AMD EPYC 9005 Turin vs. NVIDIA GH200 Grace CPU Performance Benchmarks

Benchmarks by Michael Larabel for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2411074-NE-MERGE254331&grs&sro.

AMD EPYC 9005 Turin vs. NVIDIA GH200 Grace CPU Performance BenchmarksProcessorMotherboardMemoryDiskGraphicsNetworkChipsetOSKernelDisplay DriverOpenCLCompilerFile-SystemScreen ResolutionNVIDIA GH200 Grace CPUEPYC 9575FEPYC 9655EPYC 9755ARMv8 Neoverse-V2 @ 3.47GHz (72 Cores)Pegatron JIMBO P4352 (00022432 BIOS)1 x 480GB LPDDR5-6400MT/s NVIDIA 699-2G530-0236-RC11000GB CT1000T700SSD3NVIDIA GH200 144G HBM3e 143GB2 x Intel X550Ubuntu 24.046.8.0-47-generic-64k (aarch64)NVIDIAOpenCL 3.0 CUDA 12.6.65GCC 13.2.0 + CUDA 11.8ext41920x1200AMD EPYC 9575F 64-Core @ 3.30GHz (64 Cores / 128 Threads)AMD VOLCANO (RVOT1000D BIOS)AMD Device 153a12 x 64GB DDR5-6000MT/s Samsung M321R8GA0PB1-CCPKC2 x 1920GB KIOXIA KCD8XPUG1T92ASPEEDBroadcom NetXtreme BCM5720 PCIe6.10.0-phx (x86_64)GCC 13.2.0AMD EPYC 9655 96-Core @ 2.60GHz (96 Cores / 192 Threads)AMD EPYC 9755 128-Core @ 2.70GHz (128 Cores / 256 Threads)OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- NVIDIA GH200 Grace CPU: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --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-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-dIwDw0/gcc-13-13.2.0/debian/tmp-nvptx/usr --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto --without-cuda-driver -v - EPYC 9575F: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - EPYC 9655: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - EPYC 9755: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-OiuXZC/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details- NVIDIA GH200 Grace CPU: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)- EPYC 9575F: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xb002110- EPYC 9655: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xb002110- EPYC 9755: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xb002110Python Details- NVIDIA GH200 Grace CPU: Python 3.12.6- EPYC 9575F: Python 3.12.2- EPYC 9655: Python 3.12.2- EPYC 9755: Python 3.12.2Security Details- NVIDIA GH200 Grace CPU: 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 __user pointer sanitization + spectre_v2: Not affected + srbds: Not affected + tsx_async_abort: Not affected - EPYC 9575F: 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 - EPYC 9655: 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 - EPYC 9755: 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- EPYC 9575F, EPYC 9655, EPYC 9755: OpenJDK Runtime Environment (build 21.0.3-ea+7-Ubuntu-1build1)

AMD EPYC 9005 Turin vs. NVIDIA GH200 Grace CPU Performance Benchmarksopenssl: ChaCha20minibude: OpenMP - BM2minibude: OpenMP - BM2memcached: 1:100john-the-ripper: bcryptjohn-the-ripper: Blowfishgromacs: MPI CPU - water_GMX50_bareopenssl: AES-256-GCMcoremark: CoreMark Size 666 - Iterations Per Secondastcenc: Very Thoroughastcenc: Exhaustiveastcenc: Thoroughopenfoam: drivaerFastback, Medium Mesh Size - Execution Timecompress-7zip: Compression Ratingcompress-7zip: Decompression Ratingliquid-dsp: 256 - 256 - 32build-nodejs: Time To Compileopenfoam: drivaerFastback, Small Mesh Size - Execution Timecompress-pbzip2: FreeBSD-13.0-RELEASE-amd64-memstick.img Compressionquantlib: Multi-Threadedprimesieve: 1e13primesieve: 1e12speedb: Rand Readrocksdb: Rand Readhelsing: 14 digitgraphics-magick: Sharpenamg: graphics-magick: Noise-Gaussianlammps: Rhodopsin Proteinnumpy: clickhouse: 100M Rows Hits Dataset, Second Runclickhouse: 100M Rows Hits Dataset, Third Rungraphics-magick: Enhancedclickhouse: 100M Rows Hits Dataset, First Run / Cold Cachebuild-gem5: Time To Compilelammps: 20k Atomsincompact3d: X3D-benchmarking input.i3dqmcpack: Li2_STO_aenwchem: C240 Buckyballstockfish: Chess Benchmarkxmrig: GhostRider - 1Mpennant: sedovbigpennant: leblancbigspeedb: Read While WritingNVIDIA GH200 Grace CPUEPYC 9575FEPYC 9655EPYC 97551492551695431366.39554.6563049705.5474749747566.0025156805063102396678.9962526.58854.066046.4120362.267124366004246434502400000202.93737.3738061.682797261141.531.4642.66756902111156833946356.972428225645466739759.841699.30594.01596.49369581.59166.09758.143240.28580269.6091214.7981468584015.56.4303044.473779102410427330602860536101.793244.07213808558.4219907819895414.65111936992449903884226.12849710.11976.203572.3773233.11516552415270375265866667133.85725.604241.263408297665.125.3972.12352667179250573352561.827301319761125029142.850962.28809.27809.87336779.31130.79954.786217.38342372.9331225.718213596711601.96.2211083.238866107832819054096342136705.147268.20612513777.6923773923782317.63213445994030874550652.09896411.57577.105381.6844195.147567628916266006327600000118.69824.0717521.131325341991.724.2152.04260776965057721537459.286320307673966728445.222870.03755.69763.39364740.48126.33356.103201.84742282.2111334.121918954314312.35.4787083.1365171500396511863592944509654.576386.18313316660.1032300032301022.72918377658507476042366.34304715.71959.6576110.0383160.373789081598467368494733333109.56120.5970890.934896465436.817.9681.55382210282878689658043.057403317708200031355.791795.83729.56724.64455698.12128.13267.776199.84458979.0201326.230734392319873.04.3936062.76126018687897OpenBenchmarking.org

OpenSSL

Algorithm: ChaCha20

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: ChaCha20EPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU300000M600000M900000M1200000M1500000MSE +/- 24295011.71, N = 3SE +/- 211781187.22, N = 3SE +/- 507390352.95, N = 3SE +/- 3236084.95, N = 37330602860539054096342131186359294450149255169543-m64-m64-m641. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

miniBUDE

Implementation: OpenMP - Input Deck: BM2

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2EPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU2K4K6K8K10KSE +/- 15.38, N = 3SE +/- 42.42, N = 3SE +/- 76.86, N = 4SE +/- 0.49, N = 36101.796705.159654.581366.40-march=native-march=native-march=native1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -lm

miniBUDE

Implementation: OpenMP - Input Deck: BM2

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2EPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU80160240320400SE +/- 0.62, N = 3SE +/- 1.70, N = 3SE +/- 3.07, N = 4SE +/- 0.02, N = 3244.07268.21386.1854.66-march=native-march=native-march=native-mcpu=native1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -lm

Memcached

Set To Get Ratio: 1:100

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:100EPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU3M6M9M12M15MSE +/- 30489.72, N = 3SE +/- 51257.47, N = 3SE +/- 136953.60, N = 3SE +/- 22880.28, N = 313808558.4212513777.6913316660.103049705.541. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

John The Ripper

Test: bcrypt

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: bcryptEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU70K140K210K280K350KSE +/- 77.93, N = 3SE +/- 23.62, N = 3SE +/- 45.32, N = 3SE +/- 31.26, N = 319907823773932300074749-m64 -lgmp -lbz2-m64 -lgmp -lbz2-m64 -lgmp -lbz21. (CC) gcc options: -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

John The Ripper

Test: Blowfish

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: BlowfishEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU70K140K210K280K350KSE +/- 41.32, N = 3SE +/- 7.33, N = 3SE +/- 53.35, N = 3SE +/- 19.86, N = 319895423782332301074756-m64 -lgmp -lbz2-m64 -lgmp -lbz2-m64 -lgmp -lbz21. (CC) gcc options: -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

GROMACS

Implementation: MPI CPU - Input: water_GMX50_bare

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2024Implementation: MPI CPU - Input: water_GMX50_bareEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU510152025SE +/- 0.053, N = 3SE +/- 0.014, N = 3SE +/- 0.035, N = 3SE +/- 0.005, N = 314.65117.63222.7296.002-O3 -lm-O3 -lm-O3 -lm1. (CXX) g++ options:

OpenSSL

Algorithm: AES-256-GCM

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.3Algorithm: AES-256-GCMEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU400000M800000M1200000M1600000M2000000MSE +/- 947677989.11, N = 3SE +/- 243566219.34, N = 3SE +/- 2482404044.78, N = 3SE +/- 82190365.25, N = 3119369924499013445994030871837765850747515680506310-m64-m64-m641. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

Coremark

CoreMark Size 666 - Iterations Per Second

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU1.3M2.6M3.9M5.2M6.5MSE +/- 4584.67, N = 3SE +/- 8964.82, N = 3SE +/- 10789.95, N = 3SE +/- 24238.24, N = 33884226.134550652.106042366.342396679.001. (CC) gcc options: -O2 -lrt" -lrt

ASTC Encoder

Preset: Very Thorough

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.7Preset: Very ThoroughEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU48121620SE +/- 0.0104, N = 3SE +/- 0.0010, N = 3SE +/- 0.0025, N = 3SE +/- 0.0008, N = 310.119711.575715.71956.58851. (CXX) g++ options: -O3 -flto -pthread

ASTC Encoder

Preset: Exhaustive

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.7Preset: ExhaustiveEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU3691215SE +/- 0.0055, N = 3SE +/- 0.0012, N = 3SE +/- 0.0004, N = 3SE +/- 0.0036, N = 36.20357.10539.65764.06601. (CXX) g++ options: -O3 -flto -pthread

ASTC Encoder

Preset: Thorough

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.7Preset: ThoroughEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU20406080100SE +/- 0.06, N = 5SE +/- 0.01, N = 5SE +/- 0.02, N = 6SE +/- 0.06, N = 472.3881.68110.0446.411. (CXX) g++ options: -O3 -flto -pthread

OpenFOAM

Input: drivaerFastback, Medium Mesh Size - Execution Time

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Execution TimeEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU80160240320400233.12195.15160.37362.27-m64 -lfiniteVolume -lmeshTools -lparallel -lregionModels-m64 -lfiniteVolume -lmeshTools -lparallel -lregionModels-m64 -lfiniteVolume -lmeshTools -lparallel -lregionModels-mcpu=native -lfoamToVTK -ldynamicMesh -lfileFormats1. (CXX) g++ options: -std=c++14 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -llagrangian -lgenericPatchFields -lOpenFOAM -ldl -lm

7-Zip Compression

Test: Compression Rating

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Compression RatingEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU200K400K600K800K1000KSE +/- 4011.15, N = 3SE +/- 660.09, N = 3SE +/- 3530.77, N = 3SE +/- 652.77, N = 36552417628919081594366001. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

7-Zip Compression

Test: Decompression Rating

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Decompression RatingEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU200K400K600K800K1000KSE +/- 344.02, N = 3SE +/- 257.58, N = 3SE +/- 276.36, N = 3SE +/- 414.91, N = 35270376266008467364246431. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

Liquid-DSP

Threads: 256 - Buffer Length: 256 - Filter Length: 32

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 256 - Buffer Length: 256 - Filter Length: 32EPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU2000M4000M6000M8000M10000MSE +/- 6072982.06, N = 3SE +/- 6065476.07, N = 3SE +/- 14339494.80, N = 3SE +/- 3002221.40, N = 352658666676327600000849473333345024000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Timed Node.js Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 21.7.2Time To CompileEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU4080120160200SE +/- 0.26, N = 3SE +/- 0.12, N = 3SE +/- 0.21, N = 3SE +/- 0.15, N = 3133.86118.70109.56202.94

OpenFOAM

Input: drivaerFastback, Small Mesh Size - Execution Time

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Execution TimeEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU91827364525.6024.0720.6037.37-m64 -lfiniteVolume -lmeshTools -lparallel -lregionModels-m64 -lfiniteVolume -lmeshTools -lparallel -lregionModels-m64 -lfiniteVolume -lmeshTools -lparallel -lregionModels-mcpu=native -lfoamToVTK -ldynamicMesh -lfileFormats1. (CXX) g++ options: -std=c++14 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -llagrangian -lgenericPatchFields -lOpenFOAM -ldl -lm

Parallel BZIP2 Compression

FreeBSD-13.0-RELEASE-amd64-memstick.img Compression

OpenBenchmarking.orgSeconds, Fewer Is BetterParallel BZIP2 Compression 1.1.13FreeBSD-13.0-RELEASE-amd64-memstick.img CompressionEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU0.37860.75721.13581.51441.893SE +/- 0.009364, N = 15SE +/- 0.014837, N = 15SE +/- 0.010948, N = 15SE +/- 0.004252, N = 151.2634081.1313250.9348961.6827971. (CXX) g++ options: -O2 -pthread -lbz2 -lpthread

QuantLib

Configuration: Multi-Threaded

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.32Configuration: Multi-ThreadedEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU100K200K300K400K500KSE +/- 661.19, N = 3SE +/- 414.37, N = 3SE +/- 401.79, N = 3SE +/- 316.93, N = 3297665.1341991.7465436.8261141.51. (CXX) g++ options: -O3 -march=native -fPIE -pie

Primesieve

Length: 1e13

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 12.1Length: 1e13EPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU714212835SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 325.4024.2217.9731.461. (CXX) g++ options: -O3

Primesieve

Length: 1e12

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 12.1Length: 1e12EPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU0.60011.20021.80032.40043.0005SE +/- 0.003, N = 11SE +/- 0.004, N = 11SE +/- 0.004, N = 12SE +/- 0.003, N = 102.1232.0421.5532.6671. (CXX) g++ options: -O3

Speedb

Test: Random Read

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Random ReadEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU200M400M600M800M1000MSE +/- 755684.49, N = 3SE +/- 696066.18, N = 3SE +/- 707549.54, N = 3SE +/- 126199.44, N = 3526671792607769650822102828569021111-lpthread-lpthread-lpthread1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

RocksDB

Test: Random Read

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Random ReadEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU200M400M600M800M1000MSE +/- 1563552.18, N = 3SE +/- 4322874.61, N = 3SE +/- 5260460.78, N = 3SE +/- 143829.35, N = 3505733525577215374786896580568339463-lpthread-lpthread-lpthread1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

Helsing

Digit Range: 14 digit

OpenBenchmarking.orgSeconds, Fewer Is BetterHelsing 1.0-betaDigit Range: 14 digitEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU1428425670SE +/- 0.23, N = 3SE +/- 0.26, N = 3SE +/- 0.13, N = 3SE +/- 0.02, N = 361.8359.2943.0656.971. (CC) gcc options: -O2 -pthread

GraphicsMagick

Operation: Sharpen

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: SharpenEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU90180270360450SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3301320403428-lfreetype -lbz2-lfreetype -lbz2-lfreetype -lbz21. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lz -lm -lpthread -lgomp

Algebraic Multi-Grid Benchmark

OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid Benchmark 1.2EPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU700M1400M2100M2800M3500MSE +/- 7735064.17, N = 4SE +/- 4068843.09, N = 3SE +/- 6342005.91, N = 3SE +/- 3694072.61, N = 331976112503076739667317708200022564546671. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi

GraphicsMagick

Operation: Noise-Gaussian

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: Noise-GaussianEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU90180270360450SE +/- 1.00, N = 3SE +/- 2.08, N = 3SE +/- 0.58, N = 3SE +/- 0.00, N = 3291284313397-lfreetype -lbz2-lfreetype -lbz2-lfreetype -lbz21. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lz -lm -lpthread -lgomp

LAMMPS Molecular Dynamics Simulator

Model: Rhodopsin Protein

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: Rhodopsin ProteinEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU1326395265SE +/- 0.16, N = 11SE +/- 0.32, N = 15SE +/- 0.31, N = 10SE +/- 0.13, N = 1142.8545.2255.7959.841. (CXX) g++ options: -O3 -lm -ldl

Numpy Benchmark

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU2004006008001000SE +/- 3.50, N = 3SE +/- 0.88, N = 3SE +/- 1.03, N = 3SE +/- 1.78, N = 3962.28870.03795.83699.30

ClickHouse

100M Rows Hits Dataset, Second Run

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Second RunEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU2004006008001000SE +/- 3.75, N = 3SE +/- 5.05, N = 3SE +/- 5.27, N = 3SE +/- 2.22, N = 3809.27755.69729.56594.01MIN: 65.65 / MAX: 8571.43MIN: 69.93 / MAX: 7500MIN: 81.63 / MAX: 7500MIN: 59.29 / MAX: 7500

ClickHouse

100M Rows Hits Dataset, Third Run

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third RunEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU2004006008001000SE +/- 1.25, N = 3SE +/- 0.75, N = 3SE +/- 4.70, N = 3SE +/- 2.50, N = 3809.87763.39724.64596.49MIN: 65.43 / MAX: 10000MIN: 69.36 / MAX: 8571.43MIN: 85.11 / MAX: 6666.67MIN: 59.29 / MAX: 7500

GraphicsMagick

Operation: Enhanced

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: EnhancedEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU100200300400500SE +/- 0.67, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3336364455369-lfreetype -lbz2-lfreetype -lbz2-lfreetype -lbz21. (CC) gcc options: -fopenmp -O2 -ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lzstd -llzma -lz -lm -lpthread -lgomp

ClickHouse

100M Rows Hits Dataset, First Run / Cold Cache

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, First Run / Cold CacheEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU2004006008001000SE +/- 3.67, N = 3SE +/- 1.08, N = 3SE +/- 6.12, N = 3SE +/- 2.14, N = 3779.31740.48698.12581.59MIN: 66.01 / MAX: 7500MIN: 68.57 / MAX: 8571.43MIN: 80.86 / MAX: 6666.67MIN: 58.94 / MAX: 6666.67

Timed Gem5 Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Gem5 Compilation 23.0.1Time To CompileEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU4080120160200SE +/- 0.56, N = 3SE +/- 1.17, N = 3SE +/- 1.52, N = 3SE +/- 0.26, N = 3130.80126.33128.13166.10

LAMMPS Molecular Dynamics Simulator

Model: 20k Atoms

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: 20k AtomsEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU1530456075SE +/- 0.37, N = 3SE +/- 0.52, N = 3SE +/- 0.06, N = 3SE +/- 0.14, N = 354.7956.1067.7858.141. (CXX) g++ options: -O3 -lm -ldl

Xcompact3d Incompact3d

Input: X3D-benchmarking input.i3d

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: X3D-benchmarking input.i3dEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU50100150200250SE +/- 1.26, N = 3SE +/- 0.55, N = 3SE +/- 0.78, N = 3SE +/- 0.29, N = 3217.38201.85199.84240.291. (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

QMCPACK

Input: Li2_STO_ae

OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.17.1Input: Li2_STO_aeEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU20406080100SE +/- 0.39, N = 3SE +/- 0.22, N = 3SE +/- 0.27, N = 3SE +/- 0.31, N = 372.9382.2179.0269.61-march=native-march=native-march=native-mcpu=native1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -O3 -lm -ldl

NWChem

Input: C240 Buckyball

OpenBenchmarking.orgSeconds, Fewer Is BetterNWChem 7.0.2Input: C240 BuckyballEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU300600900120015001225.71334.11326.21214.7-m64-m64-m641. (F9X) gfortran options: -lnwctask -lccsd -lmcscf -lselci -lmp2 -lmoints -lstepper -ldriver -loptim -lnwdft -lgradients -lcphf -lesp -lddscf -ldangchang -lguess -lhessian -lvib -lnwcutil -lrimp2 -lproperty -lsolvation -lnwints -lprepar -lnwmd -lnwpw -lofpw -lpaw -lpspw -lband -lnwpwlib -lcafe -lspace -lanalyze -lqhop -lpfft -ldplot -ldrdy -lvscf -lqmmm -lqmd -letrans -ltce -lbq -lmm -lcons -lperfm -ldntmc -lccca -ldimqm -lga -larmci -lpeigs -l64to32 -lopenblas -lpthread -lrt -llapack -lnwcblas -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz -lcomex -ffast-math -std=legacy -fdefault-integer-8 -finline-functions -O2

CPU Power Consumption Monitor

Phoronix Test Suite System Monitoring

OpenBenchmarking.orgWattsCPU Power Consumption MonitorPhoronix Test Suite System MonitoringEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU90180270360450Min: 19.17 / Avg: 313.09 / Max: 403.08Min: 41.04 / Avg: 253.26 / Max: 371.62Min: 44.19 / Avg: 324.1 / Max: 500.98Min: 38.42 / Avg: 170.19 / Max: 300.47

Primesieve

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.3321.0393.1EPYC 965542.2267.8344.1EPYC 975544.6342.5459.6NVIDIA GH200 Grace CPU40.9233.0271.7OpenBenchmarking.orgWatts, Fewer Is BetterPrimesieve 12.1CPU Power Consumption Monitor120240360480600

Primesieve

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.5146.3375.7EPYC 965542.2130.2315.1EPYC 975544.7147.2431.8NVIDIA GH200 Grace CPU41.8111.4254.2OpenBenchmarking.orgWatts, Fewer Is BetterPrimesieve 12.1CPU Power Consumption Monitor110220330440550

Stockfish

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.7371.1401.2EPYC 965542.3328.0363.7EPYC 975544.8440.6487.2NVIDIA GH200 Grace CPU42.7261.4290.6OpenBenchmarking.orgWatts, Fewer Is BetterStockfish 16.1CPU Power Consumption Monitor130260390520650

Stockfish

Chess Benchmark

OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterStockfish 16.1Chess BenchmarkEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU150K300K450K600K750K490782.75668223.51697627.07375426.89

Stockfish

Chess Benchmark

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 16.1Chess BenchmarkEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU70M140M210M280M350MSE +/- 2577463.33, N = 15SE +/- 3587851.54, N = 15SE +/- 4338134.07, N = 15SE +/- 1172057.21, N = 1518213596721918954330734392398146858-m64 -msse -msse3 -mpopcnt -mavx2 -mbmi -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2-m64 -msse -msse3 -mpopcnt -mavx2 -mbmi -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2-m64 -msse -msse3 -mpopcnt -mavx2 -mbmi -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi21. (CXX) g++ options: -lgcov -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -funroll-loops -flto -flto-partition=one -flto=jobserver

Helsing

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.4346.7373.1EPYC 965542.7287.0311.4EPYC 975544.9388.8429.7NVIDIA GH200 Grace CPU40.6266.7291.3OpenBenchmarking.orgWatts, Fewer Is BetterHelsing 1.0-betaCPU Power Consumption Monitor110220330440550

Xmrig

CPU Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterXmrig 6.21CPU Power Consumption MonitorEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU90180270360450Min: 38.21 / Avg: 380.2 / Max: 401.2Min: 42.04 / Avg: 321.63 / Max: 347.4Min: 44.9 / Avg: 424.95 / Max: 482.49Min: 41.25 / Avg: 139.96 / Max: 146.69

Xmrig

Variant: GhostRider - Hash Count: 1M

OpenBenchmarking.orgH/s Per Watt, More Is BetterXmrig 6.21Variant: GhostRider - Hash Count: 1MEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU112233445530.5244.5046.7728.69

Xmrig

Variant: GhostRider - Hash Count: 1M

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.21Variant: GhostRider - Hash Count: 1MEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU4K8K12K16K20KSE +/- 508.75, N = 15SE +/- 672.73, N = 15SE +/- 912.79, N = 15SE +/- 0.17, N = 311601.914312.319873.04015.5-maes-maes-maes1. (CXX) g++ options: -fexceptions -fno-rtti -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

Liquid-DSP

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.4356.3401.0EPYC 965542.5305.9344.2EPYC 975545.2422.3487.6NVIDIA GH200 Grace CPU41.2256.8292.1OpenBenchmarking.orgWatts, Fewer Is BetterLiquid-DSP 1.6CPU Power Consumption Monitor130260390520650

Liquid-DSP

Threads: 256 - Buffer Length: 256 - Filter Length: 32

OpenBenchmarking.orgsamples/s Per Watt, More Is BetterLiquid-DSP 1.6Threads: 256 - Buffer Length: 256 - Filter Length: 32EPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU4M8M12M16M20M14781256.8820685459.0520115931.6317535441.66

GraphicsMagick

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.7333.1356.2EPYC 965542.4269.3287.7EPYC 975545.0344.3371.9NVIDIA GH200 Grace CPU42.8267.5300.5OpenBenchmarking.orgWatts, Fewer Is BetterGraphicsMagick 1.3.43CPU Power Consumption Monitor100200300400500

GraphicsMagick

Operation: Sharpen

OpenBenchmarking.orgIterations Per Minute Per Watt, More Is BetterGraphicsMagick 1.3.43Operation: SharpenEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU0.360.721.081.441.80.9041.1881.1701.600

GraphicsMagick

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.3340.8367.5EPYC 965542.4256.1275.5EPYC 975544.7323.9350.4NVIDIA GH200 Grace CPU40.7239.3262.5OpenBenchmarking.orgWatts, Fewer Is BetterGraphicsMagick 1.3.43CPU Power Consumption Monitor100200300400500

GraphicsMagick

Operation: Enhanced

OpenBenchmarking.orgIterations Per Minute Per Watt, More Is BetterGraphicsMagick 1.3.43Operation: EnhancedEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU0.3470.6941.0411.3881.7350.9861.4211.4051.542

GraphicsMagick

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.4224.7243.8EPYC 965542.7181.6199.6EPYC 975544.7208.0234.1NVIDIA GH200 Grace CPU41.9166.4283.3OpenBenchmarking.orgWatts, Fewer Is BetterGraphicsMagick 1.3.43CPU Power Consumption Monitor70140210280350

GraphicsMagick

Operation: Noise-Gaussian

OpenBenchmarking.orgIterations Per Minute Per Watt, More Is BetterGraphicsMagick 1.3.43Operation: Noise-GaussianEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU0.53661.07321.60982.14642.6831.2951.5641.5052.385

ASTC Encoder

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.4312.4391.4EPYC 965542.6256.3324.5EPYC 975544.7318.6450.4NVIDIA GH200 Grace CPU42.1223.2267.3OpenBenchmarking.orgWatts, Fewer Is BetterASTC Encoder 4.7CPU Power Consumption Monitor120240360480600

ASTC Encoder

Preset: Exhaustive

OpenBenchmarking.orgMT/s Per Watt, More Is BetterASTC Encoder 4.7Preset: ExhaustiveEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU0.00680.01360.02040.02720.0340.0200.0280.0300.018

ASTC Encoder

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.4314.2390.3EPYC 965542.4254.7323.1EPYC 975544.5317.5448.4NVIDIA GH200 Grace CPU39.8218.7261.2OpenBenchmarking.orgWatts, Fewer Is BetterASTC Encoder 4.7CPU Power Consumption Monitor120240360480600

ASTC Encoder

Preset: Very Thorough

OpenBenchmarking.orgMT/s Per Watt, More Is BetterASTC Encoder 4.7Preset: Very ThoroughEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU0.01130.02260.03390.04520.05650.0320.0450.0500.030

ASTC Encoder

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.2219.9386.4EPYC 965542.2178.3319.1EPYC 975544.5203.3437.2NVIDIA GH200 Grace CPU39.1162.3263.1OpenBenchmarking.orgWatts, Fewer Is BetterASTC Encoder 4.7CPU Power Consumption Monitor110220330440550

ASTC Encoder

Preset: Thorough

OpenBenchmarking.orgMT/s Per Watt, More Is BetterASTC Encoder 4.7Preset: ThoroughEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU0.12170.24340.36510.48680.60850.3290.4580.5410.286

Numpy Benchmark

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.193.5100.9EPYC 965542.189.396.9EPYC 975544.492.599.7NVIDIA GH200 Grace CPU38.743.047.7OpenBenchmarking.orgWatts, Fewer Is BetterNumpy BenchmarkCPU Power Consumption Monitor20406080100

Numpy Benchmark

OpenBenchmarking.orgScore Per Watt, More Is BetterNumpy BenchmarkEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU4812162010.2919.7398.60316.277

Parallel BZIP2 Compression

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.2104.3361.5EPYC 965542.489.6271.2EPYC 975544.597.5319.0NVIDIA GH200 Grace CPU40.068.7207.5OpenBenchmarking.orgWatts, Fewer Is BetterParallel BZIP2 Compression 1.1.13CPU Power Consumption Monitor100200300400500

7-Zip Compression

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.4322.5400.8EPYC 965542.5283.3363.2EPYC 975544.7373.2482.8NVIDIA GH200 Grace CPU42.0162.5246.0OpenBenchmarking.orgWatts, Fewer Is Better7-Zip Compression 22.01CPU Power Consumption Monitor120240360480600

7-Zip Compression

Test: Decompression Rating

OpenBenchmarking.orgMIPS Per Watt, More Is Better7-Zip Compression 22.01Test: Decompression RatingEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU60012001800240030001634.382211.882269.162612.52

OpenFOAM

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.2388.3399.8EPYC 965542.3331.2360.0EPYC 975544.6450.2489.9NVIDIA GH200 Grace CPU41.1200.9291.9OpenBenchmarking.orgWatts, Fewer Is BetterOpenFOAM 10CPU Power Consumption Monitor130260390520650

OpenFOAM

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.8346.4399.6EPYC 965542.3304.0363.0EPYC 975544.4396.4492.6NVIDIA GH200 Grace CPU41.1200.0287.7OpenBenchmarking.orgWatts, Fewer Is BetterOpenFOAM 10CPU Power Consumption Monitor130260390520650

Xcompact3d Incompact3d

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.2389.8399.8EPYC 965542.3324.2343.2EPYC 975544.9401.7443.9NVIDIA GH200 Grace CPU42.2172.9253.2OpenBenchmarking.orgWatts, Fewer Is BetterXcompact3d Incompact3d 2021-03-11CPU Power Consumption Monitor120240360480600

NWChem

CPU Power Consumption Monitor

OpenBenchmarking.orgWatts, Fewer Is BetterNWChem 7.0.2CPU Power Consumption MonitorEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU90180270360450Min: 38.3 / Avg: 392.8 / Max: 400.06Min: 42.34 / Avg: 346.39 / Max: 369.48Min: 44.44 / Avg: 477.37 / Max: 499.23Min: 41.49 / Avg: 226.74 / Max: 281.44

Pennant

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.2248.3399.7EPYC 965542.5209.0339.7EPYC 975544.5248.9471.5NVIDIA GH200 Grace CPU41.5150.9252.9OpenBenchmarking.orgWatts, Fewer Is BetterPennant 1.0.1CPU Power Consumption Monitor120240360480600

Pennant

Test: sedovbig

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: sedovbigEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU246810SE +/- 0.034501, N = 6SE +/- 0.054129, N = 15SE +/- 0.073388, N = 15SE +/- 0.012489, N = 66.2211085.4787084.3936066.4303041. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi

Pennant

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.2195.5397.8EPYC 965542.5171.4339.9EPYC 975544.6210.5475.6NVIDIA GH200 Grace CPU41.7140.4265.6OpenBenchmarking.orgWatts, Fewer Is BetterPennant 1.0.1CPU Power Consumption Monitor120240360480600

Pennant

Test: leblancbig

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: leblancbigEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU1.00662.01323.01984.02645.033SE +/- 0.077441, N = 15SE +/- 0.087065, N = 15SE +/- 0.045208, N = 15SE +/- 0.007724, N = 73.2388663.1365172.7612604.4737791. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi

QMCPACK

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.7376.4399.8EPYC 965542.8322.1343.9EPYC 975544.9441.8480.5NVIDIA GH200 Grace CPU41.5268.3298.1OpenBenchmarking.orgWatts, Fewer Is BetterQMCPACK 3.17.1CPU Power Consumption Monitor120240360480600

QuantLib

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.5331.7399.4EPYC 965541.9300.9345.9EPYC 975544.4406.0483.0NVIDIA GH200 Grace CPU41.0215.9290.4OpenBenchmarking.orgWatts, Fewer Is BetterQuantLib 1.32CPU Power Consumption Monitor120240360480600

QuantLib

Configuration: Multi-Threaded

OpenBenchmarking.orgMFLOPS Per Watt, More Is BetterQuantLib 1.32Configuration: Multi-ThreadedEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU30060090012001500897.441136.471146.371209.80

GROMACS

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.3315.2400.1EPYC 965542.2266.4347.6EPYC 975544.6349.2487.2NVIDIA GH200 Grace CPU40.8212.1289.0OpenBenchmarking.orgWatts, Fewer Is BetterGROMACS 2024CPU Power Consumption Monitor130260390520650

GROMACS

Implementation: MPI CPU - Input: water_GMX50_bare

OpenBenchmarking.orgNs Per Day Per Watt, More Is BetterGROMACS 2024Implementation: MPI CPU - Input: water_GMX50_bareEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU0.01490.02980.04470.05960.07450.0460.0660.0650.028

LAMMPS Molecular Dynamics Simulator

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.4388.8399.6EPYC 965542.2334.9355.5EPYC 975544.3461.9488.1NVIDIA GH200 Grace CPU40.1274.7292.0OpenBenchmarking.orgWatts, Fewer Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022CPU Power Consumption Monitor130260390520650

LAMMPS Molecular Dynamics Simulator

Model: 20k Atoms

OpenBenchmarking.orgns/day Per Watt, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: 20k AtomsEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU0.04770.09540.14310.19080.23850.1410.1680.1470.212

LAMMPS Molecular Dynamics Simulator

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.192.8302.5EPYC 965542.0101.8272.7EPYC 975544.5125.0352.9NVIDIA GH200 Grace CPU39.961.9280.4OpenBenchmarking.orgWatts, Fewer Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022CPU Power Consumption Monitor100200300400500

LAMMPS Molecular Dynamics Simulator

Model: Rhodopsin Protein

OpenBenchmarking.orgns/day Per Watt, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: Rhodopsin ProteinEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU0.21760.43520.65280.87041.0880.4620.4440.4460.967

miniBUDE

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.3319.7382.8EPYC 965542.0282.8341.8EPYC 975544.7355.4459.9NVIDIA GH200 Grace CPU41.1214.0244.9OpenBenchmarking.orgWatts, Fewer Is BetterminiBUDE 20210901CPU Power Consumption Monitor120240360480600

miniBUDE

Implementation: OpenMP - Input Deck: BM2

OpenBenchmarking.orgBillion Interactions/s Per Watt, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2EPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU0.24460.48920.73380.97841.2230.7630.9481.0870.255

Algebraic Multi-Grid Benchmark

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.3294.2401.4EPYC 965541.9266.1338.1EPYC 975544.4334.4438.6NVIDIA GH200 Grace CPU39.8146.2260.2OpenBenchmarking.orgWatts, Fewer Is BetterAlgebraic Multi-Grid Benchmark 1.2CPU Power Consumption Monitor120240360480600

Algebraic Multi-Grid Benchmark

OpenBenchmarking.orgFigure Of Merit Per Watt, More Is BetterAlgebraic Multi-Grid Benchmark 1.2EPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU3M6M9M12M15M10869334.8311562385.409501967.5515435851.48

Coremark

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.0314.4399.6EPYC 965541.8318.2348.1EPYC 975544.7414.3479.8NVIDIA GH200 Grace CPU38.8119.4159.8OpenBenchmarking.orgWatts, Fewer Is BetterCoremark 1.0CPU Power Consumption Monitor120240360480600

Coremark

CoreMark Size 666 - Iterations Per Second

OpenBenchmarking.orgIterations/Sec Per Watt, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU4K8K12K16K20K12356.0814303.0314585.7420070.84

Speedb

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.5375.8400.8EPYC 965542.3313.0351.7EPYC 975544.6398.2497.6NVIDIA GH200 Grace CPU43.4181.9215.8OpenBenchmarking.orgWatts, Fewer Is BetterSpeedb 2.7CPU Power Consumption Monitor130260390520650

Speedb

Test: Read While Writing

OpenBenchmarking.orgOp/s Per Watt, More Is BetterSpeedb 2.7Test: Read While WritingEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU12K24K36K48K60K28692.8247929.3346936.4456303.84

Speedb

Test: Read While Writing

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Read While WritingEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU4M8M12M16M20MSE +/- 91725.44, N = 15SE +/- 135475.45, N = 3SE +/- 433421.14, N = 12SE +/- 87939.63, N = 1510783281150039651868789710241042-lpthread-lpthread-lpthread1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

Speedb

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.3346.4371.3EPYC 965542.3304.0324.6EPYC 975545.3417.3445.7NVIDIA GH200 Grace CPU42.0266.9286.5OpenBenchmarking.orgWatts, Fewer Is BetterSpeedb 2.7CPU Power Consumption Monitor120240360480600

Speedb

Test: Random Read

OpenBenchmarking.orgOp/s Per Watt, More Is BetterSpeedb 2.7Test: Random ReadEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU500K1000K1500K2000K2500K1520207.331998985.071969975.352132224.78

RocksDB

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.3345.7371.1EPYC 965542.4300.8323.1EPYC 975544.5415.4446.2NVIDIA GH200 Grace CPU40.1270.2291.0OpenBenchmarking.orgWatts, Fewer Is BetterRocksDB 9.0CPU Power Consumption Monitor120240360480600

RocksDB

Test: Random Read

OpenBenchmarking.orgOp/s Per Watt, More Is BetterRocksDB 9.0Test: Random ReadEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU500K1000K1500K2000K2500K1462898.911918650.381894471.542103254.00

Memcached

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.3346.3402.0EPYC 965542.2261.1299.8EPYC 975544.5352.4409.0NVIDIA GH200 Grace CPU39.5110.8124.9OpenBenchmarking.orgWatts, Fewer Is BetterMemcached 1.6.19CPU Power Consumption Monitor110220330440550

Memcached

Set To Get Ratio: 1:100

OpenBenchmarking.orgOps/sec Per Watt, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:100EPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU10K20K30K40K50K39869.2747921.2437790.6227521.07

ClickHouse

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.3193.7368.0EPYC 965541.9165.4302.9EPYC 975544.4163.5407.0NVIDIA GH200 Grace CPU40.074.7282.8OpenBenchmarking.orgWatts, Fewer Is BetterClickHouse 22.12.3.5CPU Power Consumption Monitor110220330440550

ClickHouse

100M Rows Hits Dataset, Third Run

OpenBenchmarking.orgQueries Per Minute, Geo Mean Per Watt, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third RunEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU2468104.1814.6154.4317.980

John The Ripper

Test: bcrypt

OpenBenchmarking.orgReal C/S Per Watt, More Is BetterJohn The Ripper 2023.03.14Test: bcryptEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU2004006008001000561.71789.71782.32517.44

John The Ripper

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.3354.4400.5EPYC 965542.3301.0337.8EPYC 975544.6415.1468.2NVIDIA GH200 144G HBM3e41.0144.5164.9NVIDIA GH200 Grace CPU42.2146.9165.2Xeon 6980P50.8445.3539.8Xeon 6980P - DDR5-640050.4446.6539.6OpenBenchmarking.orgWatts, Fewer Is BetterJohn The Ripper 2023.03.14CPU Power Consumption Monitor140280420560700

John The Ripper

Test: Blowfish

OpenBenchmarking.orgReal C/S Per Watt, More Is BetterJohn The Ripper 2023.03.14Test: BlowfishEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU2004006008001000562.77792.83778.22508.72

OpenSSL

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.6384.1401.2EPYC 965542.6349.5368.4EPYC 975545.0468.4501.0NVIDIA GH200 Grace CPU43.4194.2217.3OpenBenchmarking.orgWatts, Fewer Is BetterOpenSSL 3.3CPU Power Consumption Monitor130260390520650

OpenSSL

Algorithm: ChaCha20

OpenBenchmarking.orgbyte/s Per Watt, More Is BetterOpenSSL 3.3Algorithm: ChaCha20EPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU600M1200M1800M2400M3000M1908650939.882590512735.182532659749.39768573496.68

OpenSSL

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.4358.9382.8EPYC 965542.3308.5324.7EPYC 975544.7423.5443.3NVIDIA GH200 Grace CPU43.5285.0293.6OpenBenchmarking.orgWatts, Fewer Is BetterOpenSSL 3.3CPU Power Consumption Monitor120240360480600

OpenSSL

Algorithm: AES-256-GCM

OpenBenchmarking.orgbyte/s Per Watt, More Is BetterOpenSSL 3.3Algorithm: AES-256-GCMEPYC 9575FEPYC 9655EPYC 9755NVIDIA GH200 Grace CPU900M1800M2700M3600M4500M3325699800.964358545440.704339501704.861809281985.06

Timed Gem5 Compilation

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.5211.8401.3EPYC 965542.1176.5363.1EPYC 975544.4195.4485.4NVIDIA GH200 Grace CPU39.4118.8239.4OpenBenchmarking.orgWatts, Fewer Is BetterTimed Gem5 Compilation 23.0.1CPU Power Consumption Monitor130260390520650

Timed Node.js Compilation

CPU Power Consumption Monitor

MinAvgMaxEPYC 9575F38.6313.9401.2EPYC 965542.1265.8361.5EPYC 975544.5304.2485.2NVIDIA GH200 Grace CPU39.2181.0279.3OpenBenchmarking.orgWatts, Fewer Is BetterTimed Node.js Compilation 21.7.2CPU Power Consumption Monitor120240360480600

Phoronix Test Suite v10.8.5