AMD Ryzen 9 9950X DDR5 Memory Performance

AMD Ryzen 9 9950X DDR5 memory module benchmarks by Michael Larabel for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2408225-NE-RYZEN999510
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts
Allow Limiting Results To Certain Suite(s)

Statistics

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

Graph Settings

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

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Toggle/Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G
August 14
  13 Hours, 26 Minutes
2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32
August 22
  10 Hours, 21 Minutes
2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36
August 20
  10 Hours, 59 Minutes
2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38
August 21
  7 Hours, 49 Minutes
Invert Behavior (Only Show Selected Data)
  10 Hours, 39 Minutes

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


AMD Ryzen 9 9950X DDR5 Memory PerformanceOpenBenchmarking.orgPhoronix Test SuiteAMD Ryzen 9 9950X 16-Core @ 5.75GHz (16 Cores / 32 Threads)ASUS ROG STRIX X670E-E GAMING WIFI (2204 BIOS)AMD Device 14d82 x 16GB DDR5-6000MT/s G Skill F5-6000J3038F16G2 x 32GB DDR5-6400MT/s Corsair CMK64GX5M2B6400C322 x 16GB DDR5-8000MT/s Corsair CMH32GX5M2X8000C362 x 24GB DDR5-8000MT/s Corsair CMP48GX5M2X8000C382000GB Corsair MP700 PROAMD Radeon RX 7900 GRE 16GBAMD Navi 31 HDMI/DPDELL U2723QEIntel I225-V + Intel Wi-Fi 6EUbuntu 24.046.10.0-phx (x86_64)GNOME Shell 46.0X Server + Wayland4.6 Mesa 24.2~git2406040600.8112d4~oibaf~n (git-8112d44 2024-06-04 noble-oibaf-ppa) (LLVM 17.0.6 DRM 3.57)GCC 13.2.0ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLCompilerFile-SystemScreen ResolutionAMD Ryzen 9 9950X DDR5 Memory Performance BenchmarksSystem Logs- Transparent Huge Pages: madvise- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xb40401a- Python 3.12.3- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: 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; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected - 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32, 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36, 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: OpenJDK Runtime Environment (build 21.0.3+9-Ubuntu-1ubuntu1)

2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38Result OverviewPhoronix Test Suite100%123%145%168%191%OpenFOAMOpenRadiossLeelaChessZeroXcompact3d Incompact3dlibxsmmHigh Performance Conjugate GradientLlama.cppLULESHNAS Parallel BenchmarksRAMspeed SMPBuild2PyTorchNAMDStockfishSPECFEM3DTensorFlowMBWMemcachedEmbreeJava JMHXNNPACKsimdjsonx265BlenderBRL-CADXmrigNumpy BenchmarkLuxCoreRenderQuicksilverPOV-RayminiBUDEEtcpakStress-NG

AMD Ryzen 9 9950X DDR5 Memory Performancecompress-7zip: Compression Ratingcompress-7zip: Decompression Ratingblender: BMW27 - CPU-Onlyblender: Junkshop - CPU-Onlyblender: Barbershop - CPU-Onlybrl-cad: VGR Performance Metricbuild2: Time To Compileembree: Pathtracer ISPC - Crownembree: Pathtracer ISPC - Asian Dragonetcpak: Multi-Threaded - ETC2gromacs: water_GMX50_baregromacs: MPI CPU - water_GMX50_barehpcg: 104 104 104 - 60java-jmh: Throughputlczero: BLASlczero: Eigenlibxsmm: 128libxsmm: 32libxsmm: 64llama-cpp: Meta-Llama-3-8B-Instruct-Q8_0.ggufllamafile: Meta-Llama-3-8B-Instruct.F16 - CPUlulesh: luxcorerender: DLSC - CPUluxcorerender: Danish Mood - CPUluxcorerender: Orange Juice - CPUluxcorerender: LuxCore Benchmark - CPUluxcorerender: Rainbow Colors and Prism - CPUmbw: Memory Copy - 4096 MiBmbw: Memory Copy - 8192 MiBmbw: Memory Copy, Fixed Block Size - 4096 MiBmbw: Memory Copy, Fixed Block Size - 8192 MiBmemcached: 1:10memcached: 1:100minibude: OpenMP - BM1minibude: OpenMP - BM1minibude: OpenMP - BM2minibude: OpenMP - BM2namd: ATPase with 327,506 Atomsnamd: STMV with 1,066,628 Atomsnpb: BT.Cnpb: CG.Cnpb: EP.Cnpb: EP.Dnpb: FT.Cnpb: IS.Dnpb: LU.Cnpb: MG.Cnpb: SP.Bnpb: SP.Cnumpy: openfoam: motorBike - Mesh Timeopenfoam: motorBike - Execution Timeopenfoam: drivaerFastback, Small Mesh Size - Mesh Timeopenfoam: drivaerFastback, Small Mesh Size - Execution Timeopenfoam: drivaerFastback, Medium Mesh Size - Mesh Timeopenfoam: drivaerFastback, Medium Mesh Size - Execution Timeopenradioss: Bumper Beamopenradioss: Chrysler Neon 1Mopenradioss: Cell Phone Drop Testopenradioss: Bird Strike on Windshieldopenradioss: Rubber O-Ring Seal Installationopenradioss: INIVOL and Fluid Structure Interaction Drop Containerpovray: Trace Timepytorch: CPU - 1 - ResNet-50pytorch: CPU - 256 - ResNet-50quicksilver: CORAL2 P1ramspeed: Add - Integerramspeed: Copy - Integerramspeed: Scale - Integerramspeed: Triad - Integerramspeed: Average - Integersimdjson: Kostyasimdjson: TopTweetsimdjson: LargeRandspecfem3d: Mount St. Helensspecfem3d: Layered Halfspacespecfem3d: Tomographic Modelspecfem3d: Homogeneous Halfspacespecfem3d: Water-layered Halfspacestockfish: Chess Benchmarkstress-ng: Memory Copyingtensorflow: CPU - 1 - ResNet-50tensorflow: CPU - 64 - ResNet-50build-gem5: Time To Compilebuild-linux-kernel: defconfigbuild-linux-kernel: allmodconfigbuild-llvm: Ninjabuild-nodejs: Time To Compilex265: Bosphorus 4Kx265: Bosphorus 1080pincompact3d: input.i3d 129 Cells Per Directionincompact3d: input.i3d 193 Cells Per Directionxmrig: GhostRider - 1Mxnnpack: FP32MobileNetV2xnnpack: FP32MobileNetV3Largexnnpack: FP32MobileNetV3Smallxnnpack: FP16MobileNetV2xnnpack: FP16MobileNetV3Largexnnpack: FP16MobileNetV3Smallxnnpack: QU8MobileNetV2xnnpack: QU8MobileNetV3Largexnnpack: QU8MobileNetV3Smally-cruncher: 1By-cruncher: 500M2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C3819558515813346.3862.09448.9649896676.00336.553241.4914741.7841.9273.1418.9356091554717298.426226220472.0121.0237.18.364.689945.53095.404.628.645.1519.9922774.94722247.78419534.15919467.3945915791.347507615.731922.12576.8851934.61377.3853.410070.9808455287.3711144.883678.463823.8027106.181505.4759920.9624236.2121328.2814528.261068.3684.994958.569522.579849157.14898181.569052096.566476.90705.5544.90128.7653.16247.0616.10184.8957.542567666761079.9365797.0569141.0862428.7065503.937.7113.592.1225.69098664473.13879708423.35909292030.67791845772.1972605125137777810752.8116.9252.27221.13947.755590.495301.629345.24638.27134.5214.090555264.11302183616.6126415168129851291770759111875616.4527.58720306515751246.5662.21450.7149443777.00636.109940.9447739.7361.9963.2288.7973792045826799.54244236505.9128.0253.38.3210235.7225.424.698.645.1420.0822510.46422294.06220817.56920850.8835898565.347647187.481926.89177.0761931.14177.2463.604641.0160258570.7711374.513731.973714.6628891.861584.9362084.7225622.9322908.5815610.301070.6184.980257.245222.821389148.75845177.78941994.74777.18668.1643.42128.0153.23239.2216.13385.5258.892566000063967.7070747.0071754.5865845.2168588.847.6313.382.1124.95654992869.43938729923.00050134129.81843126869.5616424674919415510720.3117.3454.30219.26847.783589.331298.049341.67938.30134.4813.402805360.79924393638.6127315278159961303776765111576115.4267.24720081015700547.7263.41459.1248695477.92336.254641.7238734.5332.0093.2989.5160189012581592.653264247508.9132.2258.69.0010578.5165.364.658.555.0419.8422539.72022817.90020923.98920349.4145874747.737498825.191903.04076.1221904.95376.1983.577551.0242660124.0811908.473741.863739.1629713.551635.3464031.3126596.4623254.4016097.001046.9188.080755.573823.143631145.88323174.125721909.665577.45573.3343.25128.3253.66242.0016.39586.7958.492566666765113.5270345.5373724.8866963.0369491.427.5713.242.1025.19465363069.81742231123.28751738830.54298448768.9209356514997421010608.9617.3355.1948.258304.62138.07134.1912.974287058.19845963699.6127215538189971302773765112476415.6067.23647.5164.08459.1948978480.10435.182840.0696731.2299.1659090340365706.792227491.1127.0249.78.669832.63855.344.608.465.0519.3322093.98521444.00219747.58420124.9435694123.527350932.881899.80475.9921898.25875.9303.509580.9985258570.0311662.443722.563712.4528743.491586.5962375.4325715.3422436.8115531.761064.1785.474577.81124.3754.2716.31981.8956.192513666762771.0066545.2470445.7464832.8666822.497.5012.882.1125.45331906072.438348923.86945318131.98171567772.0794766585010139610608.8216.8153.5437.09131.2813.259573360.48058323676.512851590836101813347977831154786OpenBenchmarking.org

7-Zip Compression

OpenBenchmarking.orgMIPS, More Is Better7-Zip CompressionTest: Compression Rating2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 16GB DDR5-6000 CL30 F5-6000J3038F16G40K80K120K160K200KSE +/- 467.61, N = 3SE +/- 94.03, N = 3SE +/- 322.74, N = 32030652008101955851. 7-Zip 23.01 (x64) : Copyright (c) 1999-2023 Igor Pavlov : 2023-06-20

Test: Compression Rating

2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: The test quit with a non-zero exit status.

OpenBenchmarking.orgMIPS, More Is Better7-Zip CompressionTest: Decompression Rating2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C3630K60K90K120K150KSE +/- 14.15, N = 3SE +/- 69.90, N = 3SE +/- 33.91, N = 31581331575121570051. 7-Zip 23.01 (x64) : Copyright (c) 1999-2023 Igor Pavlov : 2023-06-20

Test: Decompression Rating

2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: The test quit with a non-zero exit status.

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: BMW27 - Compute: CPU-Only2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C361122334455SE +/- 0.01, N = 3SE +/- 0.09, N = 3SE +/- 0.02, N = 3SE +/- 0.10, N = 346.3846.5647.5147.72

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Junkshop - Compute: CPU-Only2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C381428425670SE +/- 0.12, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.17, N = 362.0962.2163.4164.08

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.2Blend File: Barbershop - Compute: CPU-Only2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38100200300400500SE +/- 0.20, N = 3SE +/- 0.26, N = 3SE +/- 0.27, N = 3SE +/- 0.20, N = 3448.96450.71459.12459.19

BRL-CAD

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

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.38.2VGR Performance Metric2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36110K220K330K440K550K4989664944374897844869541. (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

Build2

OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.17Time To Compile2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C3820406080100SE +/- 0.16, N = 3SE +/- 0.52, N = 3SE +/- 0.40, N = 3SE +/- 0.99, N = 276.0077.0177.9280.10

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: Crown2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38816243240SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 336.5536.2536.1135.18MIN: 36.13 / MAX: 37.34MIN: 35.89 / MAX: 36.94MIN: 35.69 / MAX: 36.98MIN: 34.83 / MAX: 36.01

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.3Binary: Pathtracer ISPC - Model: Asian Dragon2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C381020304050SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 341.7241.4940.9440.07MIN: 41.48 / MAX: 42.26MIN: 41.24 / MAX: 42.02MIN: 40.73 / MAX: 41.6MIN: 39.86 / MAX: 40.55

Etcpak

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 2.0Benchmark: Multi-Threaded - Configuration: ETC22 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38160320480640800SE +/- 1.25, N = 3SE +/- 0.38, N = 3SE +/- 1.36, N = 3SE +/- 0.82, N = 3741.78739.74734.53731.231. (CXX) g++ options: -flto -pthread

GROMACS

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACSInput: water_GMX50_bare2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G0.4520.9041.3561.8082.26SE +/- 0.001, N = 3SE +/- 0.000, N = 3SE +/- 0.002, N = 32.0091.9961.9271. GROMACS version: 2023.3-Ubuntu_2023.3_1ubuntu3

Input: water_GMX50_bare

2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: The test quit with a non-zero exit status. E: Fatal error:

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2024Implementation: MPI CPU - Input: water_GMX50_bare2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G0.74211.48422.22632.96843.7105SE +/- 0.002, N = 3SE +/- 0.004, N = 3SE +/- 0.001, N = 33.2983.2283.1411. (CXX) g++ options: -O3 -lm

Implementation: MPI CPU - Input: water_GMX50_bare

2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: The test quit with a non-zero exit status. E: mpirun noticed that process rank 0 with PID 0 on node phoronix-System-Product-Name exited on signal 11 (Segmentation fault).

High Performance Conjugate Gradient

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

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 104 104 104 - RT: 602 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C323691215SE +/- 0.00741, N = 3SE +/- 0.00584, N = 3SE +/- 0.00142, N = 3SE +/- 0.00125, N = 39.516019.165908.935608.797371. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

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 JMHThroughput2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C3620000M40000M60000M80000M100000M92045826799.5491554717298.4390340365706.7989012581592.65

LeelaChessZero

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.31.1Backend: BLAS2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G60120180240300SE +/- 3.79, N = 3SE +/- 2.91, N = 3SE +/- 2.65, N = 32642442261. (CXX) g++ options: -flto -pthread

Backend: BLAS

2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: The test quit with a non-zero exit status.

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.31.1Backend: Eigen2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G50100150200250SE +/- 1.15, N = 3SE +/- 2.85, N = 3SE +/- 1.45, N = 32472362272201. (CXX) g++ options: -flto -pthread

libxsmm

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

OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 1282 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G110220330440550SE +/- 0.09, N = 3SE +/- 0.10, N = 3SE +/- 0.33, N = 3SE +/- 0.38, N = 3508.9505.9491.1472.01. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -pedantic -O2 -fopenmp -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2

OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G306090120150SE +/- 0.19, N = 3SE +/- 0.07, N = 3SE +/- 0.26, N = 3SE +/- 0.07, N = 3132.2128.0127.0121.01. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -pedantic -O2 -fopenmp -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2

OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 642 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G60120180240300SE +/- 0.37, N = 3SE +/- 0.03, N = 3SE +/- 0.53, N = 3SE +/- 0.20, N = 3258.6253.3249.7237.11. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -pedantic -O2 -fopenmp -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2

Llama.cpp

OpenBenchmarking.orgTokens Per Second, More Is BetterLlama.cpp b3067Model: Meta-Llama-3-8B-Instruct-Q8_0.gguf2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C323691215SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 39.008.668.368.321. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -march=native -mtune=native -lopenblas

Llamafile

OpenBenchmarking.orgTokens Per Second, More Is BetterLlamafile 0.8.6Test: Meta-Llama-3-8B-Instruct.F16 - Acceleration: CPU2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G1.0532.1063.1594.2125.2654.68

LULESH

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

OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.32 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382K4K6K8K10KSE +/- 13.71, N = 3SE +/- 67.88, N = 3SE +/- 19.00, N = 3SE +/- 38.48, N = 310578.5210235.729945.539832.641. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi

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: CPU2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C381.21952.4393.65854.8786.0975SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 35.425.405.365.34MIN: 5.31 / MAX: 5.77MIN: 5.27 / MAX: 5.72MIN: 5.25 / MAX: 5.7MIN: 5.23 / MAX: 5.69

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Danish Mood - Acceleration: CPU2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C381.05532.11063.16594.22125.2765SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 34.694.654.624.60MIN: 2.15 / MAX: 5.26MIN: 2.02 / MAX: 5.24MIN: 2.07 / MAX: 5.22MIN: 2.13 / MAX: 5.16

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Orange Juice - Acceleration: CPU2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38246810SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 38.648.648.558.46MIN: 7.67 / MAX: 9.25MIN: 7.64 / MAX: 9.25MIN: 7.48 / MAX: 9.15MIN: 7.49 / MAX: 9.04

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: LuxCore Benchmark - Acceleration: CPU2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C361.15882.31763.47644.63525.794SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 35.155.145.055.04MIN: 2.38 / MAX: 5.77MIN: 2.39 / MAX: 5.75MIN: 2.28 / MAX: 5.66MIN: 2.28 / MAX: 5.66

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.6Scene: Rainbow Colors and Prism - Acceleration: CPU2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38510152025SE +/- 0.15, N = 3SE +/- 0.07, N = 3SE +/- 0.08, N = 3SE +/- 0.13, N = 320.0819.9919.8419.33MIN: 17.96 / MAX: 20.55MIN: 18.05 / MAX: 20.42MIN: 17.86 / MAX: 20.21MIN: 17.37 / MAX: 19.62

MBW

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

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 4096 MiB2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C385K10K15K20K25KSE +/- 72.38, N = 3SE +/- 31.03, N = 3SE +/- 265.89, N = 3SE +/- 227.86, N = 1522774.9522539.7222510.4622093.991. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 8192 MiB2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C385K10K15K20K25KSE +/- 204.03, N = 3SE +/- 188.50, N = 12SE +/- 49.10, N = 3SE +/- 56.11, N = 322817.9022294.0622247.7821444.001. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiB2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G4K8K12K16K20KSE +/- 211.85, N = 3SE +/- 32.81, N = 3SE +/- 36.42, N = 3SE +/- 24.77, N = 320923.9920817.5719747.5819534.161. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G4K8K12K16K20KSE +/- 58.10, N = 3SE +/- 144.90, N = 15SE +/- 239.88, N = 3SE +/- 10.56, N = 320850.8820349.4120124.9419467.391. (CC) gcc options: -O3 -march=native

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: 1:102 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C381.3M2.6M3.9M5.2M6.5MSE +/- 15831.56, N = 3SE +/- 11599.22, N = 3SE +/- 11157.27, N = 3SE +/- 4372.40, N = 35915791.345898565.345874747.735694123.521. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:1002 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C381.6M3.2M4.8M6.4M8MSE +/- 87376.76, N = 3SE +/- 1150.91, N = 3SE +/- 43634.73, N = 3SE +/- 71150.32, N = 37647187.487507615.737498825.197350932.881. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

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: BM12 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38400800120016002000SE +/- 0.54, N = 3SE +/- 6.84, N = 3SE +/- 0.31, N = 3SE +/- 0.87, N = 31926.891922.131903.041899.801. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM12 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C3820406080100SE +/- 0.02, N = 3SE +/- 0.27, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 377.0876.8976.1275.991. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM22 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38400800120016002000SE +/- 0.64, N = 3SE +/- 0.32, N = 3SE +/- 1.03, N = 3SE +/- 2.32, N = 31934.611931.141904.951898.261. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM22 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C3820406080100SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.09, N = 377.3977.2576.2075.931. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

NAMD

OpenBenchmarking.orgns/day, More Is BetterNAMD 3.0b6Input: ATPase with 327,506 Atoms2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G0.8111.6222.4333.2444.055SE +/- 0.03856, N = 3SE +/- 0.02618, N = 15SE +/- 0.02740, N = 9SE +/- 0.02586, N = 33.604643.577553.509583.41007

OpenBenchmarking.orgns/day, More Is BetterNAMD 3.0b6Input: STMV with 1,066,628 Atoms2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G0.23050.4610.69150.9221.1525SE +/- 0.00057, N = 3SE +/- 0.00200, N = 3SE +/- 0.00137, N = 2SE +/- 0.00027, N = 31.024261.016020.998520.98084

NAS Parallel Benchmarks

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

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: BT.C2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G13K26K39K52K65KSE +/- 85.26, N = 3SE +/- 8.10, N = 3SE +/- 74.71, N = 3SE +/- 54.79, N = 360124.0858570.7758570.0355287.371. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.C2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G3K6K9K12K15KSE +/- 10.26, N = 3SE +/- 41.76, N = 3SE +/- 36.86, N = 3SE +/- 21.24, N = 311908.4711662.4411374.5111144.881. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.C2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G8001600240032004000SE +/- 17.86, N = 3SE +/- 53.14, N = 3SE +/- 21.03, N = 3SE +/- 25.12, N = 33741.863731.973722.563678.461. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.D2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C388001600240032004000SE +/- 2.85, N = 3SE +/- 40.74, N = 3SE +/- 36.74, N = 3SE +/- 51.16, N = 33823.803739.163714.663712.451. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: FT.C2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G6K12K18K24K30KSE +/- 95.45, N = 3SE +/- 30.67, N = 3SE +/- 113.15, N = 3SE +/- 141.73, N = 329713.5528891.8628743.4927106.181. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.D2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G400800120016002000SE +/- 4.49, N = 3SE +/- 5.23, N = 3SE +/- 2.77, N = 3SE +/- 5.33, N = 31635.341586.591584.931505.471. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.C2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G14K28K42K56K70KSE +/- 72.30, N = 3SE +/- 210.02, N = 3SE +/- 130.21, N = 3SE +/- 150.93, N = 364031.3162375.4362084.7259920.961. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.C2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G6K12K18K24K30KSE +/- 34.50, N = 3SE +/- 10.24, N = 3SE +/- 20.77, N = 3SE +/- 6.88, N = 326596.4625715.3425622.9324236.211. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.B2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G5K10K15K20K25KSE +/- 10.38, N = 3SE +/- 26.91, N = 3SE +/- 40.85, N = 3SE +/- 21.50, N = 323254.4022908.5822436.8121328.281. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.C2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G3K6K9K12K15KSE +/- 15.11, N = 3SE +/- 16.93, N = 3SE +/- 42.19, N = 3SE +/- 18.27, N = 316097.0015610.3015531.7614528.261. (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

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 Benchmark2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362004006008001000SE +/- 2.01, N = 3SE +/- 8.61, N = 3SE +/- 4.91, N = 3SE +/- 1.37, N = 31070.611068.361064.171046.91

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: motorBike - Mesh Time2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362040608010084.9884.9985.4788.081. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: motorBike - Execution Time2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G132639526555.5757.2558.571. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Mesh Time2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C3661218243022.5822.8223.141. (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

Input: drivaerFastback, Small Mesh Size - Mesh Time

2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: The test quit with a non-zero exit status. E: [13] #0 Foam::error::printStack(Foam::Ostream&) at ??:?

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Execution Time2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G306090120150145.88148.76157.151. (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

Input: drivaerFastback, Small Mesh Size - Execution Time

2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: The test quit with a non-zero exit status. E: This is not a fatal error but might cause some unexpected behaviour.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Mesh Time2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G4080120160200174.13177.79181.571. (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

Input: drivaerFastback, Medium Mesh Size - Mesh Time

2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: The test quit with a non-zero exit status. E: This is not a fatal error but might cause some unexpected behaviour.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Execution Time2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G50010001500200025001909.671994.752096.571. (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

Input: drivaerFastback, Medium Mesh Size - Execution Time

2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: The test quit with a non-zero exit status. E: [13] --> FOAM FATAL ERROR:

OpenRadioss

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: Bumper Beam2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C3820406080100SE +/- 0.07, N = 3SE +/- 0.14, N = 3SE +/- 0.12, N = 3SE +/- 0.02, N = 376.9077.1877.4577.81

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: Chrysler Neon 1M2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G150300450600750SE +/- 49.36, N = 9SE +/- 1.02, N = 3SE +/- 0.28, N = 3573.33668.16705.55

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: Cell Phone Drop Test2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G1020304050SE +/- 0.15, N = 3SE +/- 0.29, N = 3SE +/- 0.18, N = 343.2543.4244.90

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: Bird Strike on Windshield2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 16GB DDR5-6000 CL30 F5-6000J3038F16G306090120150SE +/- 5.81, N = 15SE +/- 0.35, N = 3SE +/- 0.28, N = 3SE +/- 0.28, N = 3124.37128.01128.32128.76

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: Rubber O-Ring Seal Installation2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C381224364860SE +/- 0.10, N = 3SE +/- 0.15, N = 3SE +/- 0.16, N = 3SE +/- 0.06, N = 353.1653.2353.6654.27

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2023.09.15Model: INIVOL and Fluid Structure Interaction Drop Container2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 16GB DDR5-6000 CL30 F5-6000J3038F16G50100150200250SE +/- 0.99, N = 3SE +/- 0.76, N = 3SE +/- 0.74, N = 3239.22242.00247.06

Model: INIVOL and Fluid Structure Interaction Drop Container

2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: The test run did not produce a result. E: ** ERROR: FILE fsi_drop_container_0000_0001.rst NOT FOUND

POV-Ray

OpenBenchmarking.orgSeconds, Fewer Is BetterPOV-RayTrace Time2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C3648121620SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 316.1016.1316.3216.401. POV-Ray 3.7.0.10.unofficial

PyTorch

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

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: ResNet-502 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C3820406080100SE +/- 0.63, N = 3SE +/- 0.96, N = 4SE +/- 0.27, N = 3SE +/- 0.82, N = 586.7985.5284.8981.89MIN: 78.82 / MAX: 88.23MIN: 68.54 / MAX: 88.53MIN: 77.17 / MAX: 85.87MIN: 67.48 / MAX: 84.5

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 256 - Model: ResNet-502 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C381326395265SE +/- 0.40, N = 15SE +/- 0.36, N = 3SE +/- 0.34, N = 3SE +/- 0.25, N = 358.8958.4957.5456.19MIN: 34.79 / MAX: 62.62MIN: 34.6 / MAX: 60.01MIN: 53.14 / MAX: 58.34MIN: 39.35 / MAX: 57.3

Quicksilver

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

OpenBenchmarking.orgFigure Of Merit, More Is BetterQuicksilver 20230818Input: CORAL2 P12 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C385M10M15M20M25MSE +/- 21858.13, N = 3SE +/- 92074.85, N = 3SE +/- 41633.32, N = 3SE +/- 16666.67, N = 3256766672566666725660000251366671. (CXX) g++ options: -fopenmp -O3 -march=native

RAMspeed SMP

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

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Add - Benchmark: Integer2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G14K28K42K56K70KSE +/- 108.61, N = 3SE +/- 382.92, N = 3SE +/- 64.38, N = 3SE +/- 426.98, N = 365113.5263967.7062771.0061079.931. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Copy - Benchmark: Integer2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G15K30K45K60K75KSE +/- 453.50, N = 3SE +/- 762.95, N = 5SE +/- 299.96, N = 3SE +/- 820.93, N = 370747.0070345.5366545.2465797.051. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Scale - Benchmark: Integer2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G16K32K48K64K80KSE +/- 364.59, N = 3SE +/- 266.14, N = 3SE +/- 513.22, N = 3SE +/- 220.54, N = 373724.8871754.5870445.7469141.081. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Triad - Benchmark: Integer2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G14K28K42K56K70KSE +/- 439.59, N = 3SE +/- 33.42, N = 3SE +/- 130.09, N = 3SE +/- 156.92, N = 366963.0365845.2164832.8662428.701. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Average - Benchmark: Integer2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G15K30K45K60K75KSE +/- 315.00, N = 3SE +/- 204.48, N = 3SE +/- 805.54, N = 4SE +/- 119.74, N = 369491.4268588.8466822.4965503.931. (CC) gcc options: -O3 -march=native

simdjson

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 3.10Throughput Test: Kostya2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38246810SE +/- 0.05, N = 3SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 37.717.637.577.501. (CXX) g++ options: -O3 -lrt

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 3.10Throughput Test: TopTweet2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C383691215SE +/- 0.02, N = 3SE +/- 0.17, N = 3SE +/- 0.15, N = 3SE +/- 0.14, N = 313.5913.3813.2412.881. (CXX) g++ options: -O3 -lrt

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 3.10Throughput Test: LargeRandom2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C360.4770.9541.4311.9082.385SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 32.122.112.112.101. (CXX) g++ options: -O3 -lrt

SPECFEM3D

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

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Mount St. Helens2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G612182430SE +/- 0.12, N = 3SE +/- 0.05, N = 3SE +/- 0.10, N = 3SE +/- 0.17, N = 324.9625.1925.4525.691. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Layered Halfspace2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G1632486480SE +/- 0.59, N = 3SE +/- 0.59, N = 3SE +/- 0.31, N = 369.4469.8272.4473.141. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Tomographic Model2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38612182430SE +/- 0.20, N = 3SE +/- 0.14, N = 3SE +/- 0.16, N = 3SE +/- 0.16, N = 1523.0023.2923.3623.871. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Homogeneous Halfspace2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38714212835SE +/- 0.39, N = 3SE +/- 0.38, N = 3SE +/- 0.28, N = 329.8230.5430.6831.981. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.1.1Model: Water-layered Halfspace2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G1632486480SE +/- 0.24, N = 3SE +/- 0.34, N = 3SE +/- 0.16, N = 2SE +/- 0.34, N = 368.9269.5672.0872.201. (F9X) gfortran options: -O2 -fopenmp -std=f2008 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

Stockfish

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfishChess Benchmark2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C3211M22M33M44M55MSE +/- 491754.88, N = 15SE +/- 166041.67, N = 3SE +/- 515794.89, N = 15SE +/- 410305.51, N = 3513777785010139649974210491941551. Stockfish 16 by the Stockfish developers (see AUTHORS file)

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.17.08Test: Memory Copying2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382K4K6K8K10KSE +/- 44.94, N = 3SE +/- 72.54, N = 3SE +/- 39.56, N = 3SE +/- 25.18, N = 310752.8110720.3110608.9610608.821. (CXX) g++ options: -lm -lapparmor -latomic -lcrypt -ldl -ljpeg -lEGL -lGLESv2 -lgmp -lgbm -lmpfr -lsctp -lz -lrt -lpthread -lc -std=gnu99 -O2 -U_FORTIFY_SOURCE

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-502 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C3848121620SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 317.3417.3316.9216.81

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.16.1Device: CPU - Batch Size: 64 - Model: ResNet-502 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 16GB DDR5-6000 CL30 F5-6000J3038F16G1224364860SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.02, N = 355.1954.3053.5452.27

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 Compile2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G50100150200250SE +/- 0.19, N = 3SE +/- 0.11, N = 3219.27221.14

Time To Compile

2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: The test quit with a non-zero exit status. E: Warning: Protocol buffer compiler (protoc) not found.

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.8Build: defconfig2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C361122334455SE +/- 0.30, N = 3SE +/- 0.30, N = 3SE +/- 0.40, N = 347.7647.7848.26

Build: defconfig

2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: The test quit with a non-zero exit status. E: : internal compiler error: Segmentation fault

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.8Build: allmodconfig2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G130260390520650SE +/- 0.52, N = 3SE +/- 0.15, N = 3589.33590.50

Build: allmodconfig

2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: The test quit with a non-zero exit status. E: gcc: internal compiler error: Segmentation fault signal terminated program as

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: Ninja2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C3670140210280350SE +/- 0.22, N = 3SE +/- 0.12, N = 3SE +/- 0.38, N = 2298.05301.63304.62

Build System: Ninja

2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: The test quit with a non-zero exit status. E: /usr/include/c++/13/bits/vector.tcc:445:7: internal compiler error: Segmentation fault

Timed Node.js Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 21.7.2Time To Compile2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G80160240320400SE +/- 0.24, N = 3SE +/- 0.34, N = 3341.68345.25

Time To Compile

2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38: The test quit with a non-zero exit status.

x265

OpenBenchmarking.orgFrames Per Second, More Is Betterx265Video Input: Bosphorus 4K2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38918273645SE +/- 0.15, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.15, N = 338.3038.2738.0737.091. x265 [info]: HEVC encoder version 3.5+1-f0c1022b6

OpenBenchmarking.orgFrames Per Second, More Is Betterx265Video Input: Bosphorus 1080p2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38306090120150SE +/- 0.27, N = 3SE +/- 0.45, N = 3SE +/- 0.13, N = 3SE +/- 0.23, N = 3134.52134.48134.19131.281. x265 [info]: HEVC encoder version 3.5+1-f0c1022b6

Xcompact3d Incompact3d

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

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 129 Cells Per Direction2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G48121620SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.13, N = 3SE +/- 0.08, N = 312.9713.2613.4014.091. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per Direction2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G1428425670SE +/- 0.14, N = 3SE +/- 0.01, N = 2SE +/- 0.20, N = 3SE +/- 0.09, N = 358.2060.4860.8064.111. (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

Xmrig

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

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.21Variant: GhostRider - Hash Count: 1M2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G8001600240032004000SE +/- 30.71, N = 9SE +/- 36.12, N = 6SE +/- 37.04, N = 3SE +/- 6.72, N = 33699.63676.53638.63616.61. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

XNNPACK

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK 2cd86bModel: FP32MobileNetV22 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C3830060090012001500SE +/- 5.21, N = 3SE +/- 4.98, N = 3SE +/- 8.84, N = 3SE +/- 9.00, N = 312641272127312851. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK 2cd86bModel: FP32MobileNetV3Large2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C3830060090012001500SE +/- 19.19, N = 3SE +/- 9.64, N = 3SE +/- 2.08, N = 3SE +/- 3.18, N = 315161527155315901. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK 2cd86bModel: FP32MobileNetV3Small2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382004006008001000SE +/- 2.31, N = 3SE +/- 2.31, N = 3SE +/- 5.00, N = 3SE +/- 3.18, N = 38128158188361. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK 2cd86bModel: FP16MobileNetV22 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382004006008001000SE +/- 2.31, N = 3SE +/- 3.38, N = 3SE +/- 1.20, N = 3SE +/- 2.31, N = 398599699710181. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK 2cd86bModel: FP16MobileNetV3Large2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C3830060090012001500SE +/- 2.89, N = 3SE +/- 4.58, N = 3SE +/- 3.53, N = 3SE +/- 8.11, N = 312911302130313341. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK 2cd86bModel: FP16MobileNetV3Small2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382004006008001000SE +/- 1.76, N = 3SE +/- 1.20, N = 3SE +/- 1.86, N = 3SE +/- 0.58, N = 37707737767971. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK 2cd86bModel: QU8MobileNetV22 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382004006008001000SE +/- 0.33, N = 3SE +/- 0.88, N = 3SE +/- 2.96, N = 3SE +/- 3.18, N = 37597657657831. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK 2cd86bModel: QU8MobileNetV3Large2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382004006008001000SE +/- 4.91, N = 3SE +/- 1.20, N = 3SE +/- 2.73, N = 3SE +/- 3.21, N = 311151118112411541. (CXX) g++ options: -O3 -lrt -lm

OpenBenchmarking.orgus, Fewer Is BetterXNNPACK 2cd86bModel: QU8MobileNetV3Small2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C382004006008001000SE +/- 0.67, N = 3SE +/- 0.88, N = 3SE +/- 1.33, N = 37567617647861. (CXX) g++ options: -O3 -lrt -lm

Y-Cruncher

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.8.5Pi Digits To Calculate: 1B2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 16GB DDR5-6000 CL30 F5-6000J3038F16G48121620SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 315.4315.6116.45

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.8.5Pi Digits To Calculate: 500M2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C362 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C322 x 16GB DDR5-6000 CL30 F5-6000J3038F16G246810SE +/- 0.023, N = 3SE +/- 0.016, N = 3SE +/- 0.006, N = 37.2367.2477.587

104 Results Shown

7-Zip Compression:
  Compression Rating
  Decompression Rating
Blender:
  BMW27 - CPU-Only
  Junkshop - CPU-Only
  Barbershop - CPU-Only
BRL-CAD
Build2
Embree:
  Pathtracer ISPC - Crown
  Pathtracer ISPC - Asian Dragon
Etcpak
GROMACS
GROMACS
High Performance Conjugate Gradient
Java JMH
LeelaChessZero:
  BLAS
  Eigen
libxsmm:
  128
  32
  64
Llama.cpp
Llamafile
LULESH
LuxCoreRender:
  DLSC - CPU
  Danish Mood - CPU
  Orange Juice - CPU
  LuxCore Benchmark - CPU
  Rainbow Colors and Prism - CPU
MBW:
  Memory Copy - 4096 MiB
  Memory Copy - 8192 MiB
  Memory Copy, Fixed Block Size - 4096 MiB
  Memory Copy, Fixed Block Size - 8192 MiB
Memcached:
  1:10
  1:100
miniBUDE:
  OpenMP - BM1:
    GFInst/s
    Billion Interactions/s
  OpenMP - BM2:
    GFInst/s
    Billion Interactions/s
NAMD:
  ATPase with 327,506 Atoms
  STMV with 1,066,628 Atoms
NAS Parallel Benchmarks:
  BT.C
  CG.C
  EP.C
  EP.D
  FT.C
  IS.D
  LU.C
  MG.C
  SP.B
  SP.C
Numpy Benchmark
OpenFOAM:
  motorBike - Mesh Time
  motorBike - Execution Time
  drivaerFastback, Small Mesh Size - Mesh Time
  drivaerFastback, Small Mesh Size - Execution Time
  drivaerFastback, Medium Mesh Size - Mesh Time
  drivaerFastback, Medium Mesh Size - Execution Time
OpenRadioss:
  Bumper Beam
  Chrysler Neon 1M
  Cell Phone Drop Test
  Bird Strike on Windshield
  Rubber O-Ring Seal Installation
  INIVOL and Fluid Structure Interaction Drop Container
POV-Ray
PyTorch:
  CPU - 1 - ResNet-50
  CPU - 256 - ResNet-50
Quicksilver
RAMspeed SMP:
  Add - Integer
  Copy - Integer
  Scale - Integer
  Triad - Integer
  Average - Integer
simdjson:
  Kostya
  TopTweet
  LargeRand
SPECFEM3D:
  Mount St. Helens
  Layered Halfspace
  Tomographic Model
  Homogeneous Halfspace
  Water-layered Halfspace
Stockfish
Stress-NG
TensorFlow:
  CPU - 1 - ResNet-50
  CPU - 64 - ResNet-50
Timed Gem5 Compilation
Timed Linux Kernel Compilation:
  defconfig
  allmodconfig
Timed LLVM Compilation
Timed Node.js Compilation
x265:
  Bosphorus 4K
  Bosphorus 1080p
Xcompact3d Incompact3d:
  input.i3d 129 Cells Per Direction
  input.i3d 193 Cells Per Direction
Xmrig
XNNPACK:
  FP32MobileNetV2
  FP32MobileNetV3Large
  FP32MobileNetV3Small
  FP16MobileNetV2
  FP16MobileNetV3Large
  FP16MobileNetV3Small
  QU8MobileNetV2
  QU8MobileNetV3Large
  QU8MobileNetV3Small
Y-Cruncher:
  1B
  500M