb: 800 - 100 - 800 - 100clickhouse: 100M Rows Hits Dataset, Third Runonnx: T5 Encoder - CPU - Parallelapache-iotdb: 800 - 100 - 800 - 100clickhouse: 100M Rows Hits Dataset, First Run / Cold Cachevvenc: Bosphorus 1080p - Fasterz3: 2.smt2build-python: Defaultapache-iotdb: 800 - 100 - 800 - 400apache-iotdb: 800 - 100 - 800 - 400srsran: PDSCH Processor Benchmark, Throughput Threadopenfoam: drivaerFastback, Small Mesh Size - Mesh Timeonnx: GPT-2 - CPU - Standardapache-iotdb: 500 - 100 - 800 - 100onnx: Faster R-CNN R-50-FPN-int8 - CPU - Parallelapache-iotdb: 800 - 100 - 500 - 100apache-iotdb: 500 - 100 - 800 - 400apache-iotdb: 500 - 100 - 800 - 400apache-iotdb: 800 - 100 - 500 - 100apache-iotdb: 500 - 100 - 800 - 100liquid-dsp: 1 - 256 - 32tensorflow: CPU - 1 - ResNet-50apache-iotdb: 800 - 100 - 500 - 400apache-iotdb: 800 - 100 - 500 - 400apache-iotdb: 500 - 100 - 500 - 100apache-iotdb: 500 - 100 - 500 - 100apache-iotdb: 500 - 100 - 500 - 400dacapobench: PMD Source Code Analyzertensorflow: CPU - 1 - GoogLeNetrenaissance: Savina Reactors.IOapache-iotdb: 500 - 100 - 500 - 400pytorch: CPU - 1 - ResNet-50numpy: srsran: PUSCH Processor Benchmark, Throughput Threadffmpeg: libx265 - Video On Demandffmpeg: libx265 - Platformffmpeg: libx265 - Uploadmysqlslap: 128octave-benchmark: mysqlslap: 64onnx: GPT-2 - CPU - Parallelffmpeg: libx265 - Liveliquid-dsp: 1 - 256 - 512build-python: Released Build, PGO + LTO Optimizedonnx: T5 Encoder - CPU - Parallelonnx: T5 Encoder - CPU - Standardonnx: Faster R-CNN R-50-FPN-int8 - CPU - Parallelonnx: CaffeNet 12-int8 - CPU - Parallelonnx: ResNet50 v1-12-int8 - CPU - Parallelonnx: ResNet50 v1-12-int8 - CPU - Standardonnx: ResNet50 v1-12-int8 - CPU - Standardonnx: ArcFace ResNet-100 - CPU - Parallelonnx: GPT-2 - CPU - Parallelonnx: GPT-2 - CPU - Standardonnx: bertsquad-12 - CPU - Parallelonnx: super-resolution-10 - CPU - Parallelonnx: fcn-resnet101-11 - CPU - Parallelonnx: yolov4 - CPU - Parallelsvt-av1: Preset 13 - Bosphorus 4Kopenradioss: Cell Phone Drop Testopenradioss: Rubber O-Ring Seal Installationheffte: r2c - FFTW - double - 128heffte: r2c - FFTW - float - 128askap: tConvolve MPI - Griddingaskap: tConvolve MPI - Degriddinglammps: 20k Atomsmt-dgemm: Sustained Floating-Point RateXeon E-2488EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PX29560265.832.65150.643000.22908.518.227499.47174.92919791407210000.65977437.969.76101.2133.74249.021.85243196212.452311.310816.17313.73712.550401.42648127256802380026302056193249190346518558000.20565523.0215.950.899555873500.8629.7129.8321.0951559.06245.674388472652741514422751.454892928846.067.596952.63851615271.6918507.31219.861233655726347512260168982912.7349630.052474244128841506.636037.2689.14001281.54205.93215616012.23071022.1072.85213.388353.075.8864035441.70710.415.931611.463936765943472.2680.8741195618.8139332556.2244293971604740.284282.18336.942.914082.263294.49119.86123.519138.294.0803.795391437.47954476687173.17430017.4605831356.3137214.429762395.571125583.01411.1582403165.00278.0920060578244269.1681236576480.154843.7236437.41436.351.00188411878749174.948967118.453048853.566.621508.0422.130.21192.2828814.549945.700.420.4216.57579.28697.937130189.1410533.582931.951941271548.711942760915.9376.61248620.17168.3947.986.853183072373.8712750.974.9479287132.22364.42218.97323.27059392582.2939.4824.69719.3922548837108.33828.64242.9447998692.9513.65052.89378166313.48428326.361.4349499050.05167.114929508221.3323.059.9475.880316.0419.4896.3339.5932.515144.5395371756.38053468.37836.140173.7853.287162292383.6423.19834486135.416735242485.6504.16.2607059.5689.31289.26139.72295.6936.59104832115.40134.5415.219.3533.891133.99133.56783987.523.0436101.8350.47173.40854132.8676515927000059866.678.7963021.44241211.79164.5878.171.042812103.047.2361470.81538.4240491.564.34876.1383.19197.5176.8455.8964722.9814.926214.54779.8561.84323.0723.1023.19160.52259.88105.2804.9758.96911.53728.09421244296034157.96911.768170.6197653331.81432.506557.4085.0711387.0519.7521490120.53354.4492120915228.89915.9181864181068.915.46331138.9116.3548.04284.33169.620711050000225.954914152260.988122006832.2538.1340.4153.90560799.5752128187089249.503157.7873.7203613384.09983.54137.92249.90161.31356035278234.3940.06763.07616.8556494499524.13809.635.18608126.674119.4941.009377.24382.486488636461842388625580986323666713.71290.146265554980.565702839857760197112251.333695.4275.1461.49675.48153.658.7558.7528.895746.4621128123.284178.1520255000197.9216.198765.9078624.38581.793474.660404.46170224.10643.39678.108647.8908086.687313.1792960.856123.593104.61493.26177.7426.478785.00626199.095146.337.5794.20411063566531.485.25272.465652.66218.25.1514005.46301.931568333892176671.31921773.962.98146.6263.15364.222.52531705119.094477.3571330.81478.87219.155603.92982255295603542981718384060321200497071748430.38956697.696.191.207414485644.5044.4044.1716.8271804.90205.202327102072050055503391.465844220773.704.494062.91951678681.6817920.75187.281322073691351334146166145172.9501011.27293028993430761.786265.1058.07531244.70147.4762538707.51361040.1571.97912.334722.013.4476835520.72910.173.455771.032728953646655.5070.6330144386.1133902825.3142752754603175.544370.02351.062.606382.253247.54120.39136.836403.875.1783.496991364.321135018628186.24372162.8214211243.8832626.822079544.971008567.36421.3697307170.00248.8517273531463729.1296023307368.194487.2823504.26521.310.41176211765438210.11137663.291567216.546.531730.8792.060.17176.6125312.200119.860.360.3614.35769.59815.092106174.0613803.303148.811703681615.381865521916.4687.61730520.38135.9943.575.214621651621.6711155.645.5525466910.93415.39719.29341.30426270866.4227.2324.59840.9222400243125.99530.10307.0078875062.5112.84860.37583091512.69622331.461.2538293353.88077.234678151173.3324.226.6064.292110.9417.6088.4441.1135.707166.9705222107.93455540.89236.700160.2862.572651918279.5731.44425448156.906601201500.0482.96.7986050.5184.36208.12126.08213.4432.6378004105.2390.028.498.8831.31990.50136.63163936.219.837486.7947.17191.20632641.7085115643333349645.074.9351927.03160915.82077.2373.770.959441101.806.8381168.35560.3126620.694.54073.5863.76167.3973.3659.0226925.3614.133264.14660.9064.14225.4325.5025.70139.36291.7998.4504.54210.0249.154586.78226202208039132.40010.494164.4178179529.65230.185583.9833.2710028.7421.7994226119.76063.2981806410237.23616.5431331411206.015.538282910.218.0449.56680.58152.179546136667297.171385502944.551716269734.8842.1110.32140.765160115.6432394142647211.232275.7589.8703664393.131126.52136.58214.00147.27856689052195.6437.62849.32218.85156122243529.441073.639.156514134.446117.1446.564976.69406.896251824562348881635413785027233315.61295.346245919579.645792524557190648129852.363288.4275.6258.49773.69179.860.2060.2329.175175.6941009118.958162.3818426333214.3756.789316.5726021.47461.712573.785313.43791291.70550.41268.404087.43468109.23515.55841042.41147.48497.60098.85157.4827.827176.43825963.544723.736.8383.92016884197693.786.90368.267335.27583.74.0718120.38395.1922193331170726671.69120978.602.37171.6374.79436.411.92941475025.851646.2631659.06647.13125.885790.6312904662694047254230173110425587253665727746070.48570890.994.750.904553371838.4652.2452.3512.9372329.23158.448499003229338334386902.084467428866.884.516293.79488884092.3325613.29237.301740271281067959360218443433.8169111.51220121762575579.078323.6976.0119942.29192.6587731309.86741356.6454.38616.091621.704.5165046525.77613.364.525521.356338412521187.6910.8313190537.1175803122.2956443575834192.473420.87267.861.979322.944280.91157.91104.748468.863.9664.572221806.87857436485141.77491021.6860411637.7242668.629040929.241315000.02323.6873858130.42328.70225056944301111.9072930402835.325929.5630386.15516.830.41234032340749160.01407753.442056776.098.351319.9322.670.22238.8932715.635520.220.470.4718.394012.10644.770137235.4710414.302514.662217671310.592489854743.1115.70185626.04173.1955.663.934452163987.0111203.833.9488884832.82330.82424.62264.67098991569.8723.2731.30669.644318965098.29137.70275.0986597227.4216.35847.69474772516.27722976.531.52210424842.93559.630895207198.4230.145.0489.19769.1316.77116.2150.7128.617133.4743503627.79202438.61545.951204.8351.591441038351.9124.78532280130.204350609525.1532.35.4367360.97108.21237.22161.51246.2844.4393342138.9576.528.8311.5926.15976.36110.45582706.226.8609107.4762.71146.66750635.4791620627000063986.292.8566756.93954912.64474.5991.211.12810123.898.7151489.70494.7940284.235.56390.1374.71217.5290.1950.8870730.4911.861314.22970.3452.56430.5030.6431.00171.59238.13121.4955.5548.26212.32399.22219250260139158.58812.801177.9224748135.08435.604672.3453.3510088.5020.3888378142.83954.0382222556203.43219.5511660601469.818.300239110.518.3528.22786.78143.501718020000262.838735322781.236219747743.8936.9010.29124.704988101.9691992173080230.372495.8785.3411229299.02967.20122.33233.30138.37263086247213.4743.83347.72716.8263954176465.511071.033.330642134.740105.7049.810568.58350.007120389369342377702935715225166717.45260.607010703071.666420103663451680115561.883060.5250.9468.67804.51183.464.9264.7731.735655.5831095121.291176.6319365000202.267.226576.9674220.07681.486783.182062.91846343.17937.23168.242567.4212681.145711.2108886.830108.482119.56580.04127.3224.383976.27685875.195179.318.9515.31364985532723.137.14392.237697.37530.03.8718615.12417.6422343331167703331.737641029.882.21224.4382.80472.572.05139001826.294657.3451781.62658.93126.357833.7313313407467348542739663113941258877683661088100.49345991.324.530.961520502879.5054.9855.1112.5062495.55153.598510873190337458379002.104383030936.504.678403.82138889452.4126230.85275.721787635833769408254231438583.9107011.05215621332526560.148389.1579.6040925.38203.10397451310.20891417.3353.50116.462520.374.6187148527.81513.974.638281.397938776376503.4650.8557194210.3179832739.3457944053074251.393426.03259.461.999863.044377.74162.88101.728556.723.8284.667571851.25842096482139.48499103.5029051674.9943490.729927440.691371242.64315.1372250127.01330.62230787164868312.2871530413926.825972.0930375.91514.480.4232432323550156.06953248.262104490.558.611310.9622.750.23240.2333115.688919.130.480.4818.451112.49636.832139237.5710194.422491.142261981337.472519594739.3436.15446026.65178.2057.233.910892195688.4911152.313.8341285990.12326.16225.25263.51109894947.7517.7931.86662.522319497197.40538.82296.60098101214.8816.66149.57647068816.51922935.271.54610357142.19563.865203042203.1631.044.7891.58558.4417.61119.8352.3028.263133.3090441858.16800432.22147.285206.4051.286977035361.0524.49832657129.332051612520.3527.95.3387661.90111.78242.58168.08251.4445.6394217142.9772.707.9511.9325.45572.52108.65483155.827.0247109.0364.68142.93695842.9394121167000064052.793.7167506.74047312.35586.8292.371.16054129.026.8261169.68585.7537149.885.74291.9076.06226.1391.7761.9691630.7011.589322.61071.4851.94230.6330.5330.79172.89266.65125.2775.6938.14712.54839.29179255265119170.39013.141171.0213112236.02437.180677.5833.1610027.1121.8746431148.33653.7552138776205.87419.8911685931470.518.773240910.218.0546.70689.12139.477735566667296.210945272827.878920140145.1136.9020.28145.325083100.6271977177579236.472555.6387.7703425285.971003.59122.55232.47141.62962951458222.4745.14448.67616.57664182563462.141061.538.173004123.952105.5850.531068.16347.977133874069561449703052855365900017.47268.426914416470.706478128564655502111362.993042.0243.6167.48808.62183.665.3565.2631.955805.5871124120.280179.9919611667200.9787.060977.1675119.79101.475313.099053.09150323.38137.00258.311078.0638979.703710.9182864.561107.793122.90895.48162.0323.717977.51964256.283635.698.2155.330500112461934.249.24510.629585.510640.84.6423774.88541.9228070001363440002.205601286.312.7304.22102.35681.130.96183241534.598864.9622184.94866.22834.6491059.9917601876559365832975290151313824427927251611670.608131274.275.330.5059906761120.7774.2273.519.9433221.20120.858678204250724719249722.773296841241.453.806935.273331185073.1233784.74351.232387621161391726942297721835.3475312.82156215541831647.1011623101.8002697.15258.378010085313.29531835.7839.64521.989623.476.2380062286.19518.096.254181.832754316120272.1031.1237270736.0249374438.0178147859705850.594566.42199.311.973623.965939.52210.6478.0911892.693.0046.316052468.43619374441103.25648144.0033572229.5756873.239475809.781758023.97242.015372697.84453.87308859475027915.7152872539442.248184.4441291.50917.050.46314403111967120.91479358.562749241.8010.41976.2023.480.30315.3044621.240922.110.640.6425.008614.66464.506181305.227815.551973.992751671036.153130934524.5114.44790431.67216.6771.643.575682762180.2817215.923.45461108719.12246.66329.82202.530906116236.1619.6837.22491.502378371975.25044.46210.39263118177.5020.74339.28207429120.52837547.762.12313993533.02950.699621876284.3435.235.64102.6008.7827.43127.6957.6822.123103.7308204586.46357339.22057.233256.6539.010788519428.6820.09039830104.124424270798.4745.54.4378980.93117.12331.83172.91337.4649.24112383150.5180.799.2812.7020.25481.5689.62592280.730.1852130.3866.47129.08572931.8558523359666771721.1111.9975445.74667110.71469.98110.521.41349131.139.2661542.11403.4641255.986.526109.7397.27231.05109.3649.1792735.019.414336.91192.0744.34234.7934.6834.72204.70202.27145.0426.4907.21512.76249.18340284273914193.96814.536221.2227855640.17741.770719.9853.6411606.2217.2911479162.42445.5862166699177.01621.5051696251351.920.210237411.520.0581.672103.25168.616728460000254.773165262407.899819175639.5332.5560.32118.52503790.0301914168313256.892586.4975.6802317331.37936.34108.81259.35155.87770544979233.7446.15650.45815.05570226868421.831034.932.846554140.97995.0949.548162.30333.417623218575670151774340115314700014.76244.447472357966.496873598568771814110954.493338.2231.5556.96812.29183.665.8765.8232.405865.8151131121.994177.9318078200197.9756.414515.9320720.18401.387922.966922.67554373.66833.14188.192537.0924078.39269.74591707.631108.948141.76875.83126.0335.0223107.74107389.246270.8210.3937.1870621201391005.3810.00559.8810308.39792.54.3625279.97589.7730303331445570000.654251372.192.62337.45105.20772.210.90088942430.677766.9212254.38868.47034.7391124.0618773650537068550488427160618441203967122275000.270391465.715.020.5139742821191.7285.2985.119.1833338.38111.625715174519725186255302.813111641113.852.704305.105891071373.1235169.31366.702469382234389195028334537295.1857011.91158915901898597.7911867101.6138680.74259.009410679013.17241936.5540.42122.570121.406.0565967324.68219.126.068951.816451157401824.361.1143274041.4255834948.4380520513075977.768591.9197.381.935533.996229.26227.4576.3912180.802.7716.091482586.65646894459101.19625696.1502682391.4457419.442202851.971909466.06236.285420295.44469.54313999200642115.8953426568714.458480.4441268.44116.340.44328643288264118.6162756.982651410.8110.25905.0323.470.30327.4145221.775720.310.640.6425.534814.40428.254163294.917495.551890.00294891962.2083003616482.7954.40832331.29215.6872.073.345052658380.0525252.353.51065120690.43225.98729.53201.6481124133.8817.7636.71453.306378116670.29943.54168.98131123864.1619.94352.22846332319.80740212.652.28814156530.54967.558520533281.5634.715.3295.22377.7527.20124.5256.2821.046113.1673717294.02936320.09356.958248.7347.848594598422.8620.20939608131.868301941942.3888.74.4698780.87114.08330.16166.31336.7847.98115980146.0870.318.0912.3019.34970.4491.58890283.629.3651131.0563.85131.16694528.1378723015333369839.8115.4873004.54585210.90685.73113.511.49327125.349.5251508.63421.9643166.106.413114.1297.41220.31109.6242.1763936.238.969327.75193.5645.73334.9534.7634.78199.14229.49143.2776.4527.16812.61188.98958258279043182.04914.602273.1181777538.71940.009769.6923.5314130.3014.7612158156.67543.9851841970191.34220.7461722651273.919.944231313.224.0568.40699.22204.342718306667243.100326022123.994118523941.3430.4710.31138.93513287.8391850159977279.622366.1770.7055898313.09978.51105.03291.70187.78073321638258.3745.97547.23315.21472710592407.64908.833.649585161.84492.9148.141660.52326.497820408677878702793240345103100014.76237.327717216064.557075257369435597110155.893515.9231.3961.34791.79183.463.7063.6831.325615.5261076137.918171.6019125000199.9195.324534.9005720.77051.298193.051352.66341378.25134.05367.246396.2052579.312510.5013669.803111.238140.08782.02148.2822.7023113.7417529.936140.899.5395.6045021606641362.0013.52729.8414289.712852.64.1134815.39795.0541180001813310002.788461938.322.37317.25136.73701.840.722110824349.9531248.8103269.121249.61949.9851556.07251224402513921296221502153368476771298535264730.756461587.744.820.36113841661653.6988.3288.846.6914734.2782.473973406136017116172854.182028359037.143.886857.369091703734.6549485.35476.6333989652273130693063442708587.4664911.72108010691265590.8016120154.7202481.15392.237714275319.90382672.5127.91431.365821.898.9819592259.95326.468.983992.684175134161573.0421.6391373632.3346405073.23110415799238199.655172.44136.111.130005.808356.74308.2153.2516452.472.0168.978133500.1344247337072.75951674.5064613179.3283365.156410032.252571394.82165.353819266.78633.444387913497162623.1037923778238.331133758200.76816.730.4143795436959485.31792258.433844031.2815.92714.3845.000.41425.4760327.387320.090.860.8632.220922.02354.764250409.965897.721348.37397138706.6454294631421.4303.29987946.41311.9599.613.006363761626.6118660.902.46159136360.72188.52542.60152.301783154835.0625.1554.54387.011511989157.47262.56183.23451162348.3327.74830.13223158027.22344844.132.54317088825.30139.561071375364.9048.115.12133.32211.3634.65181.8381.2317.46483.9703858866.40719273.49372.744322.5333.126776518560.8116.1574952285.254370260793.8797.03.421119100.00166.81408.42248.39419.2664.16151729203.3990.499.9016.9818.21189.9872.106121061.737.2714158.9488.30102.15238129.5529729239000092195.2135.3095084.8565178.85072.88133.211.76349186.5612.4812123.37322.6747556.697.838132.62111.01301.57132.1342.2256541.028.652380.818103.5236.33140.6840.2439.99242.49183.64170.2147.8185.89015.556011.5199338348099220.90717.435220.9291575848.52548.588829.1943.3313374.3516.5884685189.41340.1102770195154.98924.1712011821611.523.190208010.718.8660.336118.06162.329795966667226.011045332351.513523331343.3829.4430.29108.64522482.3771826198538248.892785.9673.8836784301.79891.84103.37251.80156.77374075244227.3950.91448.87714.13473716025399.561049.130.765271133.16892.0852.721460.66321.297845431677605627793153195301866714.93235.167765891465.636985751969731166105357.343265.2230.4165.69803.04183.569.8169.8034.216125.7801183122.627186.4218964400193.4656.378066.1614518.96641.205042.625242.38299422.41726.82848.151557.5174464.28277.50080567.06586.8403165.28567.71110.0537.2403127.9317477.196642.9413.6919.5777531550441326.2413.19731.4813432.713807.24.2833111.82776.8438250001709080002.911831863.362.51435.77138.501011.610.628127329448.8211220.5263099.901240.06049.6021560.99246764339927906864501032117266560871278572903500.757681882.345.010.36013872541588.3096.1095.957.0304486.5885.308939706006717460184424.152151756703.062.890877.144701672014.5546506.75516.9032913806810130006737455425607.3074812.04115511401319605.1515777143.3020481.39365.317214046718.61752563.7928.53332.177521.858.4994588626.45125.548.498692.563666629443540.871.5706363062.6336936603.67106633568137928.208830.77139.571.166745.638225.26298.1554.6416010.742.1178.668323424.5746904338472.78923331.4847683149.4380769.455781750.842511125.72168.043908967.99617.794376713284399222.1140142749237.411120757204.01515.450.4341889419619287.7020857.683756154.9815.07710.7894.880.40409.8960629.165520.570.840.8434.236520.65345.346243396.856207.651355.10384183694.5154334593395.4363.52156543.89297.0397.773.113073693215.3024785.652.70782150874.91184.51340.15153.648652158717.7418.3251.63379.719526262857.47858.25141.4564160748.1826.40828.06273272025.90747856.872.68217712525.24936.314433538357.6044.595.10120.3517.8934.02168.1374.9617.66681.8791574284.24532267.07769.960315.4531.121479799554.1615.6835102278.378737947970.7961.23.568115100.05154.02410.42224.02419.9359.48145533187.7383.188.4016.4317.72983.3074.061118066.734.7164162.4780.52104.32436122.9716528256333390314.0139.1196269.1584058.56347.02135.781.76588169.8111.7662051.60324.4648014.867.296130.80114.88271.84131.3434.6400440.488.373374.229107.5337.10740.3340.4940.37239.93168.39166.3747.4946.17414.766411.0380325346495209.22416.820248.0292956145.38347.053895.9303.5115940.6113.8228168175.20540.0992706385154.85222.1741971941592.321.723212712.922.5631.942109.57199.254772616667204.598475712040.354722781346.2029.5020.3117.99397883.7161715196283293.872686.2165.1065547312.52823.26101.02291.26183.82276031850265.1848.86047.57314.58574579896398.76907.529.689215140.35292.2753.149759.61316.268018926379159441799996114958466715.12232.677860081964.527089646970817373105160.143350.9227.3468.80746.24183.665.2465.4932.145825.4801124139.017176.6919119067198.0565.439505.0182418.81961.115522.671292.41341418.46728.80377.190057.1210067.75788.30854566.85090.6426162.00267.88106.0150.2806139.0639760.758364.0714.68010.591651OpenBenchmarking.org

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 PSKEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830K60K90K120K150KSE +/- 52.29, N = 3SE +/- 166.40, N = 3SE +/- 221.78, N = 3SE +/- 646.01, N = 3SE +/- 115.44, N = 3SE +/- 425.10, N = 3SE +/- 585.52, N = 3SE +/- 135.70, N = 3635668419785532112461120139160664155044295601. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830060090012001500SE +/- 0.14, N = 3SE +/- 0.27, N = 3SE +/- 0.13, N = 3SE +/- 0.30, N = 3SE +/- 0.16, N = 3SE +/- 1.37, N = 3SE +/- 0.25, N = 3SE +/- 0.54, N = 3531.48693.78723.13934.241005.381362.001326.24265.831. (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 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24883691215SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 35.256.907.149.2410.0013.5213.192.651. (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 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488160320480640800SE +/- 3.75, N = 3SE +/- 3.35, N = 3SE +/- 1.00, N = 3SE +/- 0.66, N = 3SE +/- 1.93, N = 3SE +/- 3.18, N = 3SE +/- 0.36, N = 3SE +/- 0.33, N = 3272.46368.26392.23510.62559.88729.84731.48150.641. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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.orgsign/s, More Is BetterOpenSSL 3.3Algorithm: RSA4096EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24883K6K9K12K15KSE +/- 3.91, N = 3SE +/- 2.72, N = 3SE +/- 2.36, N = 3SE +/- 3.17, N = 3SE +/- 14.97, N = 3SE +/- 5.90, N = 3SE +/- 2.41, N = 3SE +/- 28.09, N = 35652.67335.27697.39585.510308.314289.713432.73000.21. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

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 TotalEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24883K6K9K12K15KSE +/- 13.19, N = 7SE +/- 11.93, N = 7SE +/- 38.56, N = 7SE +/- 22.68, N = 6SE +/- 99.73, N = 6SE +/- 51.53, N = 6SE +/- 68.24, N = 6SE +/- 5.11, N = 56218.27583.77530.010640.89792.512852.613807.22908.5-mavx512f -mavx512cd -mavx512bw -mavx512dq-mavx512f -mavx512cd -mavx512bw -mavx512dq-mavx512f -mavx512cd -mavx512bw -mavx512dq-mavx512f -mavx512cd -mavx512bw -mavx512dq-mavx512f -mavx512cd -mavx512bw -mavx512dq-mavx512f -mavx512cd -mavx512bw -mavx512dq-mavx512f -mavx512cd -mavx512bw -mavx512dq1. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -fno-trapping-math -fno-math-errno -ldl

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 Re-Identification Retail FP16 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248848121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.11, N = 35.154.073.874.644.364.114.2818.22MIN: 3.03 / MAX: 14.81MIN: 2.41 / MAX: 9.33MIN: 2.38 / MAX: 13.89MIN: 2.6 / MAX: 15.94MIN: 2.55 / MAX: 15.23MIN: 2.45 / MAX: 41.27MIN: 2.63 / MAX: 16.2MIN: 10.36 / MAX: 177.581. (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 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24887K14K21K28K35KSE +/- 1.45, N = 3SE +/- 12.32, N = 3SE +/- 8.32, N = 3SE +/- 16.45, N = 3SE +/- 16.68, N = 3SE +/- 16.51, N = 3SE +/- 29.98, N = 3SE +/- 31.71, N = 314005.4618120.3818615.1223774.8825279.9734815.3933111.827499.471. (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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882004006008001000SE +/- 1.40, N = 3SE +/- 0.81, N = 3SE +/- 0.59, N = 3SE +/- 0.61, N = 3SE +/- 1.97, N = 3SE +/- 1.22, N = 3SE +/- 0.33, N = 3SE +/- 0.45, N = 3301.93395.19417.64541.92589.77795.05776.84174.921. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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: MD5EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488900K1800K2700K3600K4500KSE +/- 17071.74, N = 3SE +/- 666.67, N = 3SE +/- 7423.69, N = 3SE +/- 8082.90, N = 3SE +/- 333.33, N = 3SE +/- 11590.23, N = 3SE +/- 17559.42, N = 3SE +/- 2504.83, N = 315683332219333223433328070003030333411800038250009197911. (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: HMAC-SHA512EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248840M80M120M160M200MSE +/- 24875.24, N = 3SE +/- 110045.95, N = 3SE +/- 411753.30, N = 3SE +/- 1126651.23, N = 3SE +/- 58043.09, N = 3SE +/- 580393.26, N = 3SE +/- 243467.31, N = 3SE +/- 518797.01, N = 389217667117072667116770333136344000144557000181331000170908000407210001. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

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 AtomsEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.65521.31041.96562.62083.276SE +/- 0.00340, N = 3SE +/- 0.00076, N = 3SE +/- 0.00212, N = 3SE +/- 0.00983, N = 3SE +/- 0.01721, N = 15SE +/- 0.03311, N = 4SE +/- 0.00837, N = 3SE +/- 0.00164, N = 31.319211.691201.737642.205600.654252.788462.911830.65977

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Re-Identification Retail FP16 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488400800120016002000SE +/- 0.80, N = 3SE +/- 1.30, N = 3SE +/- 0.99, N = 3SE +/- 0.33, N = 3SE +/- 1.33, N = 3SE +/- 0.22, N = 3SE +/- 0.50, N = 3SE +/- 2.58, N = 3773.96978.601029.881286.311372.191938.321863.36437.961. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24883691215SE +/- 0.01, 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 +/- 0.02, N = 32.982.372.212.702.622.372.519.76MIN: 1.64 / MAX: 9.56MIN: 1.25 / MAX: 6.4MIN: 1.23 / MAX: 13.12MIN: 1.33 / MAX: 14.55MIN: 1.36 / MAX: 13.53MIN: 1.27 / MAX: 12.75MIN: 1.32 / MAX: 13.29MIN: 4.34 / MAX: 46.891. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248890180270360450SE +/- 0.31, N = 3SE +/- 0.84, N = 3SE +/- 1.10, N = 3SE +/- 0.76, N = 3SE +/- 1.33, N = 3SE +/- 0.46, N = 3SE +/- 1.74, N = 3SE +/- 0.22, N = 3146.62171.63224.43304.22337.45317.25435.77101.211. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Machine Translation EN To DE FP16 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.39, N = 3SE +/- 0.34, N = 3SE +/- 0.17, N = 3SE +/- 0.13, N = 3SE +/- 0.10, N = 3SE +/- 0.06, N = 3SE +/- 0.12, N = 3SE +/- 0.02, N = 363.1574.7982.80102.35105.20136.73138.5033.741. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882004006008001000SE +/- 2.00, N = 3SE +/- 0.47, N = 3SE +/- 1.08, N = 3SE +/- 0.50, N = 3SE +/- 1.18, N = 3SE +/- 2.53, N = 3SE +/- 2.08, N = 3SE +/- 0.14, N = 3364.22436.41472.57681.13772.21701.841011.61249.021. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average LatencyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.56811.13621.70432.27242.8405SE +/- 0.021, N = 9SE +/- 0.010, N = 3SE +/- 0.023, N = 3SE +/- 0.007, N = 3SE +/- 0.012, N = 3SE +/- 0.006, N = 3SE +/- 0.002, N = 3SE +/- 0.005, N = 32.5251.9292.0510.9610.9000.7220.6281.8521. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 800 - Mode: Read OnlyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488300K600K900K1200K1500KSE +/- 2616.05, N = 9SE +/- 2047.43, N = 3SE +/- 4300.20, N = 3SE +/- 6188.19, N = 3SE +/- 11913.26, N = 3SE +/- 8757.92, N = 3SE +/- 4344.86, N = 3SE +/- 1154.98, N = 3317051414750390018832415889424110824312732944319621. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

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: BM1EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881122334455SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.10, N = 3SE +/- 0.10, N = 3SE +/- 0.26, N = 3SE +/- 0.03, N = 4SE +/- 0.17, N = 4SE +/- 0.06, N = 319.0925.8526.2934.6030.6849.9548.8212.451. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830060090012001500SE +/- 0.09, N = 3SE +/- 0.15, N = 3SE +/- 2.56, N = 3SE +/- 2.41, N = 3SE +/- 6.54, N = 3SE +/- 0.84, N = 4SE +/- 4.13, N = 4SE +/- 1.56, N = 3477.36646.26657.35864.96766.921248.811220.53311.311. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -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.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24887001400210028003500SE +/- 3.43, N = 3SE +/- 1.26, N = 3SE +/- 0.66, N = 3SE +/- 0.82, N = 3SE +/- 2.38, N = 3SE +/- 6.25, N = 3SE +/- 1.96, N = 3SE +/- 1.96, N = 31330.811659.061781.622184.942254.383269.123099.90816.171. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830060090012001500SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.13, N = 3SE +/- 0.58, N = 3SE +/- 0.75, N = 3SE +/- 1.54, N = 3SE +/- 2.18, N = 3SE +/- 0.15, N = 3478.87647.13658.93866.23868.471249.621240.06313.741. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881122334455SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 0.01, N = 319.1625.8926.3634.6534.7449.9949.6012.551. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -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.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Vehicle Bike Detection FP16 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830060090012001500SE +/- 2.84, N = 3SE +/- 5.03, N = 3SE +/- 0.80, N = 3SE +/- 0.91, N = 3SE +/- 5.46, N = 3SE +/- 2.52, N = 3SE +/- 2.21, N = 3SE +/- 0.87, N = 3603.92790.63833.731059.991124.061556.071560.99401.421. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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-GCMEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850000M100000M150000M200000M250000MSE +/- 30766844.46, N = 3SE +/- 18100529.72, N = 3SE +/- 48206635.82, N = 3SE +/- 82240254.30, N = 3SE +/- 100195492.71, N = 3SE +/- 197196146.33, N = 3SE +/- 834503022.99, N = 3SE +/- 45687236.99, N = 398225529560129046626940133134074673176018765593187736505370251224402513246764339927648127256801. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: ChaCha20-Poly1305EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820000M40000M60000M80000M100000MSE +/- 14961656.63, N = 3SE +/- 18319786.28, N = 3SE +/- 11858939.41, N = 3SE +/- 17029222.81, N = 3SE +/- 21541900.76, N = 3SE +/- 63328398.60, N = 3SE +/- 37115945.76, N = 3SE +/- 5622113.45, N = 335429817183472542301734854273966365832975290685504884279212962215090686450103238002630201. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: AES-256-GCMEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850000M100000M150000M200000M250000MSE +/- 23525395.17, N = 3SE +/- 28146850.41, N = 3SE +/- 101608115.02, N = 3SE +/- 10292339.65, N = 3SE +/- 51787792.79, N = 3SE +/- 27189836.88, N = 3SE +/- 124036322.98, N = 3SE +/- 5611535.27, N = 384060321200110425587253113941258877151313824427160618441203215336847677211726656087561932491901. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: ChaCha20EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830000M60000M90000M120000M150000MSE +/- 41193652.45, N = 3SE +/- 10772886.17, N = 3SE +/- 32786296.71, N = 3SE +/- 14660801.77, N = 3SE +/- 26095977.93, N = 3SE +/- 13732325.68, N = 3SE +/- 45400804.32, N = 3SE +/- 7700475.42, N = 34970717484366572774607683661088109272516116796712227500129853526473127857290350346518558001. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

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 AtomsEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.17050.3410.51150.6820.8525SE +/- 0.00046, N = 3SE +/- 0.00019, N = 3SE +/- 0.00059, N = 3SE +/- 0.00106, N = 3SE +/- 0.00185, N = 3SE +/- 0.00230, N = 3SE +/- 0.00443, N = 3SE +/- 0.00016, N = 30.389560.485700.493450.608130.270390.756460.757680.20565

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Noise Suppression Poconet-Like FP16 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488400800120016002000SE +/- 1.27, N = 3SE +/- 1.25, N = 3SE +/- 0.86, N = 3SE +/- 0.77, N = 3SE +/- 1.93, N = 3SE +/- 1.23, N = 3SE +/- 2.01, N = 3SE +/- 1.17, N = 3697.69890.99991.321274.271465.711587.741882.34523.021. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16-INT8 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248848121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.05, N = 36.194.754.535.335.024.825.0115.95MIN: 3.5 / MAX: 13.93MIN: 2.76 / MAX: 10.38MIN: 2.65 / MAX: 17.47MIN: 2.88 / MAX: 17.76MIN: 2.86 / MAX: 11.09MIN: 2.78 / MAX: 17.83MIN: 2.78 / MAX: 17.17MIN: 9.85 / MAX: 70.281. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 500 - Mode: Read Only - Average LatencyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.27160.54320.81481.08641.358SE +/- 0.014, N = 3SE +/- 0.009, N = 3SE +/- 0.012, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 3SE +/- 0.003, N = 31.2070.9040.9610.5050.5130.3610.3600.8991. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 500 - Mode: Read OnlyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488300K600K900K1200K1500KSE +/- 4829.64, N = 3SE +/- 5619.73, N = 3SE +/- 6448.13, N = 3SE +/- 2177.66, N = 3SE +/- 2685.11, N = 3SE +/- 3948.07, N = 3SE +/- 3546.18, N = 3SE +/- 1617.01, N = 3414485553371520502990676974282138416613872545558731. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -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.orgFPS, More Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16-INT8 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488400800120016002000SE +/- 1.34, N = 3SE +/- 2.16, N = 3SE +/- 0.43, N = 3SE +/- 0.37, N = 3SE +/- 1.23, N = 3SE +/- 1.19, N = 3SE +/- 0.49, N = 3SE +/- 1.52, N = 3644.50838.46879.501120.771191.721653.691588.30500.861. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Detection FP32 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100SE +/- 0.20, N = 3SE +/- 0.13, N = 3SE +/- 0.09, N = 3SE +/- 0.35, N = 3SE +/- 0.07, N = 3SE +/- 0.47, N = 3SE +/- 0.44, N = 3SE +/- 0.17, N = 344.4052.2454.9874.2285.2988.3296.1029.711. (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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.11, N = 3SE +/- 0.08, N = 3SE +/- 0.10, N = 3SE +/- 0.98, N = 3SE +/- 0.52, N = 3SE +/- 0.16, N = 344.1752.3555.1173.5185.1188.8495.9529.831. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -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: 1e12EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488510152025SE +/- 0.014, N = 3SE +/- 0.013, N = 4SE +/- 0.007, N = 4SE +/- 0.007, N = 5SE +/- 0.005, N = 5SE +/- 0.007, N = 6SE +/- 0.009, N = 6SE +/- 0.302, N = 316.82712.93712.5069.9439.1836.6917.03021.0951. (CXX) g++ options: -O3

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16-INT8 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248810002000300040005000SE +/- 0.53, N = 3SE +/- 2.13, N = 3SE +/- 1.63, N = 3SE +/- 1.50, N = 3SE +/- 2.70, N = 3SE +/- 1.99, N = 3SE +/- 3.97, N = 3SE +/- 3.53, N = 31804.902329.232495.553221.203338.384734.274486.581559.061. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -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: 1e13EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850100150200250SE +/- 0.10, N = 3SE +/- 0.11, N = 3SE +/- 0.15, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.20, N = 3SE +/- 0.07, N = 3SE +/- 0.18, N = 3205.20158.45153.60120.86111.6382.4785.31245.671. (CXX) g++ options: -O3

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 23.5Algorithm: Triple SHA-256, OnecoinEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820K40K60K80K100KSE +/- 11.55, N = 3SE +/- 25.17, N = 3SE +/- 8.82, N = 3SE +/- 72.11, N = 3SE +/- 188.09, N = 3SE +/- 146.40, N = 3SE +/- 20.82, N = 3SE +/- 3.33, N = 332710499005108767820715179734093970388471. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 23.5Algorithm: Quad SHA-256, PyriteEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248813K26K39K52K65KSE +/- 58.59, N = 3SE +/- 38.44, N = 3SE +/- 78.81, N = 3SE +/- 3.33, N = 3SE +/- 26.67, N = 3SE +/- 331.71, N = 3SE +/- 454.25, N = 3SE +/- 3.33, N = 320720322933190342507451976136060067265271. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

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: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248811K22K33K44K55KSE +/- 80.90, N = 3SE +/- 55.41, N = 3SE +/- 43.80, N = 3SE +/- 16.70, N = 3SE +/- 280.33, N = 3SE +/- 33.74, N = 3SE +/- 12.67, N = 3SE +/- 112.38, N = 35005538334374582471925186171161746041514

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 2 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248811K22K33K44K55KSE +/- 199.64, N = 3SE +/- 87.29, N = 3SE +/- 127.68, N = 3SE +/- 39.63, N = 3SE +/- 293.34, N = 3SE +/- 19.75, N = 3SE +/- 3.84, N = 3SE +/- 172.31, N = 35033938690379002497225530172851844242275

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.94051.8812.82153.7624.7025SE +/- 0.02, N = 5SE +/- 0.02, N = 3SE +/- 0.02, N = 11SE +/- 0.03, N = 3SE +/- 0.03, N = 5SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 151.462.082.102.772.814.184.151.45MIN: 0.38 / MAX: 1.85MIN: 0.66 / MAX: 2.47MIN: 0.64 / MAX: 2.54MIN: 0.99 / MAX: 3.25MIN: 1.01 / MAX: 3.32MIN: 1.81 / MAX: 4.75MIN: 1.74 / MAX: 4.72MIN: 0.39 / MAX: 1.82

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: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248813K26K39K52K65KSE +/- 91.50, N = 3SE +/- 85.35, N = 3SE +/- 62.68, N = 3SE +/- 102.05, N = 3SE +/- 529.04, N = 15SE +/- 29.87, N = 3SE +/- 45.16, N = 3SE +/- 115.84, N = 35844244674438303296831116202832151748929

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Vector Floating PointEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248813K26K39K52K65KSE +/- 108.35, N = 3SE +/- 129.67, N = 3SE +/- 415.80, N = 3SE +/- 284.33, N = 3SE +/- 59.97, N = 3SE +/- 261.75, N = 3SE +/- 280.00, N = 3SE +/- 378.33, N = 320773.7028866.8830936.5041241.4541113.8559037.1456703.0628846.061. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lmpfr -lpthread -lrt -lsctp -lz

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488246810SE +/- 0.00220, N = 5SE +/- 0.00237, N = 5SE +/- 0.00833, N = 5SE +/- 0.00774, N = 5SE +/- 0.00680, N = 5SE +/- 0.00611, N = 5SE +/- 0.01016, N = 5SE +/- 0.01930, N = 54.494064.516294.678403.806932.704303.886852.890877.59695MIN: 4.45MIN: 4.45MIN: 4.46MIN: 3.54MIN: 2.66MIN: 3.57MIN: 2.68MIN: 7.291. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -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: gravity_spheres_volume/dim_512/scivis/real_timeEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488246810SE +/- 0.01153, N = 3SE +/- 0.01618, N = 3SE +/- 0.00260, N = 3SE +/- 0.01288, N = 3SE +/- 0.01525, N = 3SE +/- 0.02548, N = 3SE +/- 0.00808, N = 3SE +/- 0.00622, N = 32.919513.794883.821385.273335.105897.369097.144702.63851

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 RatingEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248840K80K120K160K200KSE +/- 31.18, N = 3SE +/- 107.68, N = 3SE +/- 299.70, N = 3SE +/- 184.46, N = 3SE +/- 503.79, N = 3SE +/- 369.09, N = 3SE +/- 83.27, N = 3SE +/- 900.25, N = 3678688840988945118507107137170373167201615271. (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: LuxCore Benchmark - Acceleration: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881.04632.09263.13894.18525.2315SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 31.682.332.413.123.124.654.551.69MIN: 0.5 / MAX: 2.06MIN: 0.78 / MAX: 2.76MIN: 0.82 / MAX: 2.85MIN: 1.19 / MAX: 3.62MIN: 1.17 / MAX: 3.62MIN: 2.03 / MAX: 5.27MIN: 1.97 / MAX: 5.16MIN: 0.51 / MAX: 2.04

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248811K22K33K44K55KSE +/- 103.84, N = 3SE +/- 5.77, N = 3SE +/- 19.90, N = 3SE +/- 68.04, N = 3SE +/- 42.38, N = 3SE +/- 47.80, N = 3SE +/- 32.19, N = 3SE +/- 38.88, N = 317920.7525613.2926230.8533784.7435169.3149485.3546506.7518507.311. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16-INT8 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488110220330440550SE +/- 0.12, N = 3SE +/- 1.13, N = 3SE +/- 0.50, N = 3SE +/- 0.40, N = 3SE +/- 0.69, N = 3SE +/- 1.94, N = 3SE +/- 0.32, N = 3SE +/- 0.28, N = 3187.28237.30275.72351.23366.70476.63516.90219.861. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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: SHA256EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24887000M14000M21000M28000M35000MSE +/- 4935253.44, N = 3SE +/- 7769960.49, N = 3SE +/- 8102567.82, N = 3SE +/- 5636428.54, N = 3SE +/- 9363717.55, N = 3SE +/- 7892274.19, N = 3SE +/- 8957910.25, N = 3SE +/- 27097230.48, N = 313220736913174027128101787635833723876211613246938223433398965227332913806810123365572631. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

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 ReadEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830M60M90M120M150MSE +/- 45665.92, N = 3SE +/- 68428.53, N = 3SE +/- 144351.01, N = 3SE +/- 74109.11, N = 3SE +/- 342211.57, N = 3SE +/- 780691.20, N = 3SE +/- 276158.53, N = 3SE +/- 236765.68, N = 35133414667959360694082549172694289195028130693063130006737475122601. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

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 BenchmarkEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248810M20M30M40M50MSE +/- 169306.09, N = 3SE +/- 191557.41, N = 3SE +/- 160295.89, N = 3SE +/- 353999.00, N = 15SE +/- 379095.21, N = 4SE +/- 574948.55, N = 15SE +/- 611568.23, N = 15SE +/- 121681.53, N = 151661451721844343231438582977218333453729442708584554256016898291-mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl-mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl-mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl-mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl-mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl-mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl-mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl1. (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 -msse4.1 -mssse3 -msse2 -mbmi2 -flto -flto-partition=one -flto=jobserver

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_timeEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488246810SE +/- 0.00463, N = 3SE +/- 0.00148, N = 3SE +/- 0.00682, N = 3SE +/- 0.00993, N = 3SE +/- 0.01719, N = 3SE +/- 0.00259, N = 3SE +/- 0.03021, N = 3SE +/- 0.00479, N = 32.950103.816913.910705.347535.185707.466497.307482.73496

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 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488714212835SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.06, N = 311.2711.5111.0512.8211.9111.7212.0430.05MIN: 5.84 / MAX: 21.18MIN: 5.95 / MAX: 21.47MIN: 5.83 / MAX: 23.56MIN: 6.57 / MAX: 26.61MIN: 6.32 / MAX: 23.15MIN: 6 / MAX: 26.83MIN: 6.3 / MAX: 25.96MIN: 17.38 / MAX: 87.991. (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: 2 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24886001200180024003000SE +/- 3.28, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 9.29, N = 3SE +/- 2.08, N = 3SE +/- 0.58, N = 3SE +/- 2.33, N = 329302201215615621589108011552474

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 1 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24886001200180024003000SE +/- 1.45, N = 3SE +/- 1.86, N = 3SE +/- 1.76, N = 3SE +/- 1.45, N = 3SE +/- 16.15, N = 3SE +/- 1.00, N = 3SE +/- 1.53, N = 3SE +/- 0.88, N = 328992176213315541590106911402441

OpenBenchmarking.orgms, Fewer Is BetterOSPRay Studio 1.0Camera: 3 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24887001400210028003500SE +/- 5.49, N = 3SE +/- 2.52, N = 3SE +/- 4.04, N = 3SE +/- 3.18, N = 3SE +/- 2.33, N = 3SE +/- 0.33, N = 3SE +/- 15.53, N = 4SE +/- 0.88, N = 334302575252618311898126513192884

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 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830060090012001500SE +/- 0.60, N = 3SE +/- 0.62, N = 3SE +/- 0.52, N = 3SE +/- 0.16, N = 3SE +/- 0.35, N = 3SE +/- 1.73, N = 3SE +/- 0.56, N = 3SE +/- 8.08, N = 3761.78579.07560.14647.10597.79590.80605.151506.63MIN: 744.08 / MAX: 774.48MIN: 564.55 / MAX: 589.31MIN: 538 / MAX: 581.72MIN: 540.69 / MAX: 688.06MIN: 574.7 / MAX: 616.24MIN: 555.5 / MAX: 623.22MIN: 570.99 / MAX: 624.11MIN: 1180.01 / MAX: 1642.121. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 23.5Algorithm: LBC, LBRY CreditsEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24883K6K9K12K15KSE +/- 2.97, N = 3SE +/- 27.28, N = 3SE +/- 7.26, N = 3SE +/- 3.33, N = 3SE +/- 3.33, N = 3SE +/- 5.77, N = 3SE +/- 106.67, N = 3SE +/- 15.18, N = 36265.108323.698389.1511623.0011867.0016120.0015777.006037.261. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

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: MediumEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.02, N = 3SE +/- 0.04, N = 4SE +/- 0.04, N = 4SE +/- 0.03, N = 4SE +/- 0.07, N = 4SE +/- 0.18, N = 5SE +/- 0.06, N = 5SE +/- 0.00, N = 358.0876.0179.60101.80101.61154.72143.3089.141. (CXX) g++ options: -O3 -flto -pthread

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-OnlyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830060090012001500SE +/- 0.31, N = 3SE +/- 1.01, N = 3SE +/- 0.87, N = 3SE +/- 0.89, N = 3SE +/- 0.50, N = 3SE +/- 0.46, N = 3SE +/- 0.24, N = 3SE +/- 4.06, N = 31244.70942.29925.38697.15680.74481.15481.391281.54

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: FastEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248890180270360450SE +/- 0.12, N = 3SE +/- 0.39, N = 3SE +/- 0.29, N = 3SE +/- 0.15, N = 3SE +/- 0.09, N = 3SE +/- 0.54, N = 4SE +/- 0.31, N = 4SE +/- 0.04, N = 3147.48192.66203.10258.38259.01392.24365.32205.931. (CXX) g++ options: -O3 -flto -pthread

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 23.5Algorithm: Blake-2 SEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830K60K90K120K150KSE +/- 30.00, N = 3SE +/- 20.82, N = 3SE +/- 29.06, N = 3SE +/- 103.98, N = 3SE +/- 40.41, N = 3SE +/- 99.39, N = 3SE +/- 41.77, N = 3SE +/- 460.90, N = 3538707313074513100853106790142753140467561601. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

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: ThoroughEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488510152025SE +/- 0.0018, N = 3SE +/- 0.0050, N = 3SE +/- 0.0128, N = 3SE +/- 0.0021, N = 3SE +/- 0.0127, N = 3SE +/- 0.0627, N = 3SE +/- 0.0075, N = 3SE +/- 0.3204, N = 127.51369.867410.208913.295313.172419.903818.617512.23071. (CXX) g++ options: -O3 -flto -pthread

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16-INT8 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24886001200180024003000SE +/- 0.80, N = 3SE +/- 0.44, N = 3SE +/- 0.60, N = 3SE +/- 0.27, N = 3SE +/- 1.19, N = 3SE +/- 4.11, N = 3SE +/- 0.50, N = 3SE +/- 2.31, N = 31040.151356.641417.331835.781936.552672.512563.791022.101. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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 SolveEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881632486480SE +/- 0.08, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.62, N = 371.9854.3953.5039.6540.4227.9128.5372.851. (CXX) g++ options: -fopenmp -O2 -march=native

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: CrownEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488714212835SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 312.3316.0916.4621.9922.5731.3732.1813.39MIN: 12.21 / MAX: 12.61MIN: 15.93 / MAX: 16.53MIN: 16.24 / MAX: 16.97MIN: 21.74 / MAX: 22.75MIN: 22.22 / MAX: 23.17MIN: 30.96 / MAX: 32.3MIN: 31.73 / MAX: 33MIN: 13.27 / MAX: 16.49

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 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881224364860SE +/- 0.30, N = 3SE +/- 0.20, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.01, N = 3SE +/- 0.12, N = 322.0121.7020.3723.4721.4021.8921.8553.07MIN: 13.52 / MAX: 35.51MIN: 16.33 / MAX: 33.88MIN: 14.74 / MAX: 35.54MIN: 17.04 / MAX: 41.98MIN: 14.65 / MAX: 31.23MIN: 14.64 / MAX: 40.37MIN: 15.94 / MAX: 38.89MIN: 35.4 / MAX: 130.321. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -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/scivis/real_timeEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24883691215SE +/- 0.00497, N = 3SE +/- 0.00135, N = 3SE +/- 0.00347, N = 3SE +/- 0.00310, N = 3SE +/- 0.00208, N = 3SE +/- 0.00296, N = 3SE +/- 0.00215, N = 3SE +/- 0.00749, N = 33.447684.516504.618716.238006.056598.981958.499455.88640

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 1.7EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820K40K60K80K100KSE +/- 28.81, N = 3SE +/- 20.41, N = 3SE +/- 45.75, N = 3SE +/- 7.31, N = 3SE +/- 41.76, N = 3SE +/- 153.97, N = 3SE +/- 75.15, N = 3SE +/- 17.95, N = 335520.7346525.7848527.8262286.2067324.6892259.9588626.4535441.711. (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

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection FP16-INT8 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488612182430SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 310.1713.3613.9718.0919.1226.4625.5410.411. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -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/ao/real_timeEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24883691215SE +/- 0.00096, N = 3SE +/- 0.00247, N = 3SE +/- 0.00129, N = 3SE +/- 0.00160, N = 3SE +/- 0.00892, N = 3SE +/- 0.00600, N = 3SE +/- 0.00180, N = 3SE +/- 0.00839, N = 33.455774.525524.638286.254186.068958.983998.498695.93161

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 ThoroughEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.60391.20781.81172.41563.0195SE +/- 0.0004, N = 3SE +/- 0.0004, N = 3SE +/- 0.0010, N = 3SE +/- 0.0009, N = 3SE +/- 0.0004, N = 3SE +/- 0.0093, N = 3SE +/- 0.0008, N = 3SE +/- 0.0116, N = 31.03271.35631.39791.83271.81642.68412.56361.46391. (CXX) g++ options: -O3 -flto -pthread

Java JMH

This very basic test profile runs the stock benchmark of the Java JMH benchmark via Maven. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/s, More Is BetterJava JMHThroughputEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248816000M32000M48000M64000M80000M28953646655.5138412521187.6938776376503.4754316120272.1051157401824.3675134161573.0466629443540.8736765943472.27

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: ExhaustiveEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.36880.73761.10641.47521.844SE +/- 0.0004, N = 3SE +/- 0.0003, N = 3SE +/- 0.0000, N = 3SE +/- 0.0001, N = 3SE +/- 0.0003, N = 3SE +/- 0.0020, N = 3SE +/- 0.0003, N = 3SE +/- 0.0051, N = 30.63300.83130.85571.12371.11431.63911.57060.87411. (CXX) g++ options: -O3 -flto -pthread

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: RSA4096EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248880K160K240K320K400KSE +/- 34.43, N = 3SE +/- 13.93, N = 3SE +/- 20.61, N = 3SE +/- 313.76, N = 3SE +/- 923.26, N = 3SE +/- 143.01, N = 3SE +/- 494.21, N = 3SE +/- 1626.68, N = 3144386.1190537.1194210.3270736.0274041.4373632.3363062.6195618.81. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 23.5Algorithm: SkeincoinEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24887K14K21K28K35KSE +/- 11.55, N = 3SE +/- 11.55, N = 3SE +/- 39.30, N = 3SE +/- 42.56, N = 3SE +/- 134.45, N = 3SE +/- 11.55, N = 3SE +/- 8.82, N = 3SE +/- 139.08, N = 313390175801798324937255833464033693139331. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

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 - GriddingEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248814002800420056007000SE +/- 25.19, N = 7SE +/- 21.74, N = 15SE +/- 24.04, N = 8SE +/- 19.10, N = 6SE +/- 49.58, N = 6SE +/- 41.40, N = 6SE +/- 55.23, N = 6SE +/- 25.17, N = 52825.313122.292739.344438.014948.435073.236603.672556.221. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

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: SHA512EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882000M4000M6000M8000M10000MSE +/- 432194.03, N = 3SE +/- 485369.37, N = 3SE +/- 749962.39, N = 3SE +/- 1960683.64, N = 3SE +/- 7000138.46, N = 3SE +/- 2951460.39, N = 3SE +/- 5901271.36, N = 3SE +/- 2787035.99, N = 342752754605644357583579440530778147859708052051307110415799231066335681344293971601. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Floating PointEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882K4K6K8K10KSE +/- 1.77, N = 3SE +/- 0.05, N = 3SE +/- 6.00, N = 3SE +/- 1.22, N = 3SE +/- 2.90, N = 3SE +/- 15.20, N = 3SE +/- 8.85, N = 3SE +/- 7.41, N = 33175.544192.474251.395850.595977.768199.657928.204740.281. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lmpfr -lpthread -lrt -lsctp -lz

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 - DegriddingEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882K4K6K8K10KSE +/- 154.63, N = 7SE +/- 18.97, N = 15SE +/- 27.18, N = 8SE +/- 42.40, N = 6SE +/- 71.64, N = 6SE +/- 49.85, N = 6SE +/- 90.51, N = 6SE +/- 41.03, N = 54370.023420.873426.034566.428591.905172.448830.774282.181. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

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-OnlyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248880160240320400SE +/- 0.40, N = 3SE +/- 0.28, N = 3SE +/- 0.07, N = 3SE +/- 0.27, N = 3SE +/- 0.13, N = 3SE +/- 0.07, N = 3SE +/- 0.33, N = 3SE +/- 1.57, N = 3351.06267.86259.46199.31197.38136.11139.57336.94

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.65571.31141.96712.62283.2785SE +/- 0.00010, N = 4SE +/- 0.00137, N = 4SE +/- 0.00536, N = 4SE +/- 0.00923, N = 4SE +/- 0.00037, N = 4SE +/- 0.00269, N = 4SE +/- 0.00331, N = 4SE +/- 0.01544, N = 42.606381.979321.999861.973621.935531.130001.166742.91408MIN: 2.57MIN: 1.91MIN: 1.85MIN: 1.82MIN: 1.91MIN: 1.03MIN: 1.08MIN: 2.511. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

SVT-HEVC

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 1 - Input: Bosphorus 4KEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881.3052.613.9155.226.525SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 32.252.943.043.963.995.805.632.261. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 23.5Algorithm: DeepcoinEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882K4K6K8K10KSE +/- 1.04, N = 3SE +/- 1.63, N = 3SE +/- 3.02, N = 3SE +/- 7.14, N = 3SE +/- 23.29, N = 3SE +/- 5.77, N = 3SE +/- 18.11, N = 3SE +/- 36.27, N = 33247.544280.914377.745939.526229.268356.748225.263294.491. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 23.5Algorithm: scryptEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248870140210280350SE +/- 0.13, N = 3SE +/- 0.16, N = 3SE +/- 0.10, N = 3SE +/- 0.11, N = 3SE +/- 1.04, N = 3SE +/- 1.34, N = 3SE +/- 1.22, N = 3SE +/- 0.08, N = 3120.39157.91162.88210.64227.45308.21298.15119.861. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

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-OnlyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.46, N = 3SE +/- 0.10, N = 3SE +/- 0.33, N = 3SE +/- 0.16, N = 3SE +/- 0.17, N = 3SE +/- 0.05, N = 3SE +/- 0.14, N = 3SE +/- 0.18, N = 3136.83104.74101.7278.0976.3953.2554.64123.51

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Vector ShuffleEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884K8K12K16K20KSE +/- 3.05, N = 3SE +/- 1.05, N = 3SE +/- 2.63, N = 3SE +/- 1.99, N = 3SE +/- 4.46, N = 3SE +/- 12.57, N = 3SE +/- 37.39, N = 3SE +/- 104.90, N = 36403.878468.868556.7211892.6912180.8016452.4716010.749138.291. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lmpfr -lpthread -lrt -lsctp -lz

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: 12 digitEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881.16512.33023.49534.66045.8255SE +/- 0.002, N = 7SE +/- 0.003, N = 8SE +/- 0.004, N = 8SE +/- 0.003, N = 9SE +/- 0.003, N = 10SE +/- 0.003, N = 11SE +/- 0.003, N = 11SE +/- 0.028, N = 155.1783.9663.8283.0042.7712.0162.1174.0801. (CC) gcc options: -O2 -pthread

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_timeEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24883691215SE +/- 0.00315, N = 3SE +/- 0.00576, N = 3SE +/- 0.00236, N = 3SE +/- 0.00517, N = 3SE +/- 0.00278, N = 3SE +/- 0.00353, N = 3SE +/- 0.00316, N = 3SE +/- 0.00822, N = 33.496994.572224.667576.316056.091488.978138.668323.79539

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 23.5Algorithm: RingcoinEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24888001600240032004000SE +/- 1.54, N = 3SE +/- 2.14, N = 3SE +/- 4.79, N = 3SE +/- 2.43, N = 3SE +/- 2.15, N = 3SE +/- 20.59, N = 3SE +/- 6.91, N = 3SE +/- 3.73, N = 31364.321806.871851.252468.432586.653500.133424.571437.471. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

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: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820K40K60K80K100KSE +/- 16.70, N = 3SE +/- 185.23, N = 3SE +/- 273.73, N = 3SE +/- 30.37, N = 3SE +/- 84.90, N = 3SE +/- 101.69, N = 3SE +/- 195.52, N = 3SE +/- 76.92, N = 311350185743842096193764689442474690495447

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 TomcatEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882K4K6K8K10KSE +/- 4.16, N = 3SE +/- 6.08, N = 3SE +/- 9.21, N = 3SE +/- 4.80, N = 4SE +/- 5.15, N = 4SE +/- 2.95, N = 4SE +/- 2.87, N = 4SE +/- 14.71, N = 386286485648244414459337033846687

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-OnlyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 0.25, N = 3SE +/- 0.48, N = 3SE +/- 0.43, N = 3SE +/- 0.08, N = 3SE +/- 0.15, N = 3SE +/- 0.17, N = 3SE +/- 0.10, N = 3SE +/- 0.02, N = 3186.24141.77139.48103.25101.1972.7572.78173.17

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 SecondEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488200K400K600K800K1000KSE +/- 954.28, N = 3SE +/- 4350.78, N = 3SE +/- 1076.82, N = 3SE +/- 7794.37, N = 3SE +/- 7061.60, N = 3SE +/- 9073.63, N = 3SE +/- 1179.52, N = 3SE +/- 3256.58, N = 15372162.82491021.69499103.50648144.00625696.15951674.51923331.48430017.461. (CC) gcc options: -O2 -lrt" -lrt

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 23.5Algorithm: GarlicoinEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24887001400210028003500SE +/- 1.62, N = 3SE +/- 2.95, N = 3SE +/- 1.43, N = 3SE +/- 3.95, N = 3SE +/- 11.09, N = 3SE +/- 4.61, N = 3SE +/- 0.76, N = 3SE +/- 4.51, N = 31243.881637.721674.992229.572391.443179.323149.431356.311. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

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-ThreadedEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820K40K60K80K100KSE +/- 41.80, N = 3SE +/- 12.44, N = 3SE +/- 49.82, N = 3SE +/- 164.89, N = 3SE +/- 315.31, N = 3SE +/- 169.22, N = 3SE +/- 255.88, N = 3SE +/- 294.52, N = 332626.842668.643490.756873.257419.483365.180769.437214.41. (CXX) g++ options: -O3 -march=native -fPIE -pie

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Fused Multiply-AddEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248812M24M36M48M60MSE +/- 283045.68, N = 3SE +/- 319716.71, N = 3SE +/- 252666.10, N = 3SE +/- 28005.99, N = 3SE +/- 58656.32, N = 3SE +/- 500794.58, N = 3SE +/- 54521.03, N = 3SE +/- 267691.10, N = 322079544.9729040929.2429927440.6939475809.7842202851.9756410032.2555781750.8429762395.571. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lmpfr -lpthread -lrt -lsctp -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: AVX-512 VNNIEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488600K1200K1800K2400K3000KSE +/- 655.92, N = 3SE +/- 340.15, N = 3SE +/- 1225.91, N = 3SE +/- 97.63, N = 3SE +/- 1104.16, N = 3SE +/- 2539.18, N = 3SE +/- 5242.51, N = 3SE +/- 5501.16, N = 31008567.361315000.021371242.641758023.971909466.062571394.822511125.721125583.011. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lmpfr -lpthread -lrt -lsctp -lz

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-OnlyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248890180270360450SE +/- 0.44, N = 3SE +/- 0.34, N = 3SE +/- 0.40, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.18, N = 3SE +/- 0.14, N = 3SE +/- 1.05, N = 3421.36323.68315.13242.01236.28165.35168.04411.15

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: 2 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820K40K60K80K100KSE +/- 222.81, N = 3SE +/- 84.09, N = 3SE +/- 152.10, N = 3SE +/- 126.62, N = 3SE +/- 95.50, N = 3SE +/- 61.02, N = 3SE +/- 140.94, N = 3SE +/- 34.91, N = 39730773858722505372654202381923908982403

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-OnlyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 0.15, N = 3SE +/- 0.16, N = 3SE +/- 0.39, N = 3SE +/- 0.16, N = 3SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.12, N = 3SE +/- 0.24, N = 3170.00130.42127.0197.8495.4466.7867.99165.00

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 23.5Algorithm: MagiEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488140280420560700SE +/- 1.33, N = 3SE +/- 2.02, N = 3SE +/- 0.30, N = 3SE +/- 0.16, N = 3SE +/- 0.52, N = 3SE +/- 0.26, N = 3SE +/- 0.22, N = 3SE +/- 0.39, N = 3248.85328.70330.62453.87469.54633.44617.79278.091. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Chaos Group V-RAY

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

OpenBenchmarking.orgvsamples, More Is BetterChaos Group V-RAY 6.0Mode: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24889K18K27K36K45KSE +/- 50.00, N = 3SE +/- 22.00, N = 3SE +/- 11.00, N = 3SE +/- 61.57, N = 3SE +/- 54.69, N = 3SE +/- 204.38, N = 3SE +/- 96.96, N = 3SE +/- 16.84, N = 31727322505230783088531399438794376720060

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 ReadEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830M60M90M120M150MSE +/- 43642.52, N = 3SE +/- 63522.73, N = 3SE +/- 34435.41, N = 3SE +/- 36870.14, N = 3SE +/- 202059.67, N = 3SE +/- 286737.26, N = 3SE +/- 806005.16, N = 3SE +/- 170759.02, N = 35314637269443011716486839475027992006421134971626132843992578244261. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

SVT-HEVC

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 1 - Input: Bosphorus 1080pEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488612182430SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 39.1211.9012.2815.7115.8923.1022.119.161. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

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: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820K40K60K80K100KSE +/- 217.15, N = 3SE +/- 106.16, N = 3SE +/- 145.56, N = 3SE +/- 35.09, N = 3SE +/- 40.93, N = 3SE +/- 168.15, N = 3SE +/- 55.84, N = 3SE +/- 135.20, N = 39602372930715305287253426379234014281236

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Wide Vector MathEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488200K400K600K800K1000KSE +/- 31.25, N = 3SE +/- 84.63, N = 3SE +/- 205.90, N = 3SE +/- 1009.32, N = 3SE +/- 164.21, N = 3SE +/- 1811.04, N = 3SE +/- 875.08, N = 3SE +/- 373.64, N = 3307368.19402835.32413926.82539442.24568714.45778238.33749237.41576480.151. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -ljpeg -lmpfr -lpthread -lrt -lsctp -lz

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 23.5Algorithm: Myriad-GroestlEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882K4K6K8K10KSE +/- 8.61, N = 3SE +/- 3.31, N = 3SE +/- 6.11, N = 3SE +/- 39.39, N = 3SE +/- 5.30, N = 3SE +/- 13.33, N = 3SE +/- 3.33, N = 3SE +/- 30.23, N = 34487.285929.565972.098184.448480.4411337.0011207.004843.721. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

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: SharpenEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881326395265SE +/- 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 = 323303041415857361. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

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 digitEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488110220330440550SE +/- 0.52, N = 3SE +/- 0.31, N = 3SE +/- 0.18, N = 3SE +/- 0.29, N = 3SE +/- 0.10, N = 3SE +/- 0.37, N = 3SE +/- 0.19, N = 3SE +/- 0.23, N = 3504.27386.16375.92291.51268.44200.77204.02437.411. (CC) gcc options: -O2 -pthread

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: Road Segmentation ADAS FP16-INT8 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488816243240SE +/- 0.01, N = 3SE +/- 0.08, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 321.3116.8314.4817.0516.3416.7315.4536.35MIN: 11.09 / MAX: 25.08MIN: 8.78 / MAX: 22.96MIN: 8.91 / MAX: 26.44MIN: 9.42 / MAX: 28.42MIN: 9.56 / MAX: 24.55MIN: 9.66 / MAX: 30.96MIN: 9.45 / MAX: 28.8MIN: 24.38 / MAX: 96.41. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.2250.450.6750.91.125SE +/- 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 = 30.410.410.400.460.440.410.431.00MIN: 0.22 / MAX: 5.13MIN: 0.21 / MAX: 8.32MIN: 0.2 / MAX: 7.69MIN: 0.22 / MAX: 9.38MIN: 0.22 / MAX: 7.12MIN: 0.21 / MAX: 10.02MIN: 0.22 / MAX: 7.53MIN: 0.4 / MAX: 10.731. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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: BlowfishEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24889K18K27K36K45KSE +/- 17.13, N = 3SE +/- 3.00, N = 3SE +/- 44.47, N = 3SE +/- 156.57, N = 3SE +/- 7.77, N = 3SE +/- 39.87, N = 3SE +/- 175.99, N = 3SE +/- 68.54, N = 317621234032324331440328644379541889188411. (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: bcryptEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24889K18K27K36K45KSE +/- 12.53, N = 3SE +/- 7.88, N = 3SE +/- 43.32, N = 3SE +/- 157.44, N = 3SE +/- 12.90, N = 3SE +/- 104.20, N = 3SE +/- 89.13, N = 3SE +/- 127.68, N = 317654234072323531119328824369541961187871. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2

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: EnhancedEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 338495067649492491. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

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 MaterialEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850100150200250210.11160.01156.07120.91118.6285.3287.70174.95

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.39, N = 3SE +/- 0.24, N = 3SE +/- 0.10, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.08, N = 363.2953.4448.2658.5656.9858.4357.68118.45MIN: 45.67 / MAX: 70.41MIN: 25.81 / MAX: 61.36MIN: 29.16 / MAX: 69.56MIN: 42.61 / MAX: 80.74MIN: 40.42 / MAX: 73.08MIN: 31.85 / MAX: 94.55MIN: 32.9 / MAX: 82.05MIN: 68.57 / MAX: 244.711. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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: 1:100EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488800K1600K2400K3200K4000KSE +/- 2895.46, N = 3SE +/- 390.53, N = 3SE +/- 1471.26, N = 3SE +/- 2341.26, N = 3SE +/- 4210.88, N = 3SE +/- 5665.65, N = 3SE +/- 8167.06, N = 3SE +/- 11523.53, N = 31567216.542056776.092104490.552749241.802651410.813844031.283756154.983048853.561. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

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: MediumEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248848121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 36.538.358.6110.4110.2515.9215.076.62

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: allmodconfigEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488400800120016002000SE +/- 0.90, N = 3SE +/- 0.16, N = 3SE +/- 0.32, N = 3SE +/- 1.21, N = 3SE +/- 0.76, N = 3SE +/- 0.38, N = 3SE +/- 0.27, N = 3SE +/- 0.63, N = 31730.881319.931310.96976.20905.03714.38710.791508.04

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881.1252.253.3754.55.625SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 32.062.672.753.483.475.004.882.13MIN: 2 / MAX: 2.12MIN: 2.59 / MAX: 2.76MIN: 2.66 / MAX: 2.86MIN: 3.39 / MAX: 3.64MIN: 3.38 / MAX: 3.64MIN: 4.85 / MAX: 5.35MIN: 4.77 / MAX: 5.16MIN: 2.03 / MAX: 2.65

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-OnlyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.09230.18460.27690.36920.4615SE +/- 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 = 30.170.220.230.300.300.410.400.21

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: AlexNetEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248890180270360450SE +/- 0.09, N = 3SE +/- 0.04, N = 3SE +/- 0.55, N = 3SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 1.74, N = 3SE +/- 0.06, N = 3SE +/- 0.02, N = 3176.61238.89240.23315.30327.41425.47409.89192.28

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 ISPCEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488130260390520650SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3253327331446452603606288MIN: 19 / MAX: 3690MIN: 24 / MAX: 4788MIN: 25 / MAX: 4820MIN: 34 / MAX: 6230MIN: 34 / MAX: 5835MIN: 48 / MAX: 8450MIN: 47 / MAX: 8240MIN: 20 / MAX: 4338

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 ObjEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488714212835SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.10, N = 312.2015.6415.6921.2421.7827.3929.1714.55MIN: 12.12 / MAX: 12.41MIN: 15.52 / MAX: 15.91MIN: 15.56 / MAX: 16.03MIN: 21.05 / MAX: 21.73MIN: 21.53 / MAX: 22.24MIN: 27.11 / MAX: 28.25MIN: 28.89 / MAX: 30.01MIN: 14.14 / MAX: 16.68

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881020304050SE +/- 0.09, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.12, N = 319.8620.2219.1322.1120.3120.0920.5745.70MIN: 13.85 / MAX: 30.49MIN: 14.54 / MAX: 30.69MIN: 12.57 / MAX: 34MIN: 14.68 / MAX: 40.04MIN: 14.76 / MAX: 29.79MIN: 13.66 / MAX: 38.02MIN: 14.62 / MAX: 37.71MIN: 32.92 / MAX: 117.921. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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-OnlyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.19350.3870.58050.7740.9675SE +/- 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 = 30.360.470.480.640.640.860.840.42

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 2.2Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-OnlyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.19350.3870.58050.7740.9675SE +/- 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 = 30.360.470.480.640.640.860.840.42

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 DragonEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488816243240SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.12, N = 314.3618.3918.4525.0125.5332.2234.2416.58MIN: 14.27 / MAX: 14.68MIN: 18.28 / MAX: 18.74MIN: 18.3 / MAX: 18.87MIN: 24.79 / MAX: 25.66MIN: 25.23 / MAX: 26.17MIN: 31.88 / MAX: 33.23MIN: 33.88 / MAX: 35.24MIN: 16.29 / MAX: 19.34

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: MediumEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488510152025SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 39.5912.1012.4914.6614.4022.0220.659.281. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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: NinjaEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882004006008001000SE +/- 0.14, N = 3SE +/- 0.10, N = 3SE +/- 0.77, N = 3SE +/- 0.48, N = 3SE +/- 0.34, N = 3SE +/- 0.24, N = 3SE +/- 0.47, N = 3SE +/- 0.51, N = 3815.09644.77636.83464.51428.25354.76345.35697.94

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: SwirlEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850100150200250SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 31061371391811632502431301. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

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: AlexNetEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248890180270360450SE +/- 0.14, N = 3SE +/- 0.05, N = 3SE +/- 0.09, N = 3SE +/- 0.07, N = 3SE +/- 0.35, N = 3SE +/- 0.09, N = 3SE +/- 0.23, N = 3SE +/- 0.06, N = 3174.06235.47237.57305.22294.91409.96396.85189.14

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: Zxing 1D/2D Barcode Image ProcessingEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830060090012001500SE +/- 10.99, N = 7SE +/- 2.15, N = 8SE +/- 2.37, N = 8SE +/- 2.11, N = 9SE +/- 3.68, N = 9SE +/- 4.09, N = 9SE +/- 3.84, N = 9SE +/- 30.56, N = 151380104110197817495896201053

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488246810SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 33.304.304.425.555.557.727.653.58MIN: 3.26 / MAX: 3.33MIN: 4.24 / MAX: 4.36MIN: 4.32 / MAX: 4.49MIN: 5.2 / MAX: 5.66MIN: 5.19 / MAX: 5.69MIN: 6.87 / MAX: 8.15MIN: 6.89 / MAX: 7.99MIN: 3.44 / MAX: 4.42

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24887001400210028003500SE +/- 0.60, N = 3SE +/- 3.47, N = 3SE +/- 1.64, N = 3SE +/- 2.12, N = 3SE +/- 1.07, N = 3SE +/- 1.43, N = 3SE +/- 1.75, N = 3SE +/- 9.27, N = 33148.812514.662491.141973.991890.001348.371355.102931.95MIN: 3143.3MIN: 2499.44MIN: 2459.9MIN: 1936.17MIN: 1885.76MIN: 1314.05MIN: 1323.99MIN: 2882.151. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

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 MetricEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248890K180K270K360K450K1703682217672261982751672948913971383841831941271. (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

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830060090012001500SE +/- 0.51, N = 3SE +/- 0.53, N = 3SE +/- 7.18, N = 3SE +/- 1.03, N = 3SE +/- 0.47, N = 3SE +/- 1.91, N = 3SE +/- 2.47, N = 3SE +/- 6.38, N = 31615.381310.591337.471036.15962.21706.65694.521548.71MIN: 1610.14MIN: 1303.57MIN: 1299.81MIN: 1006.91MIN: 959.35MIN: 679.48MIN: 671.73MIN: 1512.561. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

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 WritingEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488900K1800K2700K3600K4500KSE +/- 20912.63, N = 3SE +/- 6544.07, N = 3SE +/- 22460.23, N = 3SE +/- 10561.84, N = 3SE +/- 24911.17, N = 3SE +/- 7045.87, N = 3SE +/- 23577.88, N = 3SE +/- 3646.67, N = 3186552124898542519594313093430036164294631433459319427601. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

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 CompileEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882004006008001000SE +/- 0.13, N = 3SE +/- 0.61, N = 3SE +/- 0.38, N = 3SE +/- 0.29, N = 3SE +/- 0.11, N = 3SE +/- 0.68, N = 3SE +/- 0.55, N = 3SE +/- 1.08, N = 3916.47743.11739.34524.51482.80421.43395.44915.94

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 CompressionEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488246810SE +/- 0.047798, N = 6SE +/- 0.051557, N = 7SE +/- 0.045809, N = 15SE +/- 0.044427, N = 15SE +/- 0.036366, N = 9SE +/- 0.025811, N = 10SE +/- 0.022219, N = 15SE +/- 0.044479, N = 67.6173055.7018566.1544604.4479044.4083233.2998793.5215656.6124861. (CXX) g++ options: -O2 -pthread -lbz2 -lpthread

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 FastEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881122334455SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 4SE +/- 0.05, N = 4SE +/- 0.08, N = 320.3826.0426.6531.6731.2946.4143.8920.17

SVT-HEVC

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080pEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248870140210280350SE +/- 0.13, N = 7SE +/- 0.17, N = 8SE +/- 0.21, N = 8SE +/- 0.24, N = 9SE +/- 0.22, N = 9SE +/- 0.26, N = 10SE +/- 0.40, N = 10SE +/- 2.05, N = 15135.99173.19178.20216.67215.68311.95297.03168.391. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 4KEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100SE +/- 0.02, N = 4SE +/- 0.04, N = 4SE +/- 0.04, N = 5SE +/- 0.05, N = 5SE +/- 0.07, N = 5SE +/- 0.06, N = 6SE +/- 0.15, N = 6SE +/- 0.08, N = 443.5755.6657.2371.6472.0799.6197.7747.981. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488246810SE +/- 0.00480, N = 3SE +/- 0.00446, N = 3SE +/- 0.00421, N = 3SE +/- 0.02716, N = 3SE +/- 0.01511, N = 3SE +/- 0.01295, N = 3SE +/- 0.01094, N = 3SE +/- 0.02381, N = 35.214623.934453.910893.575683.345053.006363.113076.85318MIN: 4.63MIN: 3.52MIN: 3.5MIN: 2.86MIN: 2.98MIN: 2.4MIN: 2.51MIN: 3.931. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

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: 1:10EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488800K1600K2400K3200K4000KSE +/- 1442.09, N = 3SE +/- 2995.56, N = 3SE +/- 2761.61, N = 3SE +/- 1507.35, N = 3SE +/- 4264.77, N = 3SE +/- 5303.54, N = 3SE +/- 5211.99, N = 3SE +/- 7388.22, N = 31651621.672163987.012195688.492762180.282658380.053761626.613693215.303072373.871. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

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.BEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24885K10K15K20K25KSE +/- 2.11, N = 3SE +/- 9.02, N = 3SE +/- 12.84, N = 3SE +/- 246.27, N = 15SE +/- 51.70, N = 4SE +/- 253.63, N = 3SE +/- 245.10, N = 4SE +/- 44.80, N = 311155.6411203.8311152.3117215.9225252.3518660.9024785.6512750.971. (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

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881.24932.49863.74794.99726.2465SE +/- 0.00099, N = 9SE +/- 0.00062, N = 9SE +/- 0.00661, N = 9SE +/- 0.00652, N = 9SE +/- 0.00129, N = 9SE +/- 0.00405, N = 9SE +/- 0.00476, N = 9SE +/- 0.10597, N = 155.552543.948883.834123.454613.510652.461592.707824.94792MIN: 5.53MIN: 3.84MIN: 3.58MIN: 3.24MIN: 3.46MIN: 2.35MIN: 2.61MIN: 4.451. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

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: 1000EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830K60K90K120K150KSE +/- 150.02, N = 3SE +/- 87.10, N = 3SE +/- 98.21, N = 3SE +/- 117.22, N = 3SE +/- 194.32, N = 3SE +/- 301.50, N = 3SE +/- 280.22, N = 3SE +/- 46.60, N = 366910.9384832.8285990.12108719.12120690.43136360.72150874.9187132.221. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

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 CompileEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248890180270360450SE +/- 0.05, N = 3SE +/- 0.16, N = 3SE +/- 0.15, N = 3SE +/- 0.49, N = 3SE +/- 0.51, N = 3SE +/- 0.40, N = 3SE +/- 0.60, N = 3SE +/- 0.34, N = 3415.40330.82326.16246.66225.99188.53184.51364.42

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 FastEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881020304050SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 4SE +/- 0.02, N = 4SE +/- 0.06, N = 319.2924.6225.2529.8229.5342.6040.1518.97

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: EmilyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248870140210280350341.30264.67263.51202.53201.65152.30153.65323.27

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: 500EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830K60K90K120K150KSE +/- 14.73, N = 3SE +/- 64.09, N = 3SE +/- 191.32, N = 3SE +/- 41.74, N = 3SE +/- 88.44, N = 3SE +/- 288.60, N = 3SE +/- 336.44, N = 3SE +/- 173.47, N = 370866.4291569.8794947.75116236.16124133.88154835.06158717.7492582.291. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

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: Road Segmentation ADAS FP16 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488918273645SE +/- 0.06, N = 3SE +/- 0.12, N = 3SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 327.2323.2717.7919.6817.7625.1518.3239.48MIN: 12.7 / MAX: 33.19MIN: 8.9 / MAX: 35.88MIN: 9.38 / MAX: 30.36MIN: 10.8 / MAX: 33.75MIN: 11.78 / MAX: 28.07MIN: 11.63 / MAX: 50.5MIN: 10.07 / MAX: 36.96MIN: 22.02 / MAX: 90.71. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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 FastEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881224364860SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 4SE +/- 0.03, N = 4SE +/- 0.08, N = 5SE +/- 0.17, N = 4SE +/- 0.01, N = 324.5931.3031.8637.2236.7154.5451.6324.69

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: Unix MakefilesEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882004006008001000SE +/- 0.63, N = 3SE +/- 0.24, N = 3SE +/- 0.38, N = 3SE +/- 0.73, N = 3SE +/- 0.93, N = 3SE +/- 1.00, N = 3SE +/- 2.07, N = 3SE +/- 0.50, N = 3840.92669.64662.52491.50453.31387.01379.72719.39

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 WritingEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881.1M2.2M3.3M4.4M5.5MSE +/- 17743.52, N = 15SE +/- 24740.75, N = 15SE +/- 14207.26, N = 3SE +/- 18321.18, N = 3SE +/- 55041.91, N = 13SE +/- 26941.48, N = 3SE +/- 35376.76, N = 15SE +/- 20209.24, N = 3240024331896503194971378371937811665119891526262825488371. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

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: defconfigEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.29, N = 3SE +/- 0.40, N = 3SE +/- 0.32, N = 3SE +/- 0.47, N = 3SE +/- 0.38, N = 3SE +/- 0.36, N = 3SE +/- 0.35, N = 3SE +/- 0.45, N = 3126.0098.2997.4175.2570.3057.4757.48108.34

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 FastEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881428425670SE +/- 0.04, N = 3SE +/- 0.05, N = 4SE +/- 0.05, N = 4SE +/- 0.04, N = 4SE +/- 0.03, N = 4SE +/- 0.07, N = 5SE +/- 0.03, N = 5SE +/- 0.31, N = 430.1037.7038.8244.4643.5462.5658.2528.641. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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 TimeEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248870140210280350307.01275.10296.60210.39168.98183.23141.46242.941. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

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: 200EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830K60K90K120K150KSE +/- 50.61, N = 3SE +/- 78.04, N = 3SE +/- 127.61, N = 3SE +/- 69.09, N = 3SE +/- 32.54, N = 3SE +/- 150.85, N = 3SE +/- 28.66, N = 3SE +/- 231.66, N = 375062.5197227.42101214.88118177.50123864.16162348.33160748.1898692.951. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

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: 100EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488714212835SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 312.8516.3616.6620.7419.9427.7526.4113.651. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

SPECFEM3D

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

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Tomographic ModelEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881428425670SE +/- 0.26, N = 3SE +/- 0.53, N = 3SE +/- 0.48, N = 3SE +/- 0.32, N = 3SE +/- 0.25, N = 3SE +/- 0.20, N = 15SE +/- 0.34, N = 3SE +/- 0.06, N = 360.3847.6949.5839.2852.2330.1328.0652.891. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

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: 100EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488612182430SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 312.7016.2816.5220.5319.8127.2225.9113.481. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -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: BT.CEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248810K20K30K40K50KSE +/- 5.49, N = 3SE +/- 20.15, N = 3SE +/- 14.25, N = 3SE +/- 229.09, N = 3SE +/- 193.15, N = 3SE +/- 110.88, N = 3SE +/- 481.60, N = 5SE +/- 96.46, N = 322331.4622976.5322935.2737547.7640212.6544844.1347856.8728326.361. (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

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_bareEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.60351.2071.81052.4143.0175SE +/- 0.001, N = 3SE +/- 0.002, N = 3SE +/- 0.002, N = 3SE +/- 0.008, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 3SE +/- 0.002, N = 3SE +/- 0.000, N = 31.2531.5221.5462.1232.2882.5432.6821.4341. (CXX) g++ options: -O3 -lm

7-Zip Compression

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

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Compression RatingEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248840K80K120K160K200KSE +/- 155.58, N = 3SE +/- 134.28, N = 3SE +/- 195.02, N = 3SE +/- 160.29, N = 3SE +/- 350.41, N = 3SE +/- 313.05, N = 3SE +/- 738.67, N = 3SE +/- 436.66, N = 382933104248103571139935141565170888177125949901. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

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 6.1Time To CompileEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881224364860SE +/- 0.13, N = 3SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.17, N = 3SE +/- 0.03, N = 3SE +/- 0.12, N = 3SE +/- 0.06, N = 3SE +/- 0.33, N = 353.8842.9442.2033.0330.5525.3025.2550.05

SPECFEM3D

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

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Homogeneous HalfspaceEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100SE +/- 1.09, N = 3SE +/- 0.54, N = 3SE +/- 0.35, N = 3SE +/- 0.57, N = 3SE +/- 0.43, N = 3SE +/- 0.28, N = 3SE +/- 0.48, N = 3SE +/- 0.41, N = 377.2359.6363.8750.7067.5639.5636.3167.111. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

SVT-VP9

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: Visual Quality Optimized - Input: Bosphorus 1080pEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248880160240320400SE +/- 0.64, N = 9SE +/- 0.69, N = 9SE +/- 0.76, N = 9SE +/- 1.08, N = 10SE +/- 1.06, N = 10SE +/- 1.71, N = 15SE +/- 2.24, N = 11SE +/- 3.49, N = 15173.33198.42203.16284.34281.56364.90357.60221.331. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

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 FastEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881122334455SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.09, N = 4SE +/- 0.02, N = 4SE +/- 0.19, N = 324.2230.1431.0435.2334.7148.1144.5923.051. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24883691215SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 36.605.044.785.645.325.125.109.94MIN: 4.29 / MAX: 16.53MIN: 3.47 / MAX: 17.02MIN: 3.18 / MAX: 16.31MIN: 3.48 / MAX: 18.22MIN: 3.77 / MAX: 11.07MIN: 3.61 / MAX: 15.43MIN: 3.41 / MAX: 14.16MIN: 6.2 / MAX: 41.651. (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.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: super-resolution-10 - Device: CPU - Executor: ParallelEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.80, N = 3SE +/- 0.21, N = 3SE +/- 0.18, N = 3SE +/- 0.21, N = 3SE +/- 0.43, N = 3SE +/- 1.10, N = 3SE +/- 0.51, N = 3SE +/- 0.57, N = 364.2989.2091.59102.6095.22133.32120.3575.881. (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: Vehicle Detection FP16 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248848121620SE +/- 0.06, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 310.949.138.448.787.7511.367.8916.04MIN: 4.95 / MAX: 20.28MIN: 4.28 / MAX: 20.17MIN: 4.55 / MAX: 19.23MIN: 4.89 / MAX: 24.3MIN: 4.8 / MAX: 13.5MIN: 4.12 / MAX: 36.81MIN: 4.2 / MAX: 25.04MIN: 9.49 / MAX: 56.981. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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 3.4Video Input: Bosphorus 4KEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488816243240SE +/- 0.19, N = 3SE +/- 0.22, N = 3SE +/- 0.15, N = 15SE +/- 0.16, N = 3SE +/- 0.11, N = 3SE +/- 0.08, N = 3SE +/- 0.20, N = 3SE +/- 0.17, N = 317.6016.7717.6127.4327.2034.6534.0219.481. (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 1080p - Video Preset: Super FastEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 0.03, N = 6SE +/- 0.05, N = 7SE +/- 0.08, N = 7SE +/- 0.05, N = 8SE +/- 0.09, N = 8SE +/- 0.16, N = 9SE +/- 0.07, N = 9SE +/- 0.97, N = 1588.44116.21119.83127.69124.52181.83168.1396.33

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 FastEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100SE +/- 0.02, N = 4SE +/- 0.04, N = 4SE +/- 0.04, N = 4SE +/- 0.03, N = 5SE +/- 0.06, N = 5SE +/- 0.07, N = 6SE +/- 0.04, N = 6SE +/- 0.30, N = 1041.1150.7152.3057.6856.2881.2374.9639.591. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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 CompileEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488816243240SE +/- 0.16, N = 3SE +/- 0.11, N = 3SE +/- 0.11, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 3SE +/- 0.21, N = 3SE +/- 0.01, N = 3SE +/- 0.07, N = 335.7128.6228.2622.1221.0517.4617.6732.52

SPECFEM3D

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

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Layered HalfspaceEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 1.62, N = 3SE +/- 1.33, N = 3SE +/- 0.49, N = 3SE +/- 0.17, N = 3SE +/- 0.40, N = 3SE +/- 0.14, N = 3SE +/- 0.13, N = 3SE +/- 1.26, N = 3166.97133.47133.31103.73113.1783.9781.88144.541. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488246810SE +/- 0.03028, N = 7SE +/- 0.01592, N = 7SE +/- 0.00717, N = 7SE +/- 0.01722, N = 7SE +/- 0.00824, N = 7SE +/- 0.01415, N = 7SE +/- 0.00632, N = 7SE +/- 0.04485, N = 77.934557.792028.168006.463574.029366.407194.245326.38053MIN: 7.8MIN: 7.68MIN: 7.7MIN: 6.03MIN: 3.95MIN: 5.93MIN: 3.87MIN: 5.921. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

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 CompileEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488120240360480600SE +/- 0.50, N = 3SE +/- 0.25, N = 3SE +/- 0.34, N = 3SE +/- 1.21, N = 3SE +/- 1.50, N = 3SE +/- 0.80, N = 3SE +/- 3.30, N = 4SE +/- 0.89, N = 3540.89438.62432.22339.22320.09273.49267.08468.38

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 4KEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881632486480SE +/- 0.11, N = 3SE +/- 0.07, N = 4SE +/- 0.10, N = 4SE +/- 0.05, N = 4SE +/- 0.07, N = 4SE +/- 0.15, N = 5SE +/- 0.11, N = 5SE +/- 0.30, N = 336.7045.9547.2957.2356.9672.7469.9636.141. (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: AlexNetEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248870140210280350SE +/- 0.07, N = 3SE +/- 0.08, N = 3SE +/- 0.21, N = 3SE +/- 0.15, N = 3SE +/- 0.15, N = 3SE +/- 0.34, N = 3SE +/- 0.33, N = 3SE +/- 0.21, N = 3160.28204.83206.40256.65248.73322.53315.45173.78

SPECFEM3D

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

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Mount St. HelensEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881428425670SE +/- 0.61, N = 5SE +/- 0.47, N = 3SE +/- 0.25, N = 3SE +/- 0.26, N = 3SE +/- 0.12, N = 3SE +/- 0.29, N = 15SE +/- 0.35, N = 5SE +/- 0.34, N = 362.5751.5951.2939.0147.8533.1331.1253.291. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

SVT-HEVC

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080pEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488120240360480600SE +/- 0.23, N = 10SE +/- 0.44, N = 11SE +/- 0.24, N = 11SE +/- 0.29, N = 11SE +/- 0.39, N = 11SE +/- 1.04, N = 12SE +/- 0.81, N = 12SE +/- 4.53, N = 15279.57351.91361.05428.68422.86560.81554.16383.641. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

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: 800 - Mode: Read Write - Average LatencyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488714212835SE +/- 0.34, N = 3SE +/- 0.16, N = 3SE +/- 0.09, N = 3SE +/- 0.23, N = 3SE +/- 0.19, N = 7SE +/- 0.16, N = 3SE +/- 0.17, N = 3SE +/- 0.06, N = 331.4424.7924.5020.0920.2116.1615.6823.201. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 800 - Mode: Read WriteEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248811K22K33K44K55KSE +/- 280.73, N = 3SE +/- 206.50, N = 3SE +/- 123.75, N = 3SE +/- 454.20, N = 3SE +/- 369.77, N = 7SE +/- 481.81, N = 3SE +/- 549.43, N = 3SE +/- 85.58, N = 325448322803265739830396084952251022344861. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

SPECFEM3D

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

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Water-layered HalfspaceEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.07, N = 3SE +/- 0.81, N = 3SE +/- 0.42, N = 3SE +/- 0.52, N = 3SE +/- 0.24, N = 3SE +/- 0.06, N = 3SE +/- 0.54, N = 3SE +/- 0.21, N = 3156.91130.20129.33104.12131.8785.2578.38135.421. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

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: 8 - Compression SpeedEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882004006008001000SE +/- 2.29, N = 3SE +/- 1.07, N = 3SE +/- 1.88, N = 3SE +/- 2.23, N = 3SE +/- 6.53, N = 3SE +/- 6.27, N = 9SE +/- 2.98, N = 3SE +/- 3.15, N = 3500.0525.1520.3798.4942.3793.8970.7485.61. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8, Long Mode - Compression SpeedEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882004006008001000SE +/- 1.62, N = 3SE +/- 0.50, N = 3SE +/- 0.95, N = 3SE +/- 1.33, N = 3SE +/- 3.16, N = 3SE +/- 3.49, N = 3SE +/- 3.15, N = 3SE +/- 1.12, N = 3482.9532.3527.9745.5888.7797.0961.2504.11. (CC) gcc options: -O3 -pthread -lz -llzma

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: 6EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488246810SE +/- 0.010, N = 6SE +/- 0.005, N = 7SE +/- 0.014, N = 7SE +/- 0.014, N = 8SE +/- 0.008, N = 8SE +/- 0.010, N = 9SE +/- 0.018, N = 9SE +/- 0.088, N = 156.7985.4365.3384.4374.4693.4213.5686.2601. (CXX) g++ options: -O3 -fPIC -lm

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: Noise-GaussianEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.00, N = 3SE +/- 1.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 36073768987119115701. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

SVT-VP9

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: Visual Quality Optimized - Input: Bosphorus 4KEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100SE +/- 0.51, N = 5SE +/- 0.64, N = 5SE +/- 0.56, N = 5SE +/- 0.79, N = 15SE +/- 0.75, N = 15SE +/- 1.42, N = 15SE +/- 1.39, N = 15SE +/- 0.47, N = 1550.5160.9761.9080.9380.87100.00100.0559.561. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

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 1080p - Video Preset: Very FastEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 0.04, N = 6SE +/- 0.04, N = 7SE +/- 0.14, N = 7SE +/- 0.05, N = 7SE +/- 0.08, N = 7SE +/- 0.11, N = 9SE +/- 0.18, N = 8SE +/- 0.95, N = 1584.36108.21111.78117.12114.08166.81154.0289.31

SVT-VP9

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 1080pEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248890180270360450SE +/- 0.18, N = 9SE +/- 0.18, N = 10SE +/- 0.20, N = 10SE +/- 0.25, N = 10SE +/- 0.34, N = 10SE +/- 0.58, N = 11SE +/- 1.10, N = 11SE +/- 0.18, N = 10208.12237.22242.58331.83330.16408.42410.42289.261. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

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 1080p - Video Preset: Super FastEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850100150200250SE +/- 0.06, N = 8SE +/- 0.04, N = 9SE +/- 0.09, N = 9SE +/- 0.12, N = 9SE +/- 0.10, N = 9SE +/- 0.26, N = 10SE +/- 0.14, N = 10SE +/- 2.29, N = 15126.08161.51168.08172.91166.31248.39224.02139.721. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

SVT-VP9

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248890180270360450SE +/- 0.22, N = 9SE +/- 0.35, N = 10SE +/- 0.28, N = 10SE +/- 0.37, N = 11SE +/- 0.48, N = 10SE +/- 0.42, N = 11SE +/- 0.58, N = 11SE +/- 0.53, N = 10213.44246.28251.44337.46336.78419.26419.93295.691. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

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 1080p - Video Preset: MediumEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881428425670SE +/- 0.04, N = 3SE +/- 0.03, N = 4SE +/- 0.03, N = 4SE +/- 0.06, N = 4SE +/- 0.06, N = 4SE +/- 0.07, N = 5SE +/- 0.10, N = 5SE +/- 0.14, N = 432.6344.4345.6349.2447.9864.1659.4836.59

ScyllaDB

This is a benchmark of ScyllaDB and is making use of Apache Cassandra's cassandra-stress for conducting the benchmark. ScyllaDB is an open-source distributed NoSQL data store that is compatible with Apache Cassandra while focusing on higher throughput and lower latency. ScyllaDB uses a sharded design on each node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterScyllaDB 5.2.9Test: WritesEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830K60K90K120K150KSE +/- 110.37, N = 3SE +/- 140.30, N = 3SE +/- 60.93, N = 3SE +/- 83.55, N = 3SE +/- 415.80, N = 3SE +/- 255.51, N = 3SE +/- 369.77, N = 3SE +/- 395.91, N = 3780049334294217112383115980151729145533104832

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 1080p - Video Preset: Ultra FastEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 0.06, N = 7SE +/- 0.16, N = 8SE +/- 0.06, N = 8SE +/- 0.12, N = 8SE +/- 0.14, N = 8SE +/- 0.20, N = 9SE +/- 0.12, N = 9SE +/- 1.21, N = 15105.23138.95142.97150.51146.08203.39187.73115.40

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 FP32 - Device: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.39, N = 3SE +/- 0.18, N = 3SE +/- 0.13, N = 3SE +/- 0.38, N = 3SE +/- 0.06, N = 3SE +/- 0.48, N = 3SE +/- 0.38, N = 3SE +/- 0.77, N = 390.0276.5272.7080.7970.3190.4983.18134.54MIN: 43.62 / MAX: 98.59MIN: 52.86 / MAX: 94.9MIN: 40.38 / MAX: 84.07MIN: 30.58 / MAX: 108.27MIN: 50.95 / MAX: 96.87MIN: 55.81 / MAX: 127.95MIN: 38.33 / MAX: 120MIN: 98.92 / MAX: 233.451. (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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248848121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 38.498.837.959.288.099.908.4015.21MIN: 5.12 / MAX: 20.17MIN: 5.3 / MAX: 14.21MIN: 5 / MAX: 23.85MIN: 5.38 / MAX: 30.18MIN: 5.45 / MAX: 16.52MIN: 6.53 / MAX: 37.69MIN: 5.53 / MAX: 31MIN: 10.08 / MAX: 64.691. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248848121620SE +/- 0.01, N = 3SE +/- 0.03, N = 4SE +/- 0.04, N = 4SE +/- 0.05, N = 4SE +/- 0.03, N = 4SE +/- 0.06, N = 5SE +/- 0.05, N = 5SE +/- 0.03, N = 38.8811.5911.9312.7012.3016.9816.439.35MIN: 8.44 / MAX: 8.94MIN: 10.79 / MAX: 11.71MIN: 11.1 / MAX: 12.12MIN: 11.78 / MAX: 12.96MIN: 11.45 / MAX: 12.4MIN: 15.46 / MAX: 17.3MIN: 15.06 / MAX: 16.65MIN: 8.52 / MAX: 10.72

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: 1BEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488816243240SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.21, N = 331.3226.1625.4620.2519.3518.2117.7333.89

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.11, N = 3SE +/- 0.06, N = 3SE +/- 0.15, N = 3SE +/- 0.10, N = 3SE +/- 0.09, N = 3SE +/- 1.01, N = 3SE +/- 0.45, N = 3SE +/- 0.73, N = 390.5076.3672.5281.5670.4489.9883.30133.99MIN: 68.02 / MAX: 98.61MIN: 56.38 / MAX: 88.42MIN: 28.47 / MAX: 87.6MIN: 55.87 / MAX: 108.66MIN: 48.46 / MAX: 97.28MIN: 63.39 / MAX: 124.53MIN: 29.23 / MAX: 114.06MIN: 103.21 / MAX: 233.721. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -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: 0EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.29, N = 3SE +/- 0.27, N = 3SE +/- 0.18, N = 3SE +/- 0.10, N = 3SE +/- 1.17, N = 3SE +/- 0.33, N = 3SE +/- 0.77, N = 5SE +/- 0.49, N = 3136.63110.46108.6589.6391.5972.1174.06133.571. (CXX) g++ options: -O3 -fPIC -lm

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: 128EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830K60K90K120K150KSE +/- 73.92, N = 3SE +/- 276.11, N = 3SE +/- 302.28, N = 3SE +/- 58.24, N = 3SE +/- 391.29, N = 3SE +/- 765.52, N = 3SE +/- 158.53, N = 3SE +/- 237.27, N = 363936.282706.283155.892280.790283.6121061.7118066.783987.5

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.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: ArcFace ResNet-100 - Device: CPU - Executor: ParallelEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488918273645SE +/- 0.12, N = 3SE +/- 0.18, N = 3SE +/- 0.11, N = 3SE +/- 0.27, N = 7SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.12, N = 3SE +/- 0.11, N = 319.8426.8627.0230.1929.3737.2734.7223.041. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

SVT-HEVC

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 4KEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 0.10, N = 6SE +/- 0.10, N = 6SE +/- 0.12, N = 6SE +/- 0.15, N = 7SE +/- 0.17, N = 7SE +/- 0.14, N = 8SE +/- 0.31, N = 8SE +/- 1.10, N = 1586.79107.47109.03130.38131.05158.94162.47101.831. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

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 1080p - Video Preset: MediumEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100SE +/- 0.05, N = 4SE +/- 0.05, N = 5SE +/- 0.04, N = 5SE +/- 0.05, N = 5SE +/- 0.10, N = 5SE +/- 0.08, N = 6SE +/- 0.13, N = 6SE +/- 0.37, N = 447.1762.7164.6866.4763.8588.3080.5250.471. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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 TesterEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200191.21146.67142.94129.09131.17102.15104.32173.41

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: leblancbigEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881020304050SE +/- 0.11, N = 3SE +/- 0.00, N = 3SE +/- 0.88, N = 15SE +/- 0.42, N = 15SE +/- 0.35, N = 3SE +/- 0.27, N = 3SE +/- 0.17, N = 3SE +/- 1.09, N = 1241.7135.4842.9431.8628.1429.5522.9732.871. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi

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: 16 - Buffer Length: 256 - Filter Length: 512EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248860M120M180M240M300MSE +/- 138604.15, N = 3SE +/- 170098.01, N = 3SE +/- 75055.53, N = 3SE +/- 1274497.72, N = 3SE +/- 168358.08, N = 3SE +/- 520000.00, N = 3SE +/- 2913213.19, N = 3SE +/- 1245404.89, N = 31564333332062700002116700002335966672301533332923900002825633331592700001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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: 128EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820K40K60K80K100KSE +/- 67.60, N = 3SE +/- 75.55, N = 3SE +/- 264.18, N = 3SE +/- 37.01, N = 3SE +/- 233.77, N = 3SE +/- 831.25, N = 3SE +/- 58.30, N = 3SE +/- 189.35, N = 349645.063986.264052.771721.169839.892195.290314.059866.6

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: 32 - Model: GoogLeNetEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.13, N = 3SE +/- 0.14, N = 3SE +/- 0.05, N = 3SE +/- 0.10, N = 374.9392.8593.71111.99115.48135.30139.1178.79

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: 128EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820K40K60K80K100KSE +/- 116.97, N = 3SE +/- 105.77, N = 3SE +/- 47.09, N = 3SE +/- 132.40, N = 3SE +/- 72.81, N = 3SE +/- 156.58, N = 3SE +/- 462.84, N = 3SE +/- 198.29, N = 351927.066756.967506.775445.773004.595084.896269.163021.4

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 500 - Mode: Read WriteEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248813K26K39K52K65KSE +/- 246.90, N = 3SE +/- 304.71, N = 3SE +/- 256.18, N = 3SE +/- 263.22, N = 3SE +/- 339.91, N = 3SE +/- 699.92, N = 3SE +/- 609.40, N = 3SE +/- 362.37, N = 331609395494047346671458525651758405424121. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 500 - Mode: Read Write - Average LatencyEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248848121620SE +/- 0.123, N = 3SE +/- 0.098, N = 3SE +/- 0.078, N = 3SE +/- 0.061, N = 3SE +/- 0.080, N = 3SE +/- 0.109, N = 3SE +/- 0.089, N = 3SE +/- 0.101, N = 315.82012.64412.35510.71410.9068.8508.56311.7911. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

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_bmEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100SE +/- 0.13, N = 3SE +/- 0.05, N = 3SE +/- 0.21, N = 3SE +/- 0.28, N = 3SE +/- 0.10, N = 3SE +/- 0.45, N = 3SE +/- 0.47, N = 6SE +/- 0.27, N = 377.2374.5986.8269.9885.7372.8847.0264.581. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

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: GoogLeNetEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.08, N = 3SE +/- 0.07, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 373.7791.2192.37110.52113.51133.21135.7878.17

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.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: fcn-resnet101-11 - Device: CPU - Executor: ParallelEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.39730.79461.19191.58921.9865SE +/- 0.007958, N = 3SE +/- 0.013735, N = 4SE +/- 0.018640, N = 15SE +/- 0.015352, N = 3SE +/- 0.014994, N = 3SE +/- 0.005924, N = 3SE +/- 0.015047, N = 15SE +/- 0.013081, N = 140.9594411.1281001.1605401.4134901.4932701.7634901.7658801.0428121. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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 1080p - Video Preset: Very FastEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 0.06, N = 7SE +/- 0.09, N = 7SE +/- 0.10, N = 8SE +/- 0.06, N = 8SE +/- 0.16, N = 8SE +/- 0.25, N = 9SE +/- 0.11, N = 9SE +/- 0.87, N = 7101.80123.89129.02131.13125.34186.56169.81103.041. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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 ProteinEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24883691215SE +/- 0.004, N = 9SE +/- 0.035, N = 10SE +/- 0.055, N = 9SE +/- 0.178, N = 15SE +/- 0.126, N = 15SE +/- 0.059, N = 11SE +/- 0.045, N = 11SE +/- 0.250, N = 156.8388.7156.8269.2669.52512.48111.7667.2361. (CXX) g++ options: -O3 -lm -ldl

NAS Parallel Benchmarks

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

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.CEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24885001000150020002500SE +/- 8.87, N = 6SE +/- 11.50, N = 10SE +/- 41.08, N = 15SE +/- 21.84, N = 15SE +/- 11.67, N = 7SE +/- 13.01, N = 8SE +/- 13.19, N = 8SE +/- 55.41, N = 151168.351489.701169.681542.111508.632123.372051.601470.811. (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

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: INIVOL and Fluid Structure Interaction Drop ContainerEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488130260390520650SE +/- 0.28, N = 3SE +/- 4.81, N = 6SE +/- 5.87, N = 9SE +/- 9.39, N = 6SE +/- 4.68, N = 5SE +/- 2.52, N = 3SE +/- 4.35, N = 9SE +/- 8.95, N = 9560.31494.79585.75403.46421.96322.67324.46538.42

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.CEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248810K20K30K40K50KSE +/- 11.58, N = 3SE +/- 38.55, N = 3SE +/- 613.23, N = 15SE +/- 221.05, N = 3SE +/- 13.15, N = 3SE +/- 521.71, N = 15SE +/- 266.87, N = 3SE +/- 714.33, N = 1526620.6940284.2337149.8841255.9843166.1047556.6948014.8640491.561. (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

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 4KEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488246810SE +/- 0.005, N = 3SE +/- 0.008, N = 3SE +/- 0.016, N = 3SE +/- 0.019, N = 3SE +/- 0.004, N = 3SE +/- 0.009, N = 3SE +/- 0.030, N = 3SE +/- 0.024, N = 34.5405.5635.7426.5266.4137.8387.2964.3481. (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: 512 - Model: GoogLeNetEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.00, N = 373.5890.1391.90109.73114.12132.62130.8076.13

SVT-VP9

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 4KEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.05, N = 5SE +/- 0.09, N = 5SE +/- 0.08, N = 5SE +/- 0.11, N = 6SE +/- 0.15, N = 6SE +/- 0.13, N = 6SE +/- 0.16, N = 6SE +/- 0.64, N = 1063.7674.7176.0697.2797.41111.01114.8883.191. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

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 1080p - Video Preset: Ultra FastEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248870140210280350SE +/- 0.11, N = 9SE +/- 0.14, N = 10SE +/- 0.14, N = 10SE +/- 0.14, N = 10SE +/- 0.16, N = 10SE +/- 0.21, N = 11SE +/- 0.34, N = 10SE +/- 4.30, N = 15167.39217.52226.13231.05220.31301.57271.84197.511. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

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: GoogLeNetEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.08, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 373.3690.1991.77109.36109.62132.13131.3476.84

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: sedovbigEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881428425670SE +/- 0.22, N = 3SE +/- 0.05, N = 3SE +/- 0.59, N = 6SE +/- 0.18, N = 3SE +/- 0.43, N = 3SE +/- 0.35, N = 8SE +/- 0.12, N = 3SE +/- 0.67, N = 459.0250.8961.9749.1842.1842.2334.6455.901. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi

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-50EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488918273645SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.31, N = 8SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 325.3630.4930.7035.0136.2341.0240.4822.98

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: 500MEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248848121620SE +/- 0.082, N = 4SE +/- 0.038, N = 4SE +/- 0.036, N = 4SE +/- 0.009, N = 5SE +/- 0.012, N = 5SE +/- 0.010, N = 5SE +/- 0.010, N = 5SE +/- 0.124, N = 414.13311.86111.5899.4148.9698.6528.37314.926

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.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: ResNet50 v1-12-int8 - Device: CPU - Executor: ParallelEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248880160240320400SE +/- 0.43, N = 3SE +/- 1.50, N = 3SE +/- 0.89, N = 3SE +/- 0.27, N = 3SE +/- 3.53, N = 4SE +/- 3.68, N = 3SE +/- 1.30, N = 3SE +/- 1.45, N = 3264.15314.23322.61336.91327.75380.82374.23214.551. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

SVT-VP9

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 4KEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100SE +/- 0.11, N = 5SE +/- 0.03, N = 5SE +/- 0.09, N = 5SE +/- 0.08, N = 6SE +/- 0.12, N = 6SE +/- 0.15, N = 6SE +/- 0.18, N = 6SE +/- 0.46, N = 1560.9070.3471.4892.0793.56103.52107.5379.851. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

libavif avifenc

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

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 2EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881428425670SE +/- 0.23, N = 3SE +/- 0.03, N = 3SE +/- 0.13, N = 3SE +/- 0.18, N = 3SE +/- 0.56, N = 4SE +/- 0.03, N = 3SE +/- 0.33, N = 7SE +/- 0.18, N = 364.1452.5651.9444.3445.7336.3337.1161.841. (CXX) g++ options: -O3 -fPIC -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: 256 - Model: ResNet-50EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488918273645SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 325.4330.5030.6334.7934.9540.6840.3323.07

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 64 - Model: ResNet-50EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488918273645SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 325.5030.6430.5334.6834.7640.2440.4923.10

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 32 - Model: ResNet-50EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488918273645SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 325.7031.0030.7934.7234.7839.9940.3723.19

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 32 - Model: AlexNetEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850100150200250SE +/- 0.15, N = 3SE +/- 0.06, N = 3SE +/- 0.23, N = 3SE +/- 0.20, N = 3SE +/- 0.25, N = 3SE +/- 0.14, N = 4SE +/- 0.35, N = 4SE +/- 1.37, N = 3139.36171.59172.89204.70199.14242.49239.93160.52

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 WindshieldEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248860120180240300SE +/- 0.57, N = 3SE +/- 0.53, N = 3SE +/- 7.66, N = 9SE +/- 4.67, N = 12SE +/- 1.41, N = 3SE +/- 2.93, N = 12SE +/- 3.16, N = 9SE +/- 3.59, N = 9291.79238.13266.65202.27229.49183.64168.39259.88

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 4KEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 0.10, N = 6SE +/- 0.12, N = 7SE +/- 0.14, N = 7SE +/- 0.26, N = 7SE +/- 0.18, N = 7SE +/- 0.33, N = 8SE +/- 0.17, N = 8SE +/- 1.00, N = 1598.45121.50125.28145.04143.28170.21166.37105.281. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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: FastEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488246810SE +/- 0.017, N = 3SE +/- 0.031, N = 3SE +/- 0.021, N = 3SE +/- 0.029, N = 3SE +/- 0.034, N = 3SE +/- 0.058, N = 3SE +/- 0.045, N = 3SE +/- 0.018, N = 34.5425.5545.6936.4906.4527.8187.4944.9751. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects

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, LosslessEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24883691215SE +/- 0.012, N = 5SE +/- 0.006, N = 6SE +/- 0.021, N = 6SE +/- 0.018, N = 6SE +/- 0.038, N = 6SE +/- 0.044, N = 7SE +/- 0.051, N = 9SE +/- 0.014, N = 510.0248.2628.1477.2157.1685.8906.1748.9691. (CXX) g++ options: -O3 -fPIC -lm

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.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: bertsquad-12 - Device: CPU - Executor: ParallelEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248848121620SE +/- 0.02536, N = 3SE +/- 0.06035, N = 3SE +/- 0.11789, N = 3SE +/- 0.10273, N = 9SE +/- 0.15346, N = 3SE +/- 0.05371, N = 3SE +/- 0.09843, N = 14SE +/- 0.10358, N = 39.1545812.3239012.5483012.7624012.6118015.5560014.7664011.537201. (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: ParallelEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24883691215SE +/- 0.08117, N = 3SE +/- 0.11284, N = 4SE +/- 0.10193, N = 15SE +/- 0.06480, N = 12SE +/- 0.01052, N = 3SE +/- 0.13812, N = 4SE +/- 0.09061, N = 9SE +/- 0.11600, N = 36.782269.222199.291799.183408.9895811.5199011.038008.094211. (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: ResizingEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248870140210280350SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.88, N = 3SE +/- 0.33, N = 3SE +/- 1.00, N = 3SE +/- 0.00, N = 3SE +/- 1.86, N = 32022502552842583383252441. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

MariaDB

This is a MariaDB MySQL database server benchmark making use of sysbench rather than the existing pts/mysqlslap test profile that uses MariaDB with mysqlslap/mariadb-slap as the benchmark driver. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 11.5Test: oltp_write_only - Threads: 64EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248870K140K210K280K350KSE +/- 493.89, N = 3SE +/- 204.75, N = 3SE +/- 419.19, N = 3SE +/- 431.10, N = 3SE +/- 971.36, N = 3SE +/- 56.85, N = 3SE +/- 391.18, N = 3SE +/- 798.38, N = 3208039260139265119273914279043348099346495296034-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-lnuma -lpcre2-8 -lcrypt -laio -lssl -lcrypto1. (CXX) g++ options: -fPIC -pie -fstack-protector -O3 -lpthread -lz -ldl -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: particle_volume/pathtracer/real_timeEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850100150200250SE +/- 0.31, N = 3SE +/- 0.28, N = 3SE +/- 0.61, N = 3SE +/- 0.64, N = 3SE +/- 1.13, N = 3SE +/- 0.68, N = 3SE +/- 1.24, N = 3SE +/- 0.42, N = 3132.40158.59170.39193.97182.05220.91209.22157.97

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: FasterEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248848121620SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 310.4912.8013.1414.5414.6017.4416.8211.771. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects

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: 12 - Compression SpeedEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248860120180240300SE +/- 0.43, N = 3SE +/- 0.62, N = 3SE +/- 0.92, N = 3SE +/- 2.73, N = 3SE +/- 1.66, N = 3SE +/- 2.97, N = 3SE +/- 6.73, N = 15SE +/- 2.27, N = 3164.4177.9171.0221.2273.1220.9248.0170.61. (CC) gcc options: -O3 -pthread -lz -llzma

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 Random Write RandomEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488600K1200K1800K2400K3000KSE +/- 24161.52, N = 3SE +/- 1649.04, N = 3SE +/- 5891.62, N = 3SE +/- 16096.70, N = 3SE +/- 1307.98, N = 3SE +/- 24549.22, N = 3SE +/- 8693.67, N = 3SE +/- 17144.87, N = 3178179522474812131122227855618177752915758292956119765331. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

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: 90EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881122334455SE +/- 0.08, N = 3SE +/- 0.16, N = 3SE +/- 0.36, N = 3SE +/- 0.44, N = 3SE +/- 0.30, N = 10SE +/- 0.28, N = 3SE +/- 0.24, N = 3SE +/- 0.19, N = 329.6535.0836.0240.1838.7248.5345.3831.811. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: JPEG - Quality: 90EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881122334455SE +/- 0.13, N = 3SE +/- 0.01, N = 3SE +/- 0.30, N = 3SE +/- 0.46, N = 3SE +/- 0.38, N = 3SE +/- 0.37, N = 10SE +/- 0.47, N = 6SE +/- 0.09, N = 330.1935.6037.1841.7740.0148.5947.0532.511. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

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.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: CaffeNet 12-int8 - Device: CPU - Executor: ParallelEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882004006008001000SE +/- 5.55, N = 6SE +/- 2.32, N = 3SE +/- 7.12, N = 3SE +/- 7.33, N = 3SE +/- 5.20, N = 3SE +/- 8.14, N = 3SE +/- 8.12, N = 7SE +/- 3.86, N = 3583.98672.35677.58719.99769.69829.19895.93557.411. (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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881.14082.28163.42244.56325.704SE +/- 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 +/- 0.01, N = 33.273.353.163.643.533.333.515.07MIN: 1.84 / MAX: 8.95MIN: 1.81 / MAX: 10.37MIN: 1.77 / MAX: 13.91MIN: 1.95 / MAX: 15.55MIN: 1.95 / MAX: 9.62MIN: 1.88 / MAX: 15.67MIN: 1.92 / MAX: 13.78MIN: 2.36 / MAX: 27.991. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

NAS Parallel Benchmarks

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

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.CEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24883K6K9K12K15KSE +/- 7.67, N = 3SE +/- 5.49, N = 3SE +/- 1.44, N = 3SE +/- 27.04, N = 3SE +/- 10.77, N = 3SE +/- 146.76, N = 5SE +/- 188.19, N = 3SE +/- 4.20, N = 310028.7410088.5010027.1111606.2214130.3013374.3515940.6111387.051. (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

Xcompact3d Incompact3d

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

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 129 Cells Per DirectionEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488510152025SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.18, N = 9SE +/- 0.17, N = 15SE +/- 0.15, N = 15SE +/- 0.12, N = 15SE +/- 0.11, N = 15SE +/- 0.22, N = 321.8020.3921.8717.2914.7616.5913.8219.751. (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

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 1080pEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 0.16, N = 7SE +/- 0.16, N = 8SE +/- 0.13, N = 8SE +/- 0.26, N = 8SE +/- 0.18, N = 8SE +/- 0.29, N = 9SE +/- 0.32, N = 8SE +/- 1.11, N = 15119.76142.84148.34162.42156.68189.41175.21120.531. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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 CompileEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881428425670SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.10, N = 3SE +/- 0.24, N = 3SE +/- 0.21, N = 3SE +/- 0.18, N = 3SE +/- 0.22, N = 3SE +/- 0.20, N = 363.3054.0453.7645.5943.9940.1140.1054.45

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 Random Write RandomEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488600K1200K1800K2400K3000KSE +/- 2410.81, N = 3SE +/- 1251.53, N = 3SE +/- 1406.43, N = 3SE +/- 6109.05, N = 3SE +/- 284.11, N = 3SE +/- 2327.92, N = 3SE +/- 1095.04, N = 3SE +/- 4914.43, N = 3180641022225562138776216669918419702770195270638521209151. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

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 NanotubeEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850100150200250SE +/- 0.19, N = 3SE +/- 0.06, N = 3SE +/- 2.92, N = 3SE +/- 2.27, N = 3SE +/- 1.91, N = 3SE +/- 1.88, N = 3SE +/- 3.67, N = 10SE +/- 8.24, N = 12237.24203.43205.87177.02191.34154.99154.85228.901. (CC) gcc options: -shared -fwrapv -O2 -lxc -lblas -lmpi

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 1080pEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488612182430SE +/- 0.02, N = 5SE +/- 0.04, N = 5SE +/- 0.07, N = 5SE +/- 0.04, N = 6SE +/- 0.07, N = 6SE +/- 0.13, N = 6SE +/- 0.08, N = 6SE +/- 0.15, N = 516.5419.5519.8921.5120.7524.1722.1715.921. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

MariaDB

This is a MariaDB MySQL database server benchmark making use of sysbench rather than the existing pts/mysqlslap test profile that uses MariaDB with mysqlslap/mariadb-slap as the benchmark driver. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 11.5Test: oltp_read_write - Threads: 64EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248840K80K120K160K200KSE +/- 193.47, N = 3SE +/- 284.19, N = 3SE +/- 211.83, N = 3SE +/- 99.82, N = 3SE +/- 296.87, N = 3SE +/- 389.87, N = 3SE +/- 91.63, N = 3SE +/- 562.40, N = 3133141166060168593169625172265201182197194186418-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-lnuma -lpcre2-8 -lcrypt -laio -lssl -lcrypto1. (CXX) g++ options: -fPIC -pie -fstack-protector -O3 -lpthread -lz -ldl -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: PUSCH Processor Benchmark, Throughput TotalEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830060090012001500SE +/- 0.65, N = 6SE +/- 0.05, N = 5SE +/- 0.10, N = 6SE +/- 0.14, N = 4SE +/- 0.05, N = 4SE +/- 0.05, N = 4SE +/- 18.52, N = 4SE +/- 9.26, N = 151206.01469.81470.51351.91273.91611.51592.31068.9-mavx512f -mavx512cd -mavx512bw -mavx512dq - MIN: 763.8 / MAX: 1207.4-mavx512f -mavx512cd -mavx512bw -mavx512dq - MIN: 849.1 / MAX: 1470-mavx512f -mavx512cd -mavx512bw -mavx512dq - MIN: 849.1 / MAX: 1470.7-mavx512f -mavx512cd -mavx512bw -mavx512dq - MIN: 955 / MAX: 1352.2-mavx512f -mavx512cd -mavx512bw -mavx512dq - MIN: 849.1 / MAX: 1274-mavx512f -mavx512cd -mavx512bw -mavx512dq - MIN: 1019.4 / MAX: 1611.6-mavx512f -mavx512cd -mavx512bw -mavx512dq - MIN: 927.1 / MAX: 1610.9MIN: 637.2 / MAX: 1111.51. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -fno-trapping-math -fno-math-errno -ldl

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 1080p - Video Preset: FastEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488612182430SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.08, N = 3SE +/- 0.08, N = 3SE +/- 0.10, N = 3SE +/- 0.07, N = 3SE +/- 0.10, N = 3SE +/- 0.06, N = 315.5418.3018.7720.2119.9423.1921.7215.461. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects

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 EngineEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24887001400210028003500SE +/- 20.21, N = 5SE +/- 19.37, N = 5SE +/- 18.64, N = 5SE +/- 18.41, N = 5SE +/- 7.90, N = 5SE +/- 9.53, N = 5SE +/- 16.10, N = 5SE +/- 50.09, N = 1528292391240923742313208021273113

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 - Compression SpeedEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24883691215SE +/- 0.00, N = 3SE +/- 0.06, N = 3SE +/- 0.00, N = 3SE +/- 0.15, N = 3SE +/- 0.15, N = 3SE +/- 0.09, N = 3SE +/- 0.14, N = 5SE +/- 0.01, N = 310.2010.5010.2011.5013.2010.7012.908.911. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19 - Compression SpeedEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488612182430SE +/- 0.09, N = 3SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.17, N = 15SE +/- 0.23, N = 15SE +/- 0.03, N = 3SE +/- 0.48, N = 12SE +/- 0.03, N = 318.018.318.020.024.018.822.516.31. (CC) gcc options: -O3 -pthread -lz -llzma

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 1080pEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488140280420560700SE +/- 0.74, N = 12SE +/- 0.93, N = 12SE +/- 0.74, N = 12SE +/- 1.70, N = 12SE +/- 0.82, N = 12SE +/- 1.66, N = 15SE +/- 1.39, N = 12SE +/- 0.76, N = 12449.57528.23546.71581.67568.41660.34631.94548.041. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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 3.4Video Input: Bosphorus 1080pEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.38, N = 6SE +/- 0.54, N = 6SE +/- 0.53, N = 6SE +/- 0.15, N = 7SE +/- 0.18, N = 7SE +/- 0.21, N = 7SE +/- 0.21, N = 7SE +/- 0.77, N = 1580.5886.7889.12103.2599.22118.06109.5784.331. (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.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: T5 Encoder - Device: CPU - Executor: StandardEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 1.77, N = 4SE +/- 1.14, N = 3SE +/- 0.71, N = 3SE +/- 2.03, N = 4SE +/- 2.12, N = 15SE +/- 2.20, N = 3SE +/- 1.28, N = 3SE +/- 2.25, N = 15152.18143.50139.48168.62204.34162.33199.25169.621. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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: 16 - Buffer Length: 256 - Filter Length: 32EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488200M400M600M800M1000MSE +/- 383507.21, N = 3SE +/- 1997031.13, N = 3SE +/- 812656.82, N = 3SE +/- 956399.50, N = 3SE +/- 3918917.08, N = 3SE +/- 67412.49, N = 3SE +/- 1995247.13, N = 3SE +/- 5153390.47, N = 35461366677180200007355666677284600007183066677959666677726166677110500001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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 - Mesh TimeEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248860120180240300297.17262.84296.21254.77243.10226.01204.60225.951. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

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 TimesEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488130260390520650SE +/- 2.17, N = 4SE +/- 2.21, N = 4SE +/- 2.53, N = 4SE +/- 5.33, N = 4SE +/- 6.12, N = 4SE +/- 6.41, N = 4SE +/- 6.56, N = 15SE +/- 2.82, N = 5550532527526602533571415

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 TimeEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248860012001800240030002944.552781.242827.882407.902123.992351.512040.352260.991. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

MariaDB

This is a MariaDB MySQL database server benchmark making use of sysbench rather than the existing pts/mysqlslap test profile that uses MariaDB with mysqlslap/mariadb-slap as the benchmark driver. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 11.5Test: oltp_update_non_index - Threads: 64EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850K100K150K200K250KSE +/- 223.22, N = 3SE +/- 120.42, N = 3SE +/- 295.88, N = 3SE +/- 159.15, N = 3SE +/- 1725.95, N = 9SE +/- 290.99, N = 3SE +/- 355.20, N = 3SE +/- 516.32, N = 3162697197477201401191756185239233313227813220068-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-lnuma -lpcre2-8 -lcrypt -laio -lssl -lcrypto1. (CXX) g++ options: -fPIC -pie -fstack-protector -O3 -lpthread -lz -ldl -lm

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-50EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881020304050SE +/- 0.17, N = 3SE +/- 0.14, N = 3SE +/- 0.16, N = 3SE +/- 0.17, N = 3SE +/- 0.01, N = 3SE +/- 0.94, N = 15SE +/- 0.34, N = 3SE +/- 1.58, N = 1534.8843.8945.1139.5341.3443.3846.2032.25MIN: 21.67 / MAX: 35.54MIN: 42.37 / MAX: 45.05MIN: 43.61 / MAX: 45.86MIN: 38.01 / MAX: 40.46MIN: 30.45 / MAX: 41.8MIN: 34.95 / MAX: 46.71MIN: 43.63 / MAX: 47.16MIN: 23.14 / MAX: 40.93

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 CompileEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881020304050SE +/- 0.12, N = 3SE +/- 0.14, N = 3SE +/- 0.23, N = 3SE +/- 0.36, N = 3SE +/- 0.14, N = 3SE +/- 0.13, N = 3SE +/- 0.25, N = 3SE +/- 0.23, N = 342.1136.9036.9032.5630.4729.4429.5038.131. (CC) gcc options: -m64 -ldl -lgcc_s -lutil -lrt -lpthread -lm -lc -pie -nodefaultlibs

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.090.180.270.360.45SE +/- 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 = 30.320.290.280.320.310.290.300.40MIN: 0.16 / MAX: 7.47MIN: 0.16 / MAX: 5.09MIN: 0.15 / MAX: 7.71MIN: 0.16 / MAX: 9.37MIN: 0.16 / MAX: 7.6MIN: 0.16 / MAX: 8.63MIN: 0.16 / MAX: 8.01MIN: 0.18 / MAX: 8.351. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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 BeamEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.20, N = 3SE +/- 0.23, N = 3SE +/- 3.91, N = 12SE +/- 2.01, N = 15SE +/- 1.43, N = 5SE +/- 1.49, N = 15SE +/- 1.50, N = 3SE +/- 2.49, N = 12140.76124.70145.32118.52138.93108.64117.99153.90

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 FacadeEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248812002400360048006000SE +/- 16.57, N = 6SE +/- 5.95, N = 7SE +/- 10.00, N = 7SE +/- 26.89, N = 7SE +/- 4.33, N = 6SE +/- 29.27, N = 6SE +/- 90.20, N = 15SE +/- 17.70, N = 6516049885083503751325224397856071. (CXX) g++ options: -O3

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: IMDBEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.29, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.14, N = 3115.64101.97100.6390.0387.8482.3883.7299.581. (CXX) g++ options: -O3 -rdynamic -lssl -lcrypto -ldl

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: H2 Database EngineEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24885001000150020002500SE +/- 14.53, N = 3SE +/- 11.72, N = 3SE +/- 7.51, N = 3SE +/- 23.54, N = 4SE +/- 9.02, N = 3SE +/- 13.96, N = 3SE +/- 5.29, N = 3SE +/- 23.84, N = 323941992197719141850182617152128

MariaDB

This is a MariaDB MySQL database server benchmark making use of sysbench rather than the existing pts/mysqlslap test profile that uses MariaDB with mysqlslap/mariadb-slap as the benchmark driver. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 11.5Test: oltp_update_index - Threads: 64EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248840K80K120K160K200KSE +/- 254.61, N = 3SE +/- 133.22, N = 3SE +/- 308.54, N = 3SE +/- 411.40, N = 3SE +/- 173.91, N = 3SE +/- 282.52, N = 3SE +/- 412.73, N = 3SE +/- 250.58, N = 3142647173080177579168313159977198538196283187089-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-lnuma -lpcre2-8 -lcrypt -laio -lssl -lcrypto1. (CXX) g++ options: -fPIC -pie -fstack-protector -O3 -lpthread -lz -ldl -lm

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, Second RunEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248860120180240300SE +/- 1.04, N = 3SE +/- 0.75, N = 3SE +/- 1.00, N = 3SE +/- 2.91, N = 3SE +/- 1.95, N = 3SE +/- 1.51, N = 5SE +/- 3.27, N = 3SE +/- 3.87, N = 3211.23230.37236.47256.89279.62248.89293.87249.50MIN: 7.98 / MAX: 10000MIN: 10.32 / MAX: 10000MIN: 10.66 / MAX: 10000MIN: 13.95 / MAX: 10000MIN: 14.87 / MAX: 8571.43MIN: 15.92 / MAX: 8571.43MIN: 18.95 / MAX: 10000MIN: 9.27 / MAX: 12000

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 SpaceEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248870140210280350SE +/- 0.58, N = 3SE +/- 1.20, N = 3SE +/- 0.33, N = 3SE +/- 1.53, N = 3SE +/- 0.88, N = 3SE +/- 0.58, N = 3SE +/- 1.73, N = 3SE +/- 0.67, N = 32272492552582362782683151. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lbz2 -lz -lm -lpthread -lgomp

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488246810SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 35.755.875.636.496.175.966.217.78MIN: 3.02 / MAX: 14.8MIN: 3.03 / MAX: 15.22MIN: 2.98 / MAX: 16.09MIN: 3.31 / MAX: 18.52MIN: 3.21 / MAX: 15.44MIN: 3.08 / MAX: 18.52MIN: 3.24 / MAX: 18.16MIN: 3.64 / MAX: 44.071. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Xcompact3d Incompact3d

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

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per DirectionEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100SE +/- 0.07, N = 3SE +/- 0.30, N = 3SE +/- 0.27, N = 3SE +/- 0.40, N = 3SE +/- 0.60, N = 15SE +/- 0.58, N = 10SE +/- 0.83, N = 3SE +/- 0.87, N = 1589.8785.3487.7775.6870.7173.8865.1173.721. (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

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: CPUEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248890180270360450SE +/- 0.03, N = 3SE +/- 0.10, N = 3SE +/- 0.15, N = 3SE +/- 0.33, N = 3SE +/- 0.11, N = 3SE +/- 0.24, N = 3SE +/- 0.41, N = 3SE +/- 0.59, N = 3393.13299.02285.97331.37313.09301.79312.52384.09MIN: 378.57 / MAX: 400.92MIN: 288.59 / MAX: 303.77MIN: 150.08 / MAX: 303.22MIN: 179.64 / MAX: 354.6MIN: 291.27 / MAX: 322.77MIN: 278.87 / MAX: 327.8MIN: 167.86 / MAX: 330.96MIN: 285.45 / MAX: 543.491. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

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 1MEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882004006008001000SE +/- 2.88, N = 3SE +/- 1.89, N = 3SE +/- 10.68, N = 9SE +/- 9.60, N = 3SE +/- 2.98, N = 3SE +/- 4.49, N = 3SE +/- 8.46, N = 9SE +/- 17.37, N = 71126.52967.201003.59936.34978.51891.84823.26983.54

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: 100EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.11, N = 3SE +/- 0.64, N = 3SE +/- 0.48, N = 3SE +/- 0.12, N = 3SE +/- 0.27, N = 3SE +/- 0.21, N = 3SE +/- 0.15, N = 3SE +/- 0.51, N = 3136.58122.33122.55108.81105.03103.37101.02137.92MAX: 23944.74MAX: 23948.29MAX: 23938.16MAX: 23915.13MAX: 23941.39MAX: 23852.59MAX: 23859.84MAX: 23917.2

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 RunEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248860120180240300SE +/- 1.41, N = 3SE +/- 2.33, N = 3SE +/- 4.93, N = 3SE +/- 4.31, N = 3SE +/- 2.40, N = 3SE +/- 2.29, N = 5SE +/- 2.43, N = 3SE +/- 1.74, N = 3214.00233.30232.47259.35291.70251.80291.26249.90MIN: 7.95 / MAX: 10000MIN: 10.46 / MAX: 10000MIN: 10.67 / MAX: 12000MIN: 13.75 / MAX: 10000MIN: 14.85 / MAX: 10000MIN: 15.75 / MAX: 10000MIN: 16.56 / MAX: 7500MIN: 9.25 / MAX: 12000

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.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: T5 Encoder - Device: CPU - Executor: ParallelEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 0.12, N = 3SE +/- 0.50, N = 3SE +/- 1.35, N = 3SE +/- 0.51, N = 3SE +/- 0.37, N = 3SE +/- 0.69, N = 3SE +/- 0.57, N = 3SE +/- 0.97, N = 3147.28138.37141.63155.88187.78156.77183.82161.311. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 100EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248816M32M48M64M80MSE +/- 45916.59, N = 3SE +/- 302032.90, N = 3SE +/- 268932.10, N = 3SE +/- 59581.06, N = 3SE +/- 81989.58, N = 3SE +/- 205604.00, N = 3SE +/- 138400.03, N = 3SE +/- 209117.79, N = 35668905263086247629514587054497973321638740752447603185056035278

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, First Run / Cold CacheEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248860120180240300SE +/- 1.29, N = 3SE +/- 1.31, N = 3SE +/- 1.84, N = 3SE +/- 0.94, N = 3SE +/- 1.12, N = 3SE +/- 2.37, N = 5SE +/- 1.16, N = 3SE +/- 1.98, N = 3195.64213.47222.47233.74258.37227.39265.18234.39MIN: 7.8 / MAX: 10000MIN: 10.21 / MAX: 8571.43MIN: 10.49 / MAX: 12000MIN: 13.67 / MAX: 10000MIN: 14.45 / MAX: 10000MIN: 13.31 / MAX: 10000MIN: 14.2 / MAX: 10000MIN: 9.14 / MAX: 12000

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 1080p - Video Preset: FasterEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881122334455SE +/- 0.06, N = 4SE +/- 0.12, N = 4SE +/- 0.04, N = 4SE +/- 0.10, N = 4SE +/- 0.07, N = 4SE +/- 0.11, N = 4SE +/- 0.12, N = 4SE +/- 0.25, N = 437.6343.8345.1446.1645.9850.9148.8640.071. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects

Z3 Theorem Prover

The Z3 Theorem Prover / SMT solver is developed by Microsoft Research under the MIT license. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterZ3 Theorem Prover 4.12.1SMT File: 2.smt2EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881428425670SE +/- 0.33, N = 3SE +/- 0.40, N = 3SE +/- 0.27, N = 3SE +/- 0.39, N = 3SE +/- 0.51, N = 3SE +/- 0.53, N = 3SE +/- 0.02, N = 3SE +/- 0.51, N = 349.3247.7348.6850.4647.2348.8847.5763.081. (CXX) g++ options: -lpthread -std=c++17 -fvisibility=hidden -mfpmath=sse -msse -msse2 -O3 -fPIC

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: DefaultEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248851015202518.8516.8216.5815.0615.2114.1314.5916.85

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.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248816M32M48M64M80MSE +/- 580368.90, N = 5SE +/- 178945.47, N = 3SE +/- 441409.11, N = 3SE +/- 367807.54, N = 3SE +/- 171497.11, N = 3SE +/- 597457.28, N = 3SE +/- 219419.50, N = 3SE +/- 207001.08, N = 35612224363954176641825637022686872710592737160257457989656494499

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.2Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488110220330440550SE +/- 8.54, N = 5SE +/- 3.04, N = 3SE +/- 4.56, N = 3SE +/- 4.87, N = 3SE +/- 1.07, N = 3SE +/- 4.31, N = 3SE +/- 1.42, N = 3SE +/- 1.92, N = 3529.44465.51462.14421.83407.64399.56398.76524.13MAX: 28194.55MAX: 27554.23MAX: 27920.53MAX: 27011.65MAX: 26798.87MAX: 27980.82MAX: 26832.58MAX: 26588.63

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 ThreadEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882004006008001000SE +/- 6.27, N = 10SE +/- 7.98, N = 10SE +/- 3.24, N = 10SE +/- 7.68, N = 9SE +/- 1.53, N = 9SE +/- 5.92, N = 10SE +/- 3.58, N = 9SE +/- 5.58, N = 91073.61071.01061.51034.9908.81049.1907.5809.6-mavx512f -mavx512cd -mavx512bw -mavx512dq-mavx512f -mavx512cd -mavx512bw -mavx512dq-mavx512f -mavx512cd -mavx512bw -mavx512dq-mavx512f -mavx512cd -mavx512bw -mavx512dq-mavx512f -mavx512cd -mavx512bw -mavx512dq-mavx512f -mavx512cd -mavx512bw -mavx512dq-mavx512f -mavx512cd -mavx512bw -mavx512dq1. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -fno-trapping-math -fno-math-errno -ldl

OpenFOAM

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Mesh TimeEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248891827364539.1633.3338.1732.8533.6530.7729.6935.191. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

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.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: GPT-2 - Device: CPU - Executor: StandardEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 0.08, N = 3SE +/- 1.56, N = 4SE +/- 0.21, N = 3SE +/- 1.56, N = 4SE +/- 3.26, N = 12SE +/- 1.39, N = 15SE +/- 0.20, N = 3SE +/- 0.39, N = 3134.45134.74123.95140.98161.84133.17140.35126.671. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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: 100EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.72, N = 3SE +/- 0.39, N = 3SE +/- 0.24, N = 3SE +/- 0.32, N = 3SE +/- 0.40, N = 3SE +/- 0.82, N = 3SE +/- 0.56, N = 3SE +/- 0.54, N = 3117.14105.70105.5895.0992.9192.0892.27119.49MAX: 15130.91MAX: 13869.85MAX: 15103.61MAX: 12619.54MAX: 12607.68MAX: 11356.99MAX: 11332.45MAX: 13845.12

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.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: ParallelEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881224364860SE +/- 0.15, N = 3SE +/- 0.32, N = 3SE +/- 0.43, N = 3SE +/- 0.46, N = 3SE +/- 0.08, N = 3SE +/- 0.21, N = 3SE +/- 0.57, N = 4SE +/- 0.30, N = 346.5649.8150.5349.5548.1452.7253.1541.011. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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: 500 - Client Number: 100EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100SE +/- 0.12, N = 3SE +/- 0.05, N = 3SE +/- 0.29, N = 3SE +/- 0.07, N = 3SE +/- 0.13, N = 3SE +/- 0.29, N = 3SE +/- 0.05, N = 3SE +/- 0.20, N = 376.6968.5868.1662.3060.5260.6659.6177.24MAX: 23999.1MAX: 23897.06MAX: 23889.9MAX: 23874.13MAX: 23884.06MAX: 23861.25MAX: 23860.38MAX: 23891.04

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248890180270360450SE +/- 2.59, N = 3SE +/- 6.58, N = 3SE +/- 6.12, N = 3SE +/- 1.97, N = 3SE +/- 5.21, N = 3SE +/- 5.40, N = 3SE +/- 1.19, N = 3SE +/- 10.20, N = 3406.89350.00347.97333.41326.49321.29316.26382.48MAX: 27507.51MAX: 27235.37MAX: 27085.8MAX: 26859.8MAX: 27383.3MAX: 27126.33MAX: 27135.69MAX: 27376.22

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820M40M60M80M100MSE +/- 173074.65, N = 3SE +/- 612351.99, N = 3SE +/- 255843.12, N = 3SE +/- 191894.24, N = 3SE +/- 707677.76, N = 3SE +/- 818556.17, N = 3SE +/- 243124.87, N = 3SE +/- 922667.78, N = 36251824571203893713387407623218578204086784543168018926364886364

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820M40M60M80M100MSE +/- 57777.29, N = 3SE +/- 77775.76, N = 3SE +/- 296062.71, N = 3SE +/- 132686.57, N = 3SE +/- 198325.80, N = 3SE +/- 390613.00, N = 3SE +/- 118053.95, N = 3SE +/- 232749.13, N = 36234888169342377695614497567015177878702776056277915944161842388

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 100EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820M40M60M80M100MSE +/- 200521.84, N = 3SE +/- 216731.24, N = 3SE +/- 282897.65, N = 3SE +/- 105860.93, N = 3SE +/- 51778.51, N = 3SE +/- 368630.91, N = 3SE +/- 356287.02, N = 3SE +/- 262085.32, N = 36354137870293571703052857743401179324034793153197999961162558098

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: 32EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248814M28M42M56M70MSE +/- 2905.93, N = 3SE +/- 1855.92, N = 3SE +/- 2645.75, N = 3SE +/- 26514.15, N = 3SE +/- 16522.71, N = 3SE +/- 318365.79, N = 3SE +/- 223381.09, N = 3SE +/- 637003.49, N = 350272333522516675365900053147000510310005301866749584667632366671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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-50EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248848121620SE +/- 0.02, N = 5SE +/- 0.03, N = 5SE +/- 0.01, N = 5SE +/- 0.02, N = 5SE +/- 0.01, N = 5SE +/- 0.03, N = 5SE +/- 0.01, N = 5SE +/- 0.01, N = 515.6117.4517.4714.7614.7614.9315.1213.71

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: 500 - Client Number: 400EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248860120180240300SE +/- 1.17, N = 3SE +/- 6.43, N = 3SE +/- 1.82, N = 3SE +/- 3.20, N = 3SE +/- 0.75, N = 3SE +/- 2.35, N = 3SE +/- 1.32, N = 3SE +/- 4.53, N = 3295.34260.60268.42244.44237.32235.16232.67290.14MAX: 26742.14MAX: 26603.07MAX: 27440.82MAX: 26631.84MAX: 26320.16MAX: 26882.16MAX: 26680.29MAX: 26535.45

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820M40M60M80M100MSE +/- 77022.60, N = 3SE +/- 676495.92, N = 3SE +/- 261270.67, N = 3SE +/- 718192.76, N = 3SE +/- 169341.63, N = 3SE +/- 360981.97, N = 3SE +/- 285870.90, N = 3SE +/- 787042.09, N = 36245919570107030691441647472357977172160776589147860081962655549

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100SE +/- 0.44, N = 3SE +/- 0.91, N = 3SE +/- 0.31, N = 3SE +/- 0.39, N = 3SE +/- 0.22, N = 3SE +/- 0.10, N = 3SE +/- 0.42, N = 3SE +/- 0.51, N = 379.6471.6670.7066.4964.5565.6364.5280.56MAX: 15110.81MAX: 13850.52MAX: 13822.83MAX: 13829.28MAX: 15090.58MAX: 12576.01MAX: 12582.24MAX: 15084.51

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248815M30M45M60M75MSE +/- 151013.56, N = 3SE +/- 364826.83, N = 3SE +/- 42980.57, N = 3SE +/- 282335.19, N = 3SE +/- 128616.25, N = 3SE +/- 41672.18, N = 3SE +/- 248696.67, N = 3SE +/- 205982.51, N = 35792524564201036647812856873598570752573698575197089646957028398

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248815M30M45M60M75MSE +/- 374209.00, N = 3SE +/- 402227.19, N = 3SE +/- 418677.10, N = 3SE +/- 162538.27, N = 3SE +/- 347434.46, N = 3SE +/- 489754.87, N = 3SE +/- 262014.17, N = 3SE +/- 245819.57, N = 35719064863451680646555026877181469435597697311667081737357760197

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: PMD Source Code AnalyzerEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830060090012001500SE +/- 9.62, N = 11SE +/- 8.67, N = 11SE +/- 9.55, N = 7SE +/- 7.50, N = 7SE +/- 9.69, N = 7SE +/- 5.68, N = 7SE +/- 3.44, N = 7SE +/- 8.96, N = 1512981155111311091101105310511122

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: GoogLeNetEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881428425670SE +/- 0.15, N = 8SE +/- 0.09, N = 8SE +/- 0.10, N = 9SE +/- 0.08, N = 8SE +/- 0.18, N = 8SE +/- 0.18, N = 8SE +/- 0.18, N = 8SE +/- 0.05, N = 852.3661.8862.9954.4955.8957.3460.1451.33

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Savina Reactors.IOEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24888001600240032004000SE +/- 12.81, N = 3SE +/- 13.60, N = 3SE +/- 28.17, N = 3SE +/- 30.89, N = 3SE +/- 38.91, N = 4SE +/- 15.03, N = 3SE +/- 41.68, N = 3SE +/- 16.31, N = 33288.43060.53042.03338.23515.93265.23350.93695.4MIN: 3264.51 / MAX: 4523.94MIN: 3036.07 / MAX: 4235.11MIN: 3011.87 / MAX: 4321.7MIN: 3297.53 / MAX: 4743.03MIN: 3407.81 / MAX: 4967.72MIN: 3236.75 / MAX: 4741.82MIN: 3285.69 / MAX: 4786.32MIN: 3664.8 / MAX: 5363.87

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: 500 - Client Number: 400EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248860120180240300SE +/- 3.65, N = 3SE +/- 3.35, N = 3SE +/- 2.74, N = 3SE +/- 1.22, N = 3SE +/- 3.51, N = 3SE +/- 3.67, N = 3SE +/- 1.59, N = 3SE +/- 1.91, N = 3275.62250.94243.61231.55231.39230.41227.34275.14MAX: 28062.74MAX: 27118.82MAX: 26975.13MAX: 27011.78MAX: 27094.18MAX: 26735.87MAX: 26429.57MAX: 27185.57

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-50EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881530456075SE +/- 0.26, N = 3SE +/- 0.65, N = 3SE +/- 1.84, N = 15SE +/- 0.11, N = 3SE +/- 0.30, N = 3SE +/- 0.90, N = 15SE +/- 0.68, N = 6SE +/- 1.37, N = 1258.4968.6767.4856.9661.3465.6968.8061.49MIN: 55.22 / MAX: 59.43MIN: 64.12 / MAX: 71.37MIN: 46.76 / MAX: 74.15MIN: 52.49 / MAX: 58MIN: 50.53 / MAX: 62.03MIN: 53.12 / MAX: 69.76MIN: 59.18 / MAX: 72.68MIN: 40.06 / MAX: 72.22

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 BenchmarkEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882004006008001000SE +/- 2.19, N = 3SE +/- 1.60, N = 3SE +/- 3.59, N = 3SE +/- 0.83, N = 3SE +/- 0.68, N = 3SE +/- 2.05, N = 3SE +/- 3.03, N = 3SE +/- 1.43, N = 3773.69804.51808.62812.29791.79803.04746.24675.48

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 ThreadEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 3.58, N = 15SE +/- 0.00, N = 4SE +/- 0.00, N = 4SE +/- 0.00, N = 4SE +/- 0.03, N = 4SE +/- 0.00, N = 4SE +/- 0.00, N = 4SE +/- 7.12, N = 15179.8183.4183.6183.6183.4183.5183.6153.6-mavx512f -mavx512cd -mavx512bw -mavx512dq - MIN: 105.9 / MAX: 183.4-mavx512f -mavx512cd -mavx512bw -mavx512dq - MIN: 105.9-mavx512f -mavx512cd -mavx512bw -mavx512dq - MIN: 106-mavx512f -mavx512cd -mavx512bw -mavx512dq - MIN: 106-mavx512f -mavx512cd -mavx512bw -mavx512dq - MIN: 105.8-mavx512f -mavx512cd -mavx512bw -mavx512dq - MIN: 105.9-mavx512f -mavx512cd -mavx512bw -mavx512dq - MIN: 106MIN: 105.1 / MAX: 182.11. (CXX) g++ options: -O3 -march=native -mavx2 -mavx -msse4.1 -mfma -fno-trapping-math -fno-math-errno -ldl

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 6.1Encoder: libx265 - Scenario: Video On DemandEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881632486480SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.09, N = 3SE +/- 0.04, N = 3SE +/- 0.09, N = 3SE +/- 0.13, N = 3SE +/- 0.10, N = 3SE +/- 0.02, N = 360.2064.9265.3565.8763.7069.8165.2458.751. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.1Encoder: libx265 - Scenario: PlatformEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881632486480SE +/- 0.01, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.10, N = 3SE +/- 0.19, N = 3SE +/- 0.22, N = 360.2364.7765.2665.8263.6869.8065.4958.751. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.1Encoder: libx265 - Scenario: UploadEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488816243240SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.08, N = 329.1731.7331.9532.4031.3234.2132.1428.891. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

MariaDB mariadb-slap

This is a MariaDB MySQL database server benchmark making use of mysqlslap (mariadb-slap). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB mariadb-slap 11.5Clients: 128EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488130260390520650SE +/- 0.54, N = 3SE +/- 0.66, N = 3SE +/- 0.74, N = 3SE +/- 0.23, N = 3SE +/- 0.93, N = 3SE +/- 0.54, N = 3SE +/- 0.71, N = 3SE +/- 1.05, N = 3517565580586561612582574-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-lnuma -lpcre2-8 -lcrypt -laio -lssl -lcrypto1. (CXX) g++ options: -fPIC -pie -fstack-protector -O3 -lpthread -lz -ldl -lm

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 8.4.0EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488246810SE +/- 0.013, N = 7SE +/- 0.016, N = 7SE +/- 0.017, N = 7SE +/- 0.018, N = 7SE +/- 0.016, N = 7SE +/- 0.021, N = 7SE +/- 0.020, N = 7SE +/- 0.011, N = 65.6945.5835.5875.8155.5265.7805.4806.462

MariaDB mariadb-slap

This is a MariaDB MySQL database server benchmark making use of mysqlslap (mariadb-slap). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB mariadb-slap 11.5Clients: 64EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248830060090012001500SE +/- 3.77, N = 3SE +/- 3.26, N = 3SE +/- 5.66, N = 3SE +/- 7.01, N = 3SE +/- 5.80, N = 3SE +/- 4.85, N = 3SE +/- 4.14, N = 3SE +/- 4.64, N = 310091095112411311076118311241128-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-shared -lrt -lstdc++-lnuma -lpcre2-8 -lcrypt -laio -lssl -lcrypto1. (CXX) g++ options: -fPIC -pie -fstack-protector -O3 -lpthread -lz -ldl -lm

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.orgInferences Per Second, More Is BetterONNX Runtime 1.17Model: GPT-2 - Device: CPU - Executor: ParallelEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 0.05, N = 3SE +/- 0.23, N = 3SE +/- 0.41, N = 3SE +/- 0.39, N = 3SE +/- 0.48, N = 3SE +/- 0.12, N = 3SE +/- 0.89, N = 3SE +/- 0.46, N = 3118.96121.29120.28121.99137.92122.63139.02123.281. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

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 6.1Encoder: libx265 - Scenario: LiveEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 0.31, N = 3SE +/- 0.58, N = 3SE +/- 0.17, N = 3SE +/- 0.29, N = 3SE +/- 1.34, N = 3SE +/- 0.71, N = 3SE +/- 0.59, N = 3SE +/- 1.39, N = 3162.38176.63179.99177.93171.60186.42176.69178.151. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

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: 512EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884M8M12M16M20MSE +/- 150617.99, N = 3SE +/- 189549.47, N = 3SE +/- 250051.55, N = 3SE +/- 421403.42, N = 15SE +/- 44508.43, N = 3SE +/- 410224.76, N = 15SE +/- 267438.19, N = 15SE +/- 13650.40, N = 318426333193650001961166718078200191250001896440019119067202550001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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 OptimizedEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850100150200250214.38202.26200.98197.98199.92193.47198.06197.92

System Power Consumption Monitor

OpenBenchmarking.orgWattsSystem Power Consumption MonitorPhoronix Test Suite System MonitoringEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248860120180240300Min: 51.6 / Avg: 115.02 / Max: 143.1Min: 51.9 / Avg: 120.42 / Max: 144Min: 67.4 / Avg: 159.31 / Max: 212.3Min: 84.9 / Avg: 130.63 / Max: 143.7Min: 53.5 / Avg: 128.7 / Max: 186.1Min: 90.7 / Avg: 202.74 / Max: 301Min: 76.2 / Avg: 149.2 / Max: 222.4Min: 51.7 / Avg: 138.43 / Max: 338.5

CPU Power Consumption Monitor

OpenBenchmarking.orgWattsCPU Power Consumption MonitorPhoronix Test Suite System MonitoringEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200Min: 12.24 / Avg: 71.17 / Max: 91.98Min: 13.54 / Avg: 75.17 / Max: 92.35Min: 29.45 / Avg: 105.15 / Max: 145.99Min: 46.51 / Avg: 84.39 / Max: 91.68Min: 16.44 / Avg: 80.5 / Max: 123.57Min: 48.93 / Avg: 138.83 / Max: 208.13Min: 37.24 / Avg: 95.52 / Max: 147.11Min: 18.57 / Avg: 89.22 / Max: 198.5

CPU Peak Freq (Highest CPU Core Frequency) Monitor

OpenBenchmarking.orgMegahertzCPU Peak Freq (Highest CPU Core Frequency) MonitorPhoronix Test Suite System MonitoringEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248810002000300040005000Min: 400 / Avg: 4805.55 / Max: 5230Min: 400 / Avg: 4794.94 / Max: 5417Min: 4543 / Avg: 5178.15 / Max: 5518Min: 3947 / Avg: 4915.82 / Max: 5476Min: 400 / Avg: 4831.8 / Max: 5638Min: 4577 / Avg: 5148.01 / Max: 5604Min: 4296 / Avg: 4927.95 / Max: 5702Min: 801 / Avg: 4466.33 / Max: 5638

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: StandardEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881.00392.00783.01174.01565.0195SE +/- 0.05218, N = 15SE +/- 0.03376, N = 15SE +/- 0.00310, N = 3SE +/- 0.00434, N = 3SE +/- 0.05991, N = 15SE +/- 0.05300, N = 15SE +/- 0.06313, N = 15SE +/- 0.03583, N = 33.437912.918463.091502.675542.663412.382992.413414.461701. (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: StandardEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248890180270360450SE +/- 4.18, N = 15SE +/- 3.81, N = 15SE +/- 0.32, N = 3SE +/- 0.61, N = 3SE +/- 9.13, N = 15SE +/- 9.29, N = 15SE +/- 11.38, N = 15SE +/- 1.81, N = 3291.71343.18323.38373.67378.25422.42418.47224.111. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt

SVT-AV1

OpenBenchmarking.orgWatts, Fewer Is BetterSVT-AV1 2.0System Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850100150200250Min: 53.4 / Avg: 96.49 / Max: 135.9Min: 53.5 / Avg: 97.56 / Max: 137.6Min: 69.1 / Avg: 135.43 / Max: 201.7Min: 92.3 / Avg: 113.17 / Max: 139.5Min: 72.5 / Avg: 106.27 / Max: 142.7Min: 96.5 / Avg: 160.58 / Max: 257.9Min: 87.4 / Avg: 116.78 / Max: 160.7Min: 53.1 / Avg: 139.35 / Max: 291.4

OpenBenchmarking.orgWatts, Fewer Is BetterSVT-AV1 2.0CPU Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150Min: 14.86 / Avg: 57.04 / Max: 90.38Min: 17.72 / Avg: 52.81 / Max: 86.02Min: 34.97 / Avg: 84.95 / Max: 119.79Min: 59.15 / Avg: 72.4 / Max: 84.51Min: 38.02 / Avg: 61.03 / Max: 80.68Min: 61.36 / Avg: 107.18 / Max: 143.06Min: 49.32 / Avg: 71.24 / Max: 87.22Min: 19.94 / Avg: 83.67 / Max: 140.47

OpenBenchmarking.orgMegahertz, More Is BetterSVT-AV1 2.0CPU Peak Freq (Highest CPU Core Frequency) MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248810002000300040005000Min: 400 / Avg: 4090.6 / Max: 5140Min: 400 / Avg: 4136 / Max: 5340Min: 5050 / Avg: 5231 / Max: 5488Min: 4265 / Avg: 5014.54 / Max: 5440Min: 4656 / Avg: 5022.46 / Max: 5265Min: 5067 / Avg: 5341.77 / Max: 5536Min: 4665 / Avg: 5036.36 / Max: 5403Min: 3900 / Avg: 4628.73 / Max: 5328

OpenBenchmarking.orgFrames Per Second Per Watt, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 13 - Input: Bosphorus 4KEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.51661.03321.54982.06642.5831.7112.2641.4471.9582.2961.5422.2741.250

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 13 - Input: Bosphorus 4KEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 0.98, N = 15SE +/- 1.88, N = 15SE +/- 2.01, N = 15SE +/- 2.97, N = 15SE +/- 2.98, N = 15SE +/- 4.27, N = 15SE +/- 4.16, N = 15SE +/- 0.99, N = 1597.60119.57122.91141.77140.09165.29162.00104.611. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenRadioss

OpenBenchmarking.orgWatts, Fewer Is BetterOpenRadioss 2023.09.15System Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200Min: 53.6 / Avg: 129.23 / Max: 134.8Min: 54.8 / Avg: 132.78 / Max: 139.8Min: 75.4 / Avg: 167.23 / Max: 182.1Min: 98.8 / Avg: 136.53 / Max: 141.3Min: 74 / Avg: 139.83 / Max: 156.7Min: 101.4 / Avg: 218.96 / Max: 242.9Min: 83.3 / Avg: 163.91 / Max: 181.2Min: 53.3 / Avg: 144.2 / Max: 243.3

OpenBenchmarking.orgWatts, Fewer Is BetterOpenRadioss 2023.09.15CPU Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150Min: 16.53 / Avg: 81.61 / Max: 87.66Min: 17.82 / Avg: 83.68 / Max: 90.4Min: 37.57 / Avg: 110.73 / Max: 122.22Min: 60.63 / Avg: 87.4 / Max: 90.41Min: 37.66 / Avg: 89 / Max: 101.5Min: 60.58 / Avg: 150.83 / Max: 170.13Min: 46.07 / Avg: 106.05 / Max: 119.17Min: 21.15 / Avg: 92.95 / Max: 162.3

OpenBenchmarking.orgMegahertz, More Is BetterOpenRadioss 2023.09.15CPU Peak Freq (Highest CPU Core Frequency) MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248810002000300040005000Min: 400 / Avg: 4945.43 / Max: 5137Min: 400 / Avg: 4730.46 / Max: 5340Min: 4849 / Avg: 5158.27 / Max: 5515Min: 4651 / Avg: 4861.54 / Max: 5304Min: 4692 / Avg: 4820.08 / Max: 5215Min: 4862 / Avg: 5088.49 / Max: 5537Min: 4692 / Avg: 4785.46 / Max: 5203Min: 3993 / Avg: 4423.44 / Max: 5416

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: Cell Phone Drop TestEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100SE +/- 0.21, N = 3SE +/- 0.37, N = 3SE +/- 3.06, N = 15SE +/- 1.62, N = 15SE +/- 1.45, N = 15SE +/- 1.68, N = 12SE +/- 0.42, N = 3SE +/- 3.30, N = 1598.8580.0495.4875.8382.0267.7167.8893.26

OpenBenchmarking.orgWatts, Fewer Is BetterOpenRadioss 2023.09.15System Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850100150200250Min: 54.7 / Avg: 132.03 / Max: 136Min: 55.4 / Avg: 132.86 / Max: 137.3Min: 75.2 / Avg: 176.99 / Max: 191.4Min: 99.5 / Avg: 135.16 / Max: 138.6Min: 74.2 / Avg: 146.13 / Max: 157Min: 101.3 / Avg: 241.39 / Max: 265.9Min: 83.7 / Avg: 174.36 / Max: 196.5Min: 54.2 / Avg: 143.97 / Max: 246.8

OpenBenchmarking.orgWatts, Fewer Is BetterOpenRadioss 2023.09.15CPU Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150Min: 16.47 / Avg: 87.18 / Max: 90.37Min: 18.71 / Avg: 86.78 / Max: 90.38Min: 37.54 / Avg: 119.54 / Max: 131.85Min: 58.52 / Avg: 88.53 / Max: 90.38Min: 38.31 / Avg: 95.37 / Max: 103.37Min: 60.42 / Avg: 171.03 / Max: 189.55Min: 38.85 / Avg: 115.17 / Max: 132.31Min: 21.95 / Avg: 94.12 / Max: 160.28

OpenBenchmarking.orgMegahertz, More Is BetterOpenRadioss 2023.09.15CPU Peak Freq (Highest CPU Core Frequency) MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248810002000300040005000Min: 400 / Avg: 4962.57 / Max: 5186Min: 4875 / Avg: 4971.99 / Max: 5340Min: 5027 / Avg: 5110.05 / Max: 5489Min: 4540 / Avg: 4751.02 / Max: 5429Min: 4722 / Avg: 4793.95 / Max: 5190Min: 4952 / Avg: 5013.45 / Max: 5489Min: 4704 / Avg: 4794.67 / Max: 5202Min: 4246 / Avg: 4387.56 / Max: 5400

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: Rubber O-Ring Seal InstallationEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24884080120160200SE +/- 0.54, N = 3SE +/- 0.24, N = 3SE +/- 8.68, N = 9SE +/- 7.92, N = 9SE +/- 3.80, N = 9SE +/- 1.00, N = 3SE +/- 3.75, N = 15SE +/- 5.45, N = 12157.48127.32162.03126.03148.28110.05106.01177.74

HeFFTe - Highly Efficient FFT for Exascale

OpenBenchmarking.orgWatts, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.4System Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100Min: 52.2 / Avg: 54.38 / Max: 61.2Min: 51.9 / Avg: 54.38 / Max: 65Min: 71.1 / Avg: 74.77 / Max: 84.5Min: 97.7 / Avg: 99.33 / Max: 104.6Min: 72.4 / Avg: 74.99 / Max: 83.7Min: 95.8 / Avg: 99.24 / Max: 114.7Min: 86.2 / Avg: 87.89 / Max: 89.1Min: 52.8 / Avg: 61.36 / Max: 82.7

OpenBenchmarking.orgWatts, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.4CPU Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881326395265Min: 17.18 / Avg: 18.45 / Max: 20.02Min: 19.23 / Avg: 20.27 / Max: 21.16Min: 38.68 / Avg: 39.26 / Max: 40.29Min: 51.07 / Avg: 60.47 / Max: 62.64Min: 38.39 / Avg: 39.98 / Max: 41.26Min: 50.14 / Avg: 61.56 / Max: 64.57Min: 48.18 / Avg: 50.59 / Max: 52.49Min: 21.18 / Avg: 25.08 / Max: 26.48

OpenBenchmarking.orgMegahertz, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.4CPU Peak Freq (Highest CPU Core Frequency) MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248810002000300040005000Min: 400 / Avg: 3114.9 / Max: 4941Min: 400 / Avg: 2142.27 / Max: 5333Min: 5263 / Avg: 5323.75 / Max: 5448Min: 5237 / Avg: 5250.46 / Max: 5268Min: 5025 / Avg: 5211.54 / Max: 5240Min: 5293 / Avg: 5472.62 / Max: 5536Min: 5016 / Avg: 5188.92 / Max: 5322Min: 3930 / Avg: 4692 / Max: 5204

OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.4Test: r2c - Backend: FFTW - Precision: double - X Y Z: 128EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.33930.67861.01791.35721.69651.5081.2030.6040.5790.5680.6050.9941.056

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.4Test: r2c - Backend: FFTW - Precision: double - X Y Z: 128EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881122334455SE +/- 0.06, N = 14SE +/- 0.13, N = 14SE +/- 0.15, N = 15SE +/- 0.59, N = 15SE +/- 0.52, N = 15SE +/- 0.58, N = 15SE +/- 1.39, N = 15SE +/- 0.39, N = 1527.8324.3823.7235.0222.7037.2450.2826.481. (CXX) g++ options: -O3

OpenBenchmarking.orgWatts, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.4System Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248820406080100Min: 52 / Avg: 54.02 / Max: 58.9Min: 52.2 / Avg: 54.78 / Max: 62.4Min: 71.5 / Avg: 73.41 / Max: 79.4Min: 99.1 / Avg: 99.42 / Max: 100.1Min: 73.2 / Avg: 73.51 / Max: 74Min: 95.4 / Avg: 99.75 / Max: 109.7Min: 83.2 / Avg: 85.68 / Max: 87.5Min: 53.5 / Avg: 59.35 / Max: 70.9

OpenBenchmarking.orgWatts, Fewer Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.4CPU Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24881224364860Min: 14.83 / Avg: 17.84 / Max: 20.5Min: 16.26 / Avg: 18.88 / Max: 21.72Min: 32.99 / Avg: 38.11 / Max: 39.6Min: 60.81 / Avg: 61.43 / Max: 62.08Min: 32.77 / Avg: 38.09 / Max: 39.13Min: 60.66 / Avg: 61.58 / Max: 63.13Min: 47.01 / Avg: 49.09 / Max: 50.56Min: 22.31 / Avg: 24.02 / Max: 25.79

OpenBenchmarking.orgMegahertz, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.4CPU Peak Freq (Highest CPU Core Frequency) MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248810002000300040005000Min: 400 / Avg: 1929.92 / Max: 5134Min: 400 / Avg: 3317 / Max: 5164Min: 5246 / Avg: 5314.36 / Max: 5350Min: 4985 / Avg: 5213.08 / Max: 5267Min: 5088 / Avg: 5203.15 / Max: 5240Min: 5434 / Avg: 5488.82 / Max: 5537Min: 5041 / Avg: 5231.64 / Max: 5462Min: 4293 / Avg: 4852.31 / Max: 5224

OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.4Test: r2c - Backend: FFTW - Precision: float - X Y Z: 128EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.96441.92882.89323.85764.8224.2864.0412.0341.7542.9862.0772.8333.540

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.4Test: r2c - Backend: FFTW - Precision: float - X Y Z: 128EPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150SE +/- 2.17, N = 15SE +/- 1.45, N = 15SE +/- 1.52, N = 15SE +/- 1.06, N = 15SE +/- 0.82, N = 15SE +/- 1.14, N = 15SE +/- 1.72, N = 15SE +/- 3.73, N = 1576.4476.2877.52107.74113.74127.93139.0685.011. (CXX) g++ options: -O3

ASKAP

OpenBenchmarking.orgWatts, Fewer Is BetterASKAP 1.0System Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850100150200250Min: 53.4 / Avg: 86.1 / Max: 108.4Min: 54.3 / Avg: 94.27 / Max: 115.1Min: 72.9 / Avg: 131.71 / Max: 175.2Min: 98.7 / Avg: 133.75 / Max: 140.5Min: 72.9 / Avg: 123.03 / Max: 156.5Min: 99.6 / Avg: 174.53 / Max: 252.9Min: 84.1 / Avg: 138.63 / Max: 191Min: 54.6 / Avg: 142.3 / Max: 224.5

OpenBenchmarking.orgWatts, Fewer Is BetterASKAP 1.0CPU Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150Min: 20.37 / Avg: 44.71 / Max: 54.1Min: 16.89 / Avg: 53.42 / Max: 64.66Min: 52.24 / Avg: 77.99 / Max: 92.16Min: 60.35 / Avg: 86.14 / Max: 90.4Min: 37.3 / Avg: 75.25 / Max: 93.98Min: 60.25 / Avg: 116.42 / Max: 162.58Min: 49.43 / Avg: 84.77 / Max: 112.51Min: 38.32 / Avg: 95.55 / Max: 124.72

OpenBenchmarking.orgMegahertz, More Is BetterASKAP 1.0CPU Peak Freq (Highest CPU Core Frequency) MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248810002000300040005000Min: 400 / Avg: 4571.25 / Max: 5136Min: 400 / Avg: 4768.77 / Max: 5141Min: 5052 / Avg: 5291.68 / Max: 5464Min: 4659 / Avg: 5160.76 / Max: 5268Min: 4772 / Avg: 5034.85 / Max: 5190Min: 4983 / Avg: 5274.91 / Max: 5500Min: 4782 / Avg: 5076.91 / Max: 5262Min: 4480 / Avg: 5025.92 / Max: 5400

OpenBenchmarking.orgMpix/sec Per Watt, More Is BetterASKAP 1.0Test: tConvolve MPI - GriddingEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150133.37109.9754.5785.78100.0664.23115.1564.88

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - GriddingEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882K4K6K8K10KSE +/- 0.00, N = 3SE +/- 44.18, N = 3SE +/- 180.65, N = 12SE +/- 528.48, N = 12SE +/- 175.34, N = 15SE +/- 298.02, N = 15SE +/- 384.12, N = 15SE +/- 48.43, N = 35963.545875.194256.287389.247529.937477.199760.756199.091. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - DegriddingEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24882K4K6K8K10KSE +/- 38.09, N = 3SE +/- 34.30, N = 3SE +/- 192.19, N = 12SE +/- 473.28, N = 12SE +/- 187.79, N = 15SE +/- 321.53, N = 15SE +/- 346.60, N = 15SE +/- 58.27, N = 34723.735179.313635.696270.826140.896642.948364.075146.331. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

LAMMPS Molecular Dynamics Simulator

OpenBenchmarking.orgWatts, Fewer Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022System Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850100150200250Min: 53.7 / Avg: 135.94 / Max: 137Min: 54.3 / Avg: 135.62 / Max: 136.3Min: 74.2 / Avg: 176.91 / Max: 190.1Min: 94.6 / Avg: 136.23 / Max: 137.3Min: 74.7 / Avg: 147.98 / Max: 151Min: 102.7 / Avg: 240.84 / Max: 257.1Min: 84.6 / Avg: 184.42 / Max: 189.6Min: 54.5 / Avg: 142.79 / Max: 236.4

OpenBenchmarking.orgWatts, Fewer Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022CPU Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150Min: 16.77 / Avg: 89.63 / Max: 90.39Min: 27.71 / Avg: 89.73 / Max: 90.38Min: 36.84 / Avg: 120.36 / Max: 131.88Min: 56.97 / Avg: 89.85 / Max: 90.41Min: 37.33 / Avg: 96.74 / Max: 100.04Min: 61.31 / Avg: 169.46 / Max: 180.85Min: 46.78 / Avg: 122.74 / Max: 127.89Min: 22.3 / Avg: 94.86 / Max: 139

OpenBenchmarking.orgMegahertz, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022CPU Peak Freq (Highest CPU Core Frequency) MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248810002000300040005000Min: 400 / Avg: 4973.27 / Max: 5113Min: 400 / Avg: 4913.12 / Max: 5036Min: 4897 / Avg: 5080.08 / Max: 5335Min: 4579 / Avg: 4728.74 / Max: 5260Min: 4731 / Avg: 4794.5 / Max: 5215Min: 4809 / Avg: 4965.28 / Max: 5499Min: 4714 / Avg: 4784.39 / Max: 5207Min: 2400 / Avg: 4332.92 / Max: 5400

OpenBenchmarking.orgns/day Per Watt, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: 20k AtomsEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.0270.0540.0810.1080.1350.0760.1000.0680.1160.0990.0810.1200.080

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: 20k AtomsEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248848121620SE +/- 0.014, N = 3SE +/- 0.004, N = 3SE +/- 0.174, N = 9SE +/- 0.191, N = 9SE +/- 0.108, N = 9SE +/- 0.297, N = 9SE +/- 0.298, N = 6SE +/- 0.177, N = 96.8388.9518.21510.3939.53913.69114.6807.5791. (CXX) g++ options: -O3 -lm -ldl

ACES DGEMM

OpenBenchmarking.orgWatts, Fewer Is BetterACES DGEMM 1.0System Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248850100150200250Min: 53 / Avg: 98.91 / Max: 111.4Min: 53.9 / Avg: 124.65 / Max: 133.5Min: 72 / Avg: 147.16 / Max: 175.9Min: 87.9 / Avg: 126.29 / Max: 137.1Min: 74 / Avg: 118.09 / Max: 149Min: 94.4 / Avg: 192.48 / Max: 264.4Min: 83.9 / Avg: 158.95 / Max: 189Min: 53.1 / Avg: 140.86 / Max: 198.1

OpenBenchmarking.orgWatts, Fewer Is BetterACES DGEMM 1.0CPU Power Consumption MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-2488306090120150Min: 17.24 / Avg: 64.71 / Max: 73.16Min: 32.7 / Avg: 81.36 / Max: 90.56Min: 36.54 / Avg: 99.71 / Max: 121.4Min: 54.86 / Avg: 82.61 / Max: 90.52Min: 45.64 / Avg: 75.73 / Max: 96.66Min: 64.62 / Avg: 140.35 / Max: 184.02Min: 40.66 / Avg: 100.01 / Max: 127.23Min: 21.45 / Avg: 96.15 / Max: 139.58

OpenBenchmarking.orgMegahertz, More Is BetterACES DGEMM 1.0CPU Peak Freq (Highest CPU Core Frequency) MonitorEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-248810002000300040005000Min: 400 / Avg: 4835.39 / Max: 5139Min: 400 / Avg: 4974.8 / Max: 5340Min: 5195 / Avg: 5279.57 / Max: 5489Min: 4850 / Avg: 5060.65 / Max: 5440Min: 5014 / Avg: 5093.65 / Max: 5190Min: 5089 / Avg: 5251.73 / Max: 5529Min: 4970 / Avg: 5072.5 / Max: 5247Min: 4014 / Avg: 4857.43 / Max: 5452

OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24880.02390.04780.07170.09560.11950.0610.0650.0530.0870.0740.0680.1060.044

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateEPYC 4244PEPYC 4344PEPYC 4364PEPYC 4464PEPYC 4484PXEPYC 4564PEPYC 4584PXXeon E-24883691215SE +/- 0.039171, N = 3SE +/- 0.037881, N = 3SE +/- 0.144416, N = 15SE +/- 0.115135, N = 15SE +/- 0.066520, N = 3SE +/- 0.258971, N = 12SE +/- 0.322380, N = 15SE +/- 0.054700, N = 33.9201685.3136495.3305007.1870625.6045029.57775310.5916514.2041101. (CC) gcc options: -O3 -march=native -fopenmp

385 Results Shown

John The Ripper
OpenVINO:
  Weld Porosity Detection FP16 - CPU
  Face Detection FP16 - CPU
  Handwritten English Recognition FP16 - CPU
OpenSSL
srsRAN Project
OpenVINO:
  Person Re-Identification Retail FP16 - CPU
  Age Gender Recognition Retail 0013 FP16 - CPU
  Handwritten English Recognition FP16-INT8 - CPU
John The Ripper:
  MD5
  HMAC-SHA512
NAMD
OpenVINO:
  Person Re-Identification Retail FP16 - CPU
  Face Detection Retail FP16 - CPU
  Road Segmentation ADAS FP16 - CPU
  Machine Translation EN To DE FP16 - CPU
  Vehicle Detection FP16 - CPU
PostgreSQL:
  100 - 800 - Read Only - Average Latency
  100 - 800 - Read Only
miniBUDE:
  OpenMP - BM1:
    Billion Interactions/s
    GFInst/s
OpenVINO
miniBUDE:
  OpenMP - BM2:
    GFInst/s
    Billion Interactions/s
OpenVINO
OpenSSL:
  AES-128-GCM
  ChaCha20-Poly1305
  AES-256-GCM
  ChaCha20
NAMD
OpenVINO:
  Noise Suppression Poconet-Like FP16 - CPU
  Vehicle Detection FP16-INT8 - CPU
PostgreSQL:
  100 - 500 - Read Only - Average Latency
  100 - 500 - Read Only
OpenVINO:
  Vehicle Detection FP16-INT8 - CPU
  Person Detection FP32 - CPU
  Person Detection FP16 - CPU
Primesieve
OpenVINO
Primesieve
Cpuminer-Opt:
  Triple SHA-256, Onecoin
  Quad SHA-256, Pyrite
OSPRay Studio:
  1 - 1080p - 16 - Path Tracer - CPU
  2 - 1080p - 16 - Path Tracer - CPU
LuxCoreRender
OSPRay Studio
Stress-NG
oneDNN
OSPRay
7-Zip Compression
LuxCoreRender
OpenVINO:
  Age Gender Recognition Retail 0013 FP16-INT8 - CPU
  Road Segmentation ADAS FP16-INT8 - CPU
OpenSSL
RocksDB
Stockfish
OSPRay
OpenVINO
OSPRay Studio:
  2 - 1080p - 1 - Path Tracer - CPU
  1 - 1080p - 1 - Path Tracer - CPU
  3 - 1080p - 1 - Path Tracer - CPU
OpenVINO
Cpuminer-Opt
ASTC Encoder
Blender
ASTC Encoder
Cpuminer-Opt
ASTC Encoder
OpenVINO
m-queens
Embree
OpenVINO
OSPRay
Aircrack-ng
OpenVINO
OSPRay
ASTC Encoder
Java JMH
ASTC Encoder
OpenSSL
Cpuminer-Opt
ASKAP
OpenSSL
Stress-NG
ASKAP
Blender
oneDNN
SVT-HEVC
Cpuminer-Opt:
  Deepcoin
  scrypt
Blender
Stress-NG
Helsing
OSPRay
Cpuminer-Opt
OSPRay Studio
DaCapo Benchmark
Blender
Coremark
Cpuminer-Opt
QuantLib
Stress-NG:
  Fused Multiply-Add
  AVX-512 VNNI
Blender
OSPRay Studio
Blender
Cpuminer-Opt
Chaos Group V-RAY
Speedb
SVT-HEVC
OSPRay Studio
Stress-NG
Cpuminer-Opt
GraphicsMagick
Helsing
OpenVINO:
  Road Segmentation ADAS FP16-INT8 - CPU
  Age Gender Recognition Retail 0013 FP16 - CPU
John The Ripper:
  Blowfish
  bcrypt
GraphicsMagick
Appleseed
OpenVINO
Memcached
uvg266
Timed Linux Kernel Compilation
LuxCoreRender
Intel Open Image Denoise
TensorFlow
OpenVKL
Embree
OpenVINO
Intel Open Image Denoise:
  RT.ldr_alb_nrm.3840x2160 - CPU-Only
  RT.hdr_alb_nrm.3840x2160 - CPU-Only
Embree
Kvazaar
Timed LLVM Compilation
GraphicsMagick
TensorFlow
DaCapo Benchmark
LuxCoreRender
oneDNN
BRL-CAD
oneDNN
RocksDB
Timed Node.js Compilation
Parallel BZIP2 Compression
uvg266
SVT-HEVC:
  7 - Bosphorus 1080p
  7 - Bosphorus 4K
oneDNN
Memcached
NAS Parallel Benchmarks
oneDNN
nginx
Timed Godot Game Engine Compilation
uvg266
Appleseed
nginx
OpenVINO
uvg266
Timed LLVM Compilation
Speedb
Timed Linux Kernel Compilation
Kvazaar
OpenFOAM
nginx
JPEG-XL libjxl
SPECFEM3D
JPEG-XL libjxl
NAS Parallel Benchmarks
GROMACS
7-Zip Compression
Timed FFmpeg Compilation
SPECFEM3D
SVT-VP9
Kvazaar
OpenVINO
ONNX Runtime
OpenVINO
x265
uvg266
Kvazaar
Timed Mesa Compilation
SPECFEM3D
oneDNN
Timed Gem5 Compilation
SVT-AV1
TensorFlow
SPECFEM3D
SVT-HEVC
PostgreSQL:
  100 - 800 - Read Write - Average Latency
  100 - 800 - Read Write
SPECFEM3D
Zstd Compression:
  8 - Compression Speed
  8, Long Mode - Compression Speed
libavif avifenc
GraphicsMagick
SVT-VP9
uvg266
SVT-VP9
Kvazaar
SVT-VP9
uvg266
ScyllaDB
uvg266
OpenVINO:
  Person Detection FP32 - CPU
  Noise Suppression Poconet-Like FP16 - CPU
LuxCoreRender
Y-Cruncher
OpenVINO
libavif avifenc
CockroachDB
ONNX Runtime
SVT-HEVC
Kvazaar
Appleseed
Pennant
Liquid-DSP
CockroachDB
TensorFlow
CockroachDB
PostgreSQL:
  100 - 500 - Read Write
  100 - 500 - Read Write - Average Latency
CloverLeaf
TensorFlow
ONNX Runtime
Kvazaar
LAMMPS Molecular Dynamics Simulator
NAS Parallel Benchmarks
OpenRadioss
NAS Parallel Benchmarks
SVT-AV1
TensorFlow
SVT-VP9
Kvazaar
TensorFlow
Pennant
TensorFlow
Y-Cruncher
ONNX Runtime
SVT-VP9
libavif avifenc
TensorFlow:
  CPU - 256 - ResNet-50
  CPU - 64 - ResNet-50
  CPU - 32 - ResNet-50
  CPU - 32 - AlexNet
OpenRadioss
SVT-AV1
VVenC
libavif avifenc
ONNX Runtime:
  bertsquad-12 - CPU - Parallel
  yolov4 - CPU - Parallel
GraphicsMagick
MariaDB
OSPRay
VVenC
Zstd Compression
RocksDB
JPEG-XL libjxl:
  PNG - 90
  JPEG - 90
ONNX Runtime
OpenVINO
NAS Parallel Benchmarks
Xcompact3d Incompact3d
SVT-AV1
Timed PHP Compilation
Speedb
GPAW
SVT-AV1
MariaDB
srsRAN Project
VVenC
DaCapo Benchmark
Zstd Compression:
  19, Long Mode - Compression Speed
  19 - Compression Speed
SVT-AV1
x265
ONNX Runtime
Liquid-DSP
OpenFOAM
PyBench
OpenFOAM
MariaDB
PyTorch
Timed Wasmer Compilation
OpenVINO
OpenRadioss
Google Draco
DuckDB
DaCapo Benchmark
MariaDB
ClickHouse
GraphicsMagick
OpenVINO
Xcompact3d Incompact3d
OpenVINO
OpenRadioss
Apache IoTDB
ClickHouse
ONNX Runtime
Apache IoTDB
ClickHouse
VVenC
Z3 Theorem Prover
Timed CPython Compilation
Apache IoTDB:
  800 - 100 - 800 - 400:
    point/sec
    Average Latency
srsRAN Project
OpenFOAM
ONNX Runtime
Apache IoTDB
ONNX Runtime
Apache IoTDB:
  800 - 100 - 500 - 100
  500 - 100 - 800 - 400
  500 - 100 - 800 - 400
  800 - 100 - 500 - 100
  500 - 100 - 800 - 100
Liquid-DSP
TensorFlow
Apache IoTDB:
  800 - 100 - 500 - 400:
    Average Latency
    point/sec
  500 - 100 - 500 - 100:
    Average Latency
    point/sec
  500 - 100 - 500 - 400:
    point/sec
DaCapo Benchmark
TensorFlow
Renaissance
Apache IoTDB
PyTorch
Numpy Benchmark
srsRAN Project
FFmpeg:
  libx265 - Video On Demand
  libx265 - Platform
  libx265 - Upload
MariaDB mariadb-slap
GNU Octave Benchmark
MariaDB mariadb-slap
ONNX Runtime
FFmpeg
Liquid-DSP
Timed CPython Compilation
System Power Consumption Monitor:
  Phoronix Test Suite System Monitoring:
    Watts
    Watts
    Megahertz
ONNX Runtime:
  ResNet50 v1-12-int8 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
SVT-AV1:
  System Power Consumption Monitor
  CPU Power Consumption Monitor
  CPU Peak Freq (Highest CPU Core Frequency) Monitor
  Preset 13 - Bosphorus 4K
SVT-AV1
OpenRadioss:
  System Power Consumption Monitor
  CPU Power Consumption Monitor
  CPU Peak Freq (Highest CPU Core Frequency) Monitor
OpenRadioss
OpenRadioss:
  System Power Consumption Monitor
  CPU Power Consumption Monitor
  CPU Peak Freq (Highest CPU Core Frequency) Monitor
OpenRadioss
HeFFTe - Highly Efficient FFT for Exascale:
  System Power Consumption Monitor
  CPU Power Consumption Monitor
  CPU Peak Freq (Highest CPU Core Frequency) Monitor
  r2c - FFTW - double - 128
HeFFTe - Highly Efficient FFT for Exascale
HeFFTe - Highly Efficient FFT for Exascale:
  System Power Consumption Monitor
  CPU Power Consumption Monitor
  CPU Peak Freq (Highest CPU Core Frequency) Monitor
  r2c - FFTW - float - 128
HeFFTe - Highly Efficient FFT for Exascale
ASKAP:
  System Power Consumption Monitor
  CPU Power Consumption Monitor
  CPU Peak Freq (Highest CPU Core Frequency) Monitor
  tConvolve MPI - Gridding
ASKAP:
  tConvolve MPI - Gridding
  tConvolve MPI - Degridding
LAMMPS Molecular Dynamics Simulator:
  System Power Consumption Monitor
  CPU Power Consumption Monitor
  CPU Peak Freq (Highest CPU Core Frequency) Monitor
  20k Atoms
LAMMPS Molecular Dynamics Simulator
ACES DGEMM:
  System Power Consumption Monitor
  CPU Power Consumption Monitor
  CPU Peak Freq (Highest CPU Core Frequency) Monitor
  Sustained Floating-Point Rate
ACES DGEMM