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 AMD Ryzen 9 9950X DDR5 Memory Performance - Phoronix Test Suite AMD Ryzen 9 9950X DDR5 Memory Performance AMD Ryzen 9 9950X DDR5 memory module benchmarks by Michael Larabel for a future article.
HTML result view exported from: https://openbenchmarking.org/result/2408225-NE-RYZEN999510&grr&sor&rro .
AMD Ryzen 9 9950X DDR5 Memory Performance Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL Compiler File-System Screen Resolution 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 AMD Ryzen 9 9950X 16-Core @ 5.75GHz (16 Cores / 32 Threads) ASUS ROG STRIX X670E-E GAMING WIFI (2204 BIOS) AMD Device 14d8 2 x 16GB DDR5-6000MT/s G Skill F5-6000J3038F16G 2000GB Corsair MP700 PRO AMD Radeon RX 7900 GRE 16GB AMD Navi 31 HDMI/DP DELL U2723QE Intel I225-V + Intel Wi-Fi 6E Ubuntu 24.04 6.10.0-phx (x86_64) GNOME Shell 46.0 X Server + Wayland 4.6 Mesa 24.2~git2406040600.8112d4~oibaf~n (git-8112d44 2024-06-04 noble-oibaf-ppa) (LLVM 17.0.6 DRM 3.57) GCC 13.2.0 ext4 3840x2160 2 x 32GB DDR5-6400MT/s Corsair CMK64GX5M2B6400C32 2 x 16GB DDR5-8000MT/s Corsair CMH32GX5M2X8000C36 2 x 24GB DDR5-8000MT/s Corsair CMP48GX5M2X8000C38 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler 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-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 Processor Details - Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xb40401a Python Details - Python 3.12.3 Security Details - gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected Java Details - 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)
AMD Ryzen 9 9950X DDR5 Memory Performance llamafile: Meta-Llama-3-8B-Instruct.F16 - CPU openradioss: Chrysler Neon 1M openfoam: drivaerFastback, Medium Mesh Size - Mesh Time openfoam: drivaerFastback, Medium Mesh Size - Execution Time xmrig: GhostRider - 1M blender: Barbershop - CPU-Only build-linux-kernel: allmodconfig lczero: Eigen openradioss: Bird Strike on Windshield lczero: BLAS xnnpack: QU8MobileNetV3Small xnnpack: QU8MobileNetV3Large xnnpack: QU8MobileNetV2 xnnpack: FP16MobileNetV3Small xnnpack: FP16MobileNetV3Large xnnpack: FP16MobileNetV2 xnnpack: FP32MobileNetV3Small xnnpack: FP32MobileNetV3Large xnnpack: FP32MobileNetV2 build-llvm: Ninja build-nodejs: Time To Compile openradioss: INIVOL and Fluid Structure Interaction Drop Container brl-cad: VGR Performance Metric stockfish: Chess Benchmark java-jmh: Throughput hpcg: 104 104 104 - 60 build-gem5: Time To Compile openfoam: motorBike - Execution Time openfoam: motorBike - Mesh Time tensorflow: CPU - 64 - ResNet-50 npb: SP.C simdjson: Kostya mbw: Memory Copy, Fixed Block Size - 8192 MiB build2: Time To Compile gromacs: water_GMX50_bare numpy: openradioss: Bumper Beam specfem3d: Layered Halfspace simdjson: LargeRand specfem3d: Water-layered Halfspace simdjson: TopTweet mbw: Memory Copy - 8192 MiB pytorch: CPU - 256 - ResNet-50 libxsmm: 128 minibude: OpenMP - BM2 minibude: OpenMP - BM2 memcached: 1:10 memcached: 1:100 blender: Junkshop - CPU-Only luxcorerender: Orange Juice - CPU luxcorerender: Danish Mood - CPU luxcorerender: LuxCore Benchmark - CPU llama-cpp: Meta-Llama-3-8B-Instruct-Q8_0.gguf luxcorerender: DLSC - CPU incompact3d: input.i3d 193 Cells Per Direction openradioss: Rubber O-Ring Seal Installation gromacs: MPI CPU - water_GMX50_bare openfoam: drivaerFastback, Small Mesh Size - Mesh Time npb: BT.C namd: STMV with 1,066,628 Atoms specfem3d: Tomographic Model openfoam: drivaerFastback, Small Mesh Size - Execution Time blender: BMW27 - CPU-Only openradioss: Cell Phone Drop Test libxsmm: 64 libxsmm: 32 build-linux-kernel: defconfig quicksilver: CORAL2 P1 mbw: Memory Copy - 4096 MiB namd: ATPase with 327,506 Atoms npb: EP.D npb: LU.C ramspeed: Copy - Integer specfem3d: Homogeneous Halfspace stress-ng: Memory Copying ramspeed: Average - Integer specfem3d: Mount St. Helens ramspeed: Add - Integer npb: IS.D ramspeed: Triad - Integer ramspeed: Scale - Integer mbw: Memory Copy, Fixed Block Size - 4096 MiB etcpak: Multi-Threaded - ETC2 embree: Pathtracer ISPC - Crown pytorch: CPU - 1 - ResNet-50 embree: Pathtracer ISPC - Asian Dragon povray: Trace Time compress-7zip: Compression Rating y-cruncher: 1B compress-7zip: Decompression Rating npb: SP.B x265: Bosphorus 4K npb: FT.C incompact3d: input.i3d 129 Cells Per Direction npb: CG.C minibude: OpenMP - BM1 minibude: OpenMP - BM1 luxcorerender: Rainbow Colors and Prism - CPU tensorflow: CPU - 1 - ResNet-50 y-cruncher: 500M npb: MG.C lulesh: x265: Bosphorus 1080p npb: EP.C 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 4.68 705.55 181.56905 2096.5664 3616.6 448.96 590.495 220 128.76 226 756 1118 759 770 1291 985 812 1516 1264 301.629 345.246 247.06 498966 51377778 91554717298.426 8.93560 221.139 58.5695 84.9949 52.27 14528.26 7.71 19467.394 76.003 1.927 1068.36 76.90 73.138797084 2.12 72.197260512 13.59 22247.784 57.54 472.0 77.385 1934.613 5915791.34 7507615.73 62.09 8.64 4.62 5.15 8.36 5.40 64.1130218 53.16 3.141 22.579849 55287.37 0.98084 23.359092920 157.14898 46.38 44.90 237.1 121.0 47.755 25676667 22774.947 3.41007 3823.80 59920.96 65797.05 30.677918457 10752.81 65503.93 25.690986644 61079.93 1505.47 62428.70 69141.08 19534.159 741.784 36.5532 84.89 41.4914 16.101 195585 16.452 158133 21328.28 38.27 27106.18 14.0905552 11144.88 76.885 1922.125 19.99 16.92 7.587 24236.21 9945.5309 134.52 3678.46 668.16 177.7894 1994.747 3638.6 450.71 589.331 236 128.01 244 761 1115 765 776 1303 996 815 1527 1273 298.049 341.679 239.22 494437 49194155 92045826799.54 8.79737 219.268 57.2452 84.9802 54.30 15610.30 7.63 20850.883 77.006 1.996 1070.61 77.18 69.439387299 2.11 69.561642467 13.38 22294.062 58.89 505.9 77.246 1931.141 5898565.34 7647187.48 62.21 8.64 4.69 5.14 8.32 5.42 60.7992439 53.23 3.228 22.821389 58570.77 1.01602 23.000501341 148.75845 46.56 43.42 253.3 128.0 47.783 25660000 22510.464 3.60464 3714.66 62084.72 70747.00 29.818431268 10720.31 68588.84 24.956549928 63967.70 1584.93 65845.21 71754.58 20817.569 739.736 36.1099 85.52 40.9447 16.133 203065 15.426 157512 22908.58 38.30 28891.86 13.4028053 11374.51 77.076 1926.891 20.08 17.34 7.247 25622.93 10235.722 134.48 3731.97 573.33 174.12572 1909.6655 3699.6 459.12 247 128.32 264 764 1124 765 773 1302 997 818 1553 1272 304.621 242.00 486954 49974210 89012581592.653 9.51601 55.5738 88.0807 55.19 16097.00 7.57 20349.414 77.923 2.009 1046.91 77.45 69.817422311 2.10 68.920935651 13.24 22817.900 58.49 508.9 76.198 1904.953 5874747.73 7498825.19 63.41 8.55 4.65 5.04 9.00 5.36 58.1984596 53.66 3.298 23.143631 60124.08 1.02426 23.287517388 145.88323 47.72 43.25 258.6 132.2 48.258 25666667 22539.720 3.57755 3739.16 64031.31 70345.53 30.542984487 10608.96 69491.42 25.194653630 65113.52 1635.34 66963.03 73724.88 20923.989 734.533 36.2546 86.79 41.7238 16.395 200810 15.606 157005 23254.40 38.07 29713.55 12.9742870 11908.47 76.122 1903.040 19.84 17.33 7.236 26596.46 10578.516 134.19 3741.86 3676.5 459.19 227 124.37 786 1154 783 797 1334 1018 836 1590 1285 489784 50101396 90340365706.792 9.16590 85.4745 53.54 15531.76 7.50 20124.943 80.104 1064.17 77.81 72.4383489 2.11 72.079476658 12.88 21444.002 56.19 491.1 75.930 1898.258 5694123.52 7350932.88 64.08 8.46 4.60 5.05 8.66 5.34 60.4805832 54.27 58570.03 0.99852 23.869453181 47.51 249.7 127.0 25136667 22093.985 3.50958 3712.45 62375.43 66545.24 31.981715677 10608.82 66822.49 25.453319060 62771.00 1586.59 64832.86 70445.74 19747.584 731.229 35.1828 81.89 40.0696 16.319 22436.81 37.09 28743.49 13.2595733 11662.44 75.992 1899.804 19.33 16.81 25715.34 9832.6385 131.28 3722.56 OpenBenchmarking.org
Llamafile Test: Meta-Llama-3-8B-Instruct.F16 - Acceleration: CPU OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.8.6 Test: Meta-Llama-3-8B-Instruct.F16 - Acceleration: CPU 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 1.053 2.106 3.159 4.212 5.265 4.68
OpenRadioss Model: Chrysler Neon 1M OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Chrysler Neon 1M 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 150 300 450 600 750 SE +/- 0.28, N = 3 SE +/- 1.02, N = 3 SE +/- 49.36, N = 9 705.55 668.16 573.33
OpenFOAM Input: drivaerFastback, Medium Mesh Size - Mesh Time OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Mesh Time 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 40 80 120 160 200 181.57 177.79 174.13 1. (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
OpenFOAM Input: drivaerFastback, Medium Mesh Size - Execution Time OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Execution Time 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 500 1000 1500 2000 2500 2096.57 1994.75 1909.67 1. (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
Xmrig Variant: GhostRider - Hash Count: 1M OpenBenchmarking.org H/s, More Is Better Xmrig 6.21 Variant: GhostRider - Hash Count: 1M 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 800 1600 2400 3200 4000 SE +/- 6.72, N = 3 SE +/- 37.04, N = 3 SE +/- 36.12, N = 6 SE +/- 30.71, N = 9 3616.6 3638.6 3676.5 3699.6 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
Blender Blend File: Barbershop - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.2 Blend File: Barbershop - Compute: CPU-Only 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 100 200 300 400 500 SE +/- 0.20, N = 3 SE +/- 0.27, N = 3 SE +/- 0.26, N = 3 SE +/- 0.20, N = 3 459.19 459.12 450.71 448.96
Timed Linux Kernel Compilation Build: allmodconfig OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.8 Build: allmodconfig 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 130 260 390 520 650 SE +/- 0.15, N = 3 SE +/- 0.52, N = 3 590.50 589.33
LeelaChessZero Backend: Eigen OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.31.1 Backend: Eigen 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 50 100 150 200 250 SE +/- 1.45, N = 3 SE +/- 2.85, N = 3 SE +/- 1.15, N = 3 220 227 236 247 1. (CXX) g++ options: -flto -pthread
OpenRadioss Model: Bird Strike on Windshield OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Bird Strike on Windshield 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 30 60 90 120 150 SE +/- 0.28, N = 3 SE +/- 0.28, N = 3 SE +/- 0.35, N = 3 SE +/- 5.81, N = 15 128.76 128.32 128.01 124.37
LeelaChessZero Backend: BLAS OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.31.1 Backend: BLAS 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 60 120 180 240 300 SE +/- 2.65, N = 3 SE +/- 2.91, N = 3 SE +/- 3.79, N = 3 226 244 264 1. (CXX) g++ options: -flto -pthread
XNNPACK Model: QU8MobileNetV3Small OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: QU8MobileNetV3Small 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 200 400 600 800 1000 SE +/- 1.33, N = 3 SE +/- 0.88, N = 3 SE +/- 0.67, N = 3 786 764 761 756 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: QU8MobileNetV3Large OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: QU8MobileNetV3Large 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 200 400 600 800 1000 SE +/- 3.21, N = 3 SE +/- 2.73, N = 3 SE +/- 1.20, N = 3 SE +/- 4.91, N = 3 1154 1124 1118 1115 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: QU8MobileNetV2 OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: QU8MobileNetV2 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 200 400 600 800 1000 SE +/- 3.18, N = 3 SE +/- 2.96, N = 3 SE +/- 0.88, N = 3 SE +/- 0.33, N = 3 783 765 765 759 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP16MobileNetV3Small OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP16MobileNetV3Small 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 200 400 600 800 1000 SE +/- 0.58, N = 3 SE +/- 1.86, N = 3 SE +/- 1.20, N = 3 SE +/- 1.76, N = 3 797 776 773 770 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP16MobileNetV3Large OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP16MobileNetV3Large 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 300 600 900 1200 1500 SE +/- 8.11, N = 3 SE +/- 3.53, N = 3 SE +/- 4.58, N = 3 SE +/- 2.89, N = 3 1334 1303 1302 1291 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP16MobileNetV2 OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP16MobileNetV2 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 200 400 600 800 1000 SE +/- 2.31, N = 3 SE +/- 1.20, N = 3 SE +/- 3.38, N = 3 SE +/- 2.31, N = 3 1018 997 996 985 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP32MobileNetV3Small OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP32MobileNetV3Small 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 200 400 600 800 1000 SE +/- 3.18, N = 3 SE +/- 5.00, N = 3 SE +/- 2.31, N = 3 SE +/- 2.31, N = 3 836 818 815 812 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP32MobileNetV3Large OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP32MobileNetV3Large 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 300 600 900 1200 1500 SE +/- 3.18, N = 3 SE +/- 2.08, N = 3 SE +/- 9.64, N = 3 SE +/- 19.19, N = 3 1590 1553 1527 1516 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP32MobileNetV2 OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP32MobileNetV2 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 300 600 900 1200 1500 SE +/- 9.00, N = 3 SE +/- 8.84, N = 3 SE +/- 4.98, N = 3 SE +/- 5.21, N = 3 1285 1273 1272 1264 1. (CXX) g++ options: -O3 -lrt -lm
Timed LLVM Compilation Build System: Ninja OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 16.0 Build System: Ninja 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 70 140 210 280 350 SE +/- 0.38, N = 2 SE +/- 0.12, N = 3 SE +/- 0.22, N = 3 304.62 301.63 298.05
Timed Node.js Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Node.js Compilation 21.7.2 Time To Compile 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 80 160 240 320 400 SE +/- 0.34, N = 3 SE +/- 0.24, N = 3 345.25 341.68
OpenRadioss Model: INIVOL and Fluid Structure Interaction Drop Container OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: INIVOL and Fluid Structure Interaction Drop Container 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 50 100 150 200 250 SE +/- 0.74, N = 3 SE +/- 0.76, N = 3 SE +/- 0.99, N = 3 247.06 242.00 239.22
BRL-CAD VGR Performance Metric OpenBenchmarking.org VGR Performance Metric, More Is Better BRL-CAD 7.38.2 VGR Performance Metric 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 110K 220K 330K 440K 550K 486954 489784 494437 498966 1. (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
Stockfish Chess Benchmark OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish Chess Benchmark 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 11M 22M 33M 44M 55M SE +/- 410305.51, N = 3 SE +/- 515794.89, N = 15 SE +/- 166041.67, N = 3 SE +/- 491754.88, N = 15 49194155 49974210 50101396 51377778 1. Stockfish 16 by the Stockfish developers (see AUTHORS file)
Java JMH Throughput OpenBenchmarking.org Ops/s, More Is Better Java JMH Throughput 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 20000M 40000M 60000M 80000M 100000M 89012581592.65 90340365706.79 91554717298.43 92045826799.54
High Performance Conjugate Gradient X Y Z: 104 104 104 - RT: 60 OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.1 X Y Z: 104 104 104 - RT: 60 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 3 6 9 12 15 SE +/- 0.00125, N = 3 SE +/- 0.00142, N = 3 SE +/- 0.00584, N = 3 SE +/- 0.00741, N = 3 8.79737 8.93560 9.16590 9.51601 1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi
Timed Gem5 Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Gem5 Compilation 23.0.1 Time To Compile 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 50 100 150 200 250 SE +/- 0.11, N = 3 SE +/- 0.19, N = 3 221.14 219.27
OpenFOAM Input: motorBike - Execution Time OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: motorBike - Execution Time 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 13 26 39 52 65 58.57 57.25 55.57 1. (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
OpenFOAM Input: motorBike - Mesh Time OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: motorBike - Mesh Time 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 20 40 60 80 100 88.08 85.47 84.99 84.98 1. (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
TensorFlow Device: CPU - Batch Size: 64 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.16.1 Device: CPU - Batch Size: 64 - Model: ResNet-50 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 12 24 36 48 60 SE +/- 0.02, N = 3 SE +/- 0.09, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 52.27 53.54 54.30 55.19
NAS Parallel Benchmarks Test / Class: SP.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.C 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 3K 6K 9K 12K 15K SE +/- 18.27, N = 3 SE +/- 42.19, N = 3 SE +/- 16.93, N = 3 SE +/- 15.11, N = 3 14528.26 15531.76 15610.30 16097.00 1. (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
simdjson Throughput Test: Kostya OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: Kostya 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 4 6 8 10 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 7.50 7.57 7.63 7.71 1. (CXX) g++ options: -O3 -lrt
MBW Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 4K 8K 12K 16K 20K SE +/- 10.56, N = 3 SE +/- 239.88, N = 3 SE +/- 144.90, N = 15 SE +/- 58.10, N = 3 19467.39 20124.94 20349.41 20850.88 1. (CC) gcc options: -O3 -march=native
Build2 Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Build2 0.17 Time To Compile 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 20 40 60 80 100 SE +/- 0.99, N = 2 SE +/- 0.40, N = 3 SE +/- 0.52, N = 3 SE +/- 0.16, N = 3 80.10 77.92 77.01 76.00
GROMACS Input: water_GMX50_bare OpenBenchmarking.org Ns Per Day, More Is Better GROMACS Input: water_GMX50_bare 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 0.452 0.904 1.356 1.808 2.26 SE +/- 0.002, N = 3 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 1.927 1.996 2.009 1. GROMACS version: 2023.3-Ubuntu_2023.3_1ubuntu3
Numpy Benchmark OpenBenchmarking.org Score, More Is Better Numpy Benchmark 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 200 400 600 800 1000 SE +/- 1.37, N = 3 SE +/- 4.91, N = 3 SE +/- 8.61, N = 3 SE +/- 2.01, N = 3 1046.91 1064.17 1068.36 1070.61
OpenRadioss Model: Bumper Beam OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Bumper Beam 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 20 40 60 80 100 SE +/- 0.02, N = 3 SE +/- 0.12, N = 3 SE +/- 0.14, N = 3 SE +/- 0.07, N = 3 77.81 77.45 77.18 76.90
SPECFEM3D Model: Layered Halfspace OpenBenchmarking.org Seconds, Fewer Is Better SPECFEM3D 4.1.1 Model: Layered Halfspace 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 16 32 48 64 80 SE +/- 0.31, N = 3 SE +/- 0.59, N = 3 SE +/- 0.59, N = 3 73.14 72.44 69.82 69.44 1. (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
simdjson Throughput Test: LargeRandom OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: LargeRandom 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 0.477 0.954 1.431 1.908 2.385 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 2.10 2.11 2.11 2.12 1. (CXX) g++ options: -O3 -lrt
SPECFEM3D Model: Water-layered Halfspace OpenBenchmarking.org Seconds, Fewer Is Better SPECFEM3D 4.1.1 Model: Water-layered Halfspace 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 16 32 48 64 80 SE +/- 0.34, N = 3 SE +/- 0.16, N = 2 SE +/- 0.34, N = 3 SE +/- 0.24, N = 3 72.20 72.08 69.56 68.92 1. (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
simdjson Throughput Test: TopTweet OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: TopTweet 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 3 6 9 12 15 SE +/- 0.14, N = 3 SE +/- 0.15, N = 3 SE +/- 0.17, N = 3 SE +/- 0.02, N = 3 12.88 13.24 13.38 13.59 1. (CXX) g++ options: -O3 -lrt
MBW Test: Memory Copy - Array Size: 8192 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 8192 MiB 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 5K 10K 15K 20K 25K SE +/- 56.11, N = 3 SE +/- 49.10, N = 3 SE +/- 188.50, N = 12 SE +/- 204.03, N = 3 21444.00 22247.78 22294.06 22817.90 1. (CC) gcc options: -O3 -march=native
PyTorch Device: CPU - Batch Size: 256 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 256 - Model: ResNet-50 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 13 26 39 52 65 SE +/- 0.25, N = 3 SE +/- 0.34, N = 3 SE +/- 0.36, N = 3 SE +/- 0.40, N = 15 56.19 57.54 58.49 58.89 MIN: 39.35 / MAX: 57.3 MIN: 53.14 / MAX: 58.34 MIN: 34.6 / MAX: 60.01 MIN: 34.79 / MAX: 62.62
libxsmm M N K: 128 OpenBenchmarking.org GFLOPS/s, More Is Better libxsmm 2-1.17-3645 M N K: 128 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 110 220 330 440 550 SE +/- 0.38, N = 3 SE +/- 0.33, N = 3 SE +/- 0.10, N = 3 SE +/- 0.09, N = 3 472.0 491.1 505.9 508.9 1. (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
miniBUDE Implementation: OpenMP - Input Deck: BM2 OpenBenchmarking.org Billion Interactions/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 20 40 60 80 100 SE +/- 0.09, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 75.93 76.20 77.25 77.39 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
miniBUDE Implementation: OpenMP - Input Deck: BM2 OpenBenchmarking.org GFInst/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 400 800 1200 1600 2000 SE +/- 2.32, N = 3 SE +/- 1.03, N = 3 SE +/- 0.32, N = 3 SE +/- 0.64, N = 3 1898.26 1904.95 1931.14 1934.61 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
Memcached Set To Get Ratio: 1:10 OpenBenchmarking.org Ops/sec, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:10 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 1.3M 2.6M 3.9M 5.2M 6.5M SE +/- 4372.40, N = 3 SE +/- 11157.27, N = 3 SE +/- 11599.22, N = 3 SE +/- 15831.56, N = 3 5694123.52 5874747.73 5898565.34 5915791.34 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
Memcached Set To Get Ratio: 1:100 OpenBenchmarking.org Ops/sec, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:100 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 1.6M 3.2M 4.8M 6.4M 8M SE +/- 71150.32, N = 3 SE +/- 43634.73, N = 3 SE +/- 1150.91, N = 3 SE +/- 87376.76, N = 3 7350932.88 7498825.19 7507615.73 7647187.48 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
Blender Blend File: Junkshop - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.2 Blend File: Junkshop - Compute: CPU-Only 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 14 28 42 56 70 SE +/- 0.17, N = 3 SE +/- 0.08, N = 3 SE +/- 0.03, N = 3 SE +/- 0.12, N = 3 64.08 63.41 62.21 62.09
LuxCoreRender Scene: Orange Juice - Acceleration: CPU OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Orange Juice - Acceleration: CPU 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 8.46 8.55 8.64 8.64 MIN: 7.49 / MAX: 9.04 MIN: 7.48 / MAX: 9.15 MIN: 7.64 / MAX: 9.25 MIN: 7.67 / MAX: 9.25
LuxCoreRender Scene: Danish Mood - Acceleration: CPU OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Danish Mood - Acceleration: CPU 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 1.0553 2.1106 3.1659 4.2212 5.2765 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 4.60 4.62 4.65 4.69 MIN: 2.13 / MAX: 5.16 MIN: 2.07 / MAX: 5.22 MIN: 2.02 / MAX: 5.24 MIN: 2.15 / MAX: 5.26
LuxCoreRender Scene: LuxCore Benchmark - Acceleration: CPU OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: LuxCore Benchmark - Acceleration: CPU 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 1.1588 2.3176 3.4764 4.6352 5.794 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 5.04 5.05 5.14 5.15 MIN: 2.28 / MAX: 5.66 MIN: 2.28 / MAX: 5.66 MIN: 2.39 / MAX: 5.75 MIN: 2.38 / MAX: 5.77
Llama.cpp Model: Meta-Llama-3-8B-Instruct-Q8_0.gguf OpenBenchmarking.org Tokens Per Second, More Is Better Llama.cpp b3067 Model: Meta-Llama-3-8B-Instruct-Q8_0.gguf 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 8.32 8.36 8.66 9.00 1. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -march=native -mtune=native -lopenblas
LuxCoreRender Scene: DLSC - Acceleration: CPU OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: DLSC - Acceleration: CPU 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 1.2195 2.439 3.6585 4.878 6.0975 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 5.34 5.36 5.40 5.42 MIN: 5.23 / MAX: 5.69 MIN: 5.25 / MAX: 5.7 MIN: 5.27 / MAX: 5.72 MIN: 5.31 / MAX: 5.77
Xcompact3d Incompact3d Input: input.i3d 193 Cells Per Direction OpenBenchmarking.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: input.i3d 193 Cells Per Direction 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 14 28 42 56 70 SE +/- 0.09, N = 3 SE +/- 0.20, N = 3 SE +/- 0.01, N = 2 SE +/- 0.14, N = 3 64.11 60.80 60.48 58.20 1. (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
OpenRadioss Model: Rubber O-Ring Seal Installation OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Rubber O-Ring Seal Installation 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 12 24 36 48 60 SE +/- 0.06, N = 3 SE +/- 0.16, N = 3 SE +/- 0.15, N = 3 SE +/- 0.10, N = 3 54.27 53.66 53.23 53.16
GROMACS Implementation: MPI CPU - Input: water_GMX50_bare OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2024 Implementation: MPI CPU - Input: water_GMX50_bare 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 0.7421 1.4842 2.2263 2.9684 3.7105 SE +/- 0.001, N = 3 SE +/- 0.004, N = 3 SE +/- 0.002, N = 3 3.141 3.228 3.298 1. (CXX) g++ options: -O3 -lm
OpenFOAM Input: drivaerFastback, Small Mesh Size - Mesh Time OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Mesh Time 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 6 12 18 24 30 23.14 22.82 22.58 1. (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
NAS Parallel Benchmarks Test / Class: BT.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: BT.C 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 13K 26K 39K 52K 65K SE +/- 54.79, N = 3 SE +/- 74.71, N = 3 SE +/- 8.10, N = 3 SE +/- 85.26, N = 3 55287.37 58570.03 58570.77 60124.08 1. (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
NAMD Input: STMV with 1,066,628 Atoms OpenBenchmarking.org ns/day, More Is Better NAMD 3.0b6 Input: STMV with 1,066,628 Atoms 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 0.2305 0.461 0.6915 0.922 1.1525 SE +/- 0.00027, N = 3 SE +/- 0.00137, N = 2 SE +/- 0.00200, N = 3 SE +/- 0.00057, N = 3 0.98084 0.99852 1.01602 1.02426
SPECFEM3D Model: Tomographic Model OpenBenchmarking.org Seconds, Fewer Is Better SPECFEM3D 4.1.1 Model: Tomographic Model 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 6 12 18 24 30 SE +/- 0.16, N = 15 SE +/- 0.16, N = 3 SE +/- 0.14, N = 3 SE +/- 0.20, N = 3 23.87 23.36 23.29 23.00 1. (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
OpenFOAM Input: drivaerFastback, Small Mesh Size - Execution Time OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Execution Time 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 30 60 90 120 150 157.15 148.76 145.88 1. (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
Blender Blend File: BMW27 - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.2 Blend File: BMW27 - Compute: CPU-Only 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 11 22 33 44 55 SE +/- 0.10, N = 3 SE +/- 0.02, N = 3 SE +/- 0.09, N = 3 SE +/- 0.01, N = 3 47.72 47.51 46.56 46.38
OpenRadioss Model: Cell Phone Drop Test OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Cell Phone Drop Test 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 10 20 30 40 50 SE +/- 0.18, N = 3 SE +/- 0.29, N = 3 SE +/- 0.15, N = 3 44.90 43.42 43.25
libxsmm M N K: 64 OpenBenchmarking.org GFLOPS/s, More Is Better libxsmm 2-1.17-3645 M N K: 64 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 60 120 180 240 300 SE +/- 0.20, N = 3 SE +/- 0.53, N = 3 SE +/- 0.03, N = 3 SE +/- 0.37, N = 3 237.1 249.7 253.3 258.6 1. (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
libxsmm M N K: 32 OpenBenchmarking.org GFLOPS/s, More Is Better libxsmm 2-1.17-3645 M N K: 32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 30 60 90 120 150 SE +/- 0.07, N = 3 SE +/- 0.26, N = 3 SE +/- 0.07, N = 3 SE +/- 0.19, N = 3 121.0 127.0 128.0 132.2 1. (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
Timed Linux Kernel Compilation Build: defconfig OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.8 Build: defconfig 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 11 22 33 44 55 SE +/- 0.40, N = 3 SE +/- 0.30, N = 3 SE +/- 0.30, N = 3 48.26 47.78 47.76
Quicksilver Input: CORAL2 P1 OpenBenchmarking.org Figure Of Merit, More Is Better Quicksilver 20230818 Input: CORAL2 P1 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 5M 10M 15M 20M 25M SE +/- 16666.67, N = 3 SE +/- 41633.32, N = 3 SE +/- 92074.85, N = 3 SE +/- 21858.13, N = 3 25136667 25660000 25666667 25676667 1. (CXX) g++ options: -fopenmp -O3 -march=native
MBW Test: Memory Copy - Array Size: 4096 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 4096 MiB 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 5K 10K 15K 20K 25K SE +/- 227.86, N = 15 SE +/- 265.89, N = 3 SE +/- 31.03, N = 3 SE +/- 72.38, N = 3 22093.99 22510.46 22539.72 22774.95 1. (CC) gcc options: -O3 -march=native
NAMD Input: ATPase with 327,506 Atoms OpenBenchmarking.org ns/day, More Is Better NAMD 3.0b6 Input: ATPase with 327,506 Atoms 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 0.811 1.622 2.433 3.244 4.055 SE +/- 0.02586, N = 3 SE +/- 0.02740, N = 9 SE +/- 0.02618, N = 15 SE +/- 0.03856, N = 3 3.41007 3.50958 3.57755 3.60464
NAS Parallel Benchmarks Test / Class: EP.D OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 800 1600 2400 3200 4000 SE +/- 51.16, N = 3 SE +/- 36.74, N = 3 SE +/- 40.74, N = 3 SE +/- 2.85, N = 3 3712.45 3714.66 3739.16 3823.80 1. (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
NAS Parallel Benchmarks Test / Class: LU.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 14K 28K 42K 56K 70K SE +/- 150.93, N = 3 SE +/- 130.21, N = 3 SE +/- 210.02, N = 3 SE +/- 72.30, N = 3 59920.96 62084.72 62375.43 64031.31 1. (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
RAMspeed SMP Type: Copy - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Copy - Benchmark: Integer 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 15K 30K 45K 60K 75K SE +/- 820.93, N = 3 SE +/- 299.96, N = 3 SE +/- 762.95, N = 5 SE +/- 453.50, N = 3 65797.05 66545.24 70345.53 70747.00 1. (CC) gcc options: -O3 -march=native
SPECFEM3D Model: Homogeneous Halfspace OpenBenchmarking.org Seconds, Fewer Is Better SPECFEM3D 4.1.1 Model: Homogeneous Halfspace 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 7 14 21 28 35 SE +/- 0.28, N = 3 SE +/- 0.38, N = 3 SE +/- 0.39, N = 3 31.98 30.68 30.54 29.82 1. (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
Stress-NG Test: Memory Copying OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.17.08 Test: Memory Copying 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2K 4K 6K 8K 10K SE +/- 25.18, N = 3 SE +/- 39.56, N = 3 SE +/- 72.54, N = 3 SE +/- 44.94, N = 3 10608.82 10608.96 10720.31 10752.81 1. (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
RAMspeed SMP Type: Average - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Average - Benchmark: Integer 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 15K 30K 45K 60K 75K SE +/- 119.74, N = 3 SE +/- 805.54, N = 4 SE +/- 204.48, N = 3 SE +/- 315.00, N = 3 65503.93 66822.49 68588.84 69491.42 1. (CC) gcc options: -O3 -march=native
SPECFEM3D Model: Mount St. Helens OpenBenchmarking.org Seconds, Fewer Is Better SPECFEM3D 4.1.1 Model: Mount St. Helens 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 6 12 18 24 30 SE +/- 0.17, N = 3 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 SE +/- 0.12, N = 3 25.69 25.45 25.19 24.96 1. (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
RAMspeed SMP Type: Add - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Add - Benchmark: Integer 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 14K 28K 42K 56K 70K SE +/- 426.98, N = 3 SE +/- 64.38, N = 3 SE +/- 382.92, N = 3 SE +/- 108.61, N = 3 61079.93 62771.00 63967.70 65113.52 1. (CC) gcc options: -O3 -march=native
NAS Parallel Benchmarks Test / Class: IS.D OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: IS.D 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 400 800 1200 1600 2000 SE +/- 5.33, N = 3 SE +/- 2.77, N = 3 SE +/- 5.23, N = 3 SE +/- 4.49, N = 3 1505.47 1584.93 1586.59 1635.34 1. (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
RAMspeed SMP Type: Triad - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Triad - Benchmark: Integer 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 14K 28K 42K 56K 70K SE +/- 156.92, N = 3 SE +/- 130.09, N = 3 SE +/- 33.42, N = 3 SE +/- 439.59, N = 3 62428.70 64832.86 65845.21 66963.03 1. (CC) gcc options: -O3 -march=native
RAMspeed SMP Type: Scale - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Scale - Benchmark: Integer 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 16K 32K 48K 64K 80K SE +/- 220.54, N = 3 SE +/- 513.22, N = 3 SE +/- 266.14, N = 3 SE +/- 364.59, N = 3 69141.08 70445.74 71754.58 73724.88 1. (CC) gcc options: -O3 -march=native
MBW Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiB 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 4K 8K 12K 16K 20K SE +/- 24.77, N = 3 SE +/- 36.42, N = 3 SE +/- 32.81, N = 3 SE +/- 211.85, N = 3 19534.16 19747.58 20817.57 20923.99 1. (CC) gcc options: -O3 -march=native
Etcpak Benchmark: Multi-Threaded - Configuration: ETC2 OpenBenchmarking.org Mpx/s, More Is Better Etcpak 2.0 Benchmark: Multi-Threaded - Configuration: ETC2 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 160 320 480 640 800 SE +/- 0.82, N = 3 SE +/- 1.36, N = 3 SE +/- 0.38, N = 3 SE +/- 1.25, N = 3 731.23 734.53 739.74 741.78 1. (CXX) g++ options: -flto -pthread
Embree Binary: Pathtracer ISPC - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Crown 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 8 16 24 32 40 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 35.18 36.11 36.25 36.55 MIN: 34.83 / MAX: 36.01 MIN: 35.69 / MAX: 36.98 MIN: 35.89 / MAX: 36.94 MIN: 36.13 / MAX: 37.34
PyTorch Device: CPU - Batch Size: 1 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 1 - Model: ResNet-50 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 20 40 60 80 100 SE +/- 0.82, N = 5 SE +/- 0.27, N = 3 SE +/- 0.96, N = 4 SE +/- 0.63, N = 3 81.89 84.89 85.52 86.79 MIN: 67.48 / MAX: 84.5 MIN: 77.17 / MAX: 85.87 MIN: 68.54 / MAX: 88.53 MIN: 78.82 / MAX: 88.23
Embree Binary: Pathtracer ISPC - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 10 20 30 40 50 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 40.07 40.94 41.49 41.72 MIN: 39.86 / MAX: 40.55 MIN: 40.73 / MAX: 41.6 MIN: 41.24 / MAX: 42.02 MIN: 41.48 / MAX: 42.26
POV-Ray Trace Time OpenBenchmarking.org Seconds, Fewer Is Better POV-Ray Trace Time 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 16.40 16.32 16.13 16.10 1. POV-Ray 3.7.0.10.unofficial
7-Zip Compression Test: Compression Rating OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression Test: Compression Rating 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 40K 80K 120K 160K 200K SE +/- 322.74, N = 3 SE +/- 94.03, N = 3 SE +/- 467.61, N = 3 195585 200810 203065 1. 7-Zip 23.01 (x64) : Copyright (c) 1999-2023 Igor Pavlov : 2023-06-20
Y-Cruncher Pi Digits To Calculate: 1B OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.8.5 Pi Digits To Calculate: 1B 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 16.45 15.61 15.43
7-Zip Compression Test: Decompression Rating OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression Test: Decompression Rating 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 30K 60K 90K 120K 150K SE +/- 33.91, N = 3 SE +/- 69.90, N = 3 SE +/- 14.15, N = 3 157005 157512 158133 1. 7-Zip 23.01 (x64) : Copyright (c) 1999-2023 Igor Pavlov : 2023-06-20
NAS Parallel Benchmarks Test / Class: SP.B OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.B 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 5K 10K 15K 20K 25K SE +/- 21.50, N = 3 SE +/- 40.85, N = 3 SE +/- 26.91, N = 3 SE +/- 10.38, N = 3 21328.28 22436.81 22908.58 23254.40 1. (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
x265 Video Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better x265 Video Input: Bosphorus 4K 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 9 18 27 36 45 SE +/- 0.15, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.15, N = 3 37.09 38.07 38.27 38.30 1. x265 [info]: HEVC encoder version 3.5+1-f0c1022b6
NAS Parallel Benchmarks Test / Class: FT.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: FT.C 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 6K 12K 18K 24K 30K SE +/- 141.73, N = 3 SE +/- 113.15, N = 3 SE +/- 30.67, N = 3 SE +/- 95.45, N = 3 27106.18 28743.49 28891.86 29713.55 1. (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 Input: input.i3d 129 Cells Per Direction OpenBenchmarking.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: input.i3d 129 Cells Per Direction 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 4 8 12 16 20 SE +/- 0.08, N = 3 SE +/- 0.13, N = 3 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 14.09 13.40 13.26 12.97 1. (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
NAS Parallel Benchmarks Test / Class: CG.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: CG.C 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 3K 6K 9K 12K 15K SE +/- 21.24, N = 3 SE +/- 36.86, N = 3 SE +/- 41.76, N = 3 SE +/- 10.26, N = 3 11144.88 11374.51 11662.44 11908.47 1. (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
miniBUDE Implementation: OpenMP - Input Deck: BM1 OpenBenchmarking.org Billion Interactions/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM1 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 20 40 60 80 100 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.27, N = 3 SE +/- 0.02, N = 3 75.99 76.12 76.89 77.08 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
miniBUDE Implementation: OpenMP - Input Deck: BM1 OpenBenchmarking.org GFInst/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM1 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 400 800 1200 1600 2000 SE +/- 0.87, N = 3 SE +/- 0.31, N = 3 SE +/- 6.84, N = 3 SE +/- 0.54, N = 3 1899.80 1903.04 1922.13 1926.89 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
LuxCoreRender Scene: Rainbow Colors and Prism - Acceleration: CPU OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Rainbow Colors and Prism - Acceleration: CPU 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 5 10 15 20 25 SE +/- 0.13, N = 3 SE +/- 0.08, N = 3 SE +/- 0.07, N = 3 SE +/- 0.15, N = 3 19.33 19.84 19.99 20.08 MIN: 17.37 / MAX: 19.62 MIN: 17.86 / MAX: 20.21 MIN: 18.05 / MAX: 20.42 MIN: 17.96 / MAX: 20.55
TensorFlow Device: CPU - Batch Size: 1 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.16.1 Device: CPU - Batch Size: 1 - Model: ResNet-50 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 16.81 16.92 17.33 17.34
Y-Cruncher Pi Digits To Calculate: 500M OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.8.5 Pi Digits To Calculate: 500M 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 4 6 8 10 SE +/- 0.006, N = 3 SE +/- 0.016, N = 3 SE +/- 0.023, N = 3 7.587 7.247 7.236
NAS Parallel Benchmarks Test / Class: MG.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: MG.C 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 6K 12K 18K 24K 30K SE +/- 6.88, N = 3 SE +/- 20.77, N = 3 SE +/- 10.24, N = 3 SE +/- 34.50, N = 3 24236.21 25622.93 25715.34 26596.46 1. (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
LULESH OpenBenchmarking.org z/s, More Is Better LULESH 2.0.3 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2K 4K 6K 8K 10K SE +/- 38.48, N = 3 SE +/- 19.00, N = 3 SE +/- 67.88, N = 3 SE +/- 13.71, N = 3 9832.64 9945.53 10235.72 10578.52 1. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi
x265 Video Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better x265 Video Input: Bosphorus 1080p 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 30 60 90 120 150 SE +/- 0.23, N = 3 SE +/- 0.13, N = 3 SE +/- 0.45, N = 3 SE +/- 0.27, N = 3 131.28 134.19 134.48 134.52 1. x265 [info]: HEVC encoder version 3.5+1-f0c1022b6
NAS Parallel Benchmarks Test / Class: EP.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.C 2 x 16GB DDR5-6000 CL30 F5-6000J3038F16G 2 x 24GB DDR5-8000 CL38 CMP48GX5M2X8000C38 2 x 32GB DDR5-6400 CL32 CMK64GX5M2B6400C32 2 x 16GB DDR5-8000 CL36 CMH32GX5M2X8000C36 800 1600 2400 3200 4000 SE +/- 25.12, N = 3 SE +/- 21.03, N = 3 SE +/- 53.14, N = 3 SE +/- 17.86, N = 3 3678.46 3722.56 3731.97 3741.86 1. (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
Phoronix Test Suite v10.8.5