Extra GPTshop.ai GH200 Linux Benchmarks Benchmarks by Michael Larabel for a future article. GPTshop.ai GH200: Processor: ARMv8 Neoverse-V2 @ 3.39GHz (72 Cores), Motherboard: Quanta Cloud QuantaGrid S74G-2U 1S7GZ9Z0000 S7G MB (CG1) (3A06 BIOS), Memory: 1 x 480GB DRAM-6400MT/s, Disk: 960GB SAMSUNG MZ1L2960HCJR-00A07 + 1920GB SAMSUNG MZTL21T9, Graphics: ASPEED, Network: 2 x Mellanox MT2910 + 2 x QLogic FastLinQ QL41000 10/25/40/50GbE OS: Ubuntu 23.10, Kernel: 6.5.0-15-generic (aarch64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200 Ampere Altra Max M128-30: Processor: ARMv8 Neoverse-N1 @ 3.00GHz (128 Cores), Motherboard: GIGABYTE G242-P36-00 MP32-AR2-00 v01000100 (F31k SCP: 2.10.20220531 BIOS), Chipset: Ampere Computing LLC Altra PCI Root Complex A, Memory: 16 x 32GB DDR4-3200MT/s Samsung M393A4K40DB3-CWE, Disk: 800GB Micron_7450_MTFDKBA800TFS, Graphics: ASPEED, Monitor: VGA HDMI, Network: 2 x Intel I350 OS: Ubuntu 23.10, Kernel: 6.5.0-13-generic (aarch64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1080 HP Z6 G5 A - Threadripper PRO 7995WX: Processor: AMD Ryzen Threadripper PRO 7995WX 96-Cores @ 6.44GHz (96 Cores / 192 Threads), Motherboard: HP Z6 G5 A Workstation 8B24 (U65 Ver. 01.01.04 BIOS), Chipset: AMD Device 14a4, Memory: 8 x 16GB DRAM-5200MT/s Hynix HMCG78AGBRA190N, Disk: 2 x 1024GB SAMSUNG MZVL21T0HCLR-00BH1, Graphics: NVIDIA RTX A4000 16GB, Audio: NVIDIA GA104 HD Audio, Monitor: ASUS VP28U, Network: Realtek RTL8111/8168/8411 OS: Ubuntu 23.10, Kernel: 6.5.0-17-generic (x86_64), Desktop: GNOME Shell 45.2, Display Server: X Server 1.21.1.7, Display Driver: NVIDIA 535.154.05, OpenGL: 4.6.0, OpenCL: OpenCL 3.0 CUDA 12.2.148, Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 3840x2160 High Performance Conjugate Gradient 3.1 X Y Z: 144 144 144 - RT: 60 GFLOP/s > Higher Is Better GPTshop.ai GH200 ..................... 41.69 |================================= Ampere Altra Max M128-30 ............. 21.24 |================= Rodinia 3.1 Test: OpenMP LavaMD Seconds < Lower Is Better GPTshop.ai GH200 ..................... 30.31 |================================ Ampere Altra Max M128-30 ............. 31.67 |================================= HP Z6 G5 A - Threadripper PRO 7995WX . 26.83 |============================ Algebraic Multi-Grid Benchmark 1.2 Figure Of Merit > Higher Is Better GPTshop.ai GH200 ..................... 1997929111 |============================ Ampere Altra Max M128-30 ............. 1059875333 |=============== HP Z6 G5 A - Threadripper PRO 7995WX . 1665755333 |======================= NWChem 7.0.2 Input: C240 Buckyball Seconds < Lower Is Better GPTshop.ai GH200 ..................... 1403.5 |================= Ampere Altra Max M128-30 ............. 2707.9 |================================ HP Z6 G5 A - Threadripper PRO 7995WX . 1594.0 |=================== Xcompact3d Incompact3d 2021-03-11 Input: X3D-benchmarking input.i3d Seconds < Lower Is Better GPTshop.ai GH200 ..................... 254.49 |============= Ampere Altra Max M128-30 ............. 607.68 |================================ HP Z6 G5 A - Threadripper PRO 7995WX . 380.67 |==================== Xcompact3d Incompact3d 2021-03-11 Input: input.i3d 193 Cells Per Direction Seconds < Lower Is Better GPTshop.ai GH200 ..................... 9.81172053 |=========== Ampere Altra Max M128-30 ............. 23.79401910 |=========================== HP Z6 G5 A - Threadripper PRO 7995WX . 10.73783740 |============ LULESH 2.0.3 z/s > Higher Is Better GPTshop.ai GH200 ..................... 23185.18 |============================= Ampere Altra Max M128-30 ............. 16161.79 |==================== HP Z6 G5 A - Threadripper PRO 7995WX . 23737.17 |============================== Xmrig 6.18.1 Variant: Monero - Hash Count: 1M H/s > Higher Is Better GPTshop.ai GH200 ..................... 17253.0 |========= Ampere Altra Max M128-30 ............. 4298.5 |== HP Z6 G5 A - Threadripper PRO 7995WX . 56878.0 |=============================== John The Ripper 2023.03.14 Test: bcrypt Real C/S > Higher Is Better GPTshop.ai GH200 ..................... 68817 |============= Ampere Altra Max M128-30 ............. 109117 |==================== HP Z6 G5 A - Threadripper PRO 7995WX . 172569 |================================ GraphicsMagick 1.3.38 Operation: Sharpen Iterations Per Minute > Higher Is Better GPTshop.ai GH200 ..................... 1363 |================================== Ampere Altra Max M128-30 ............. 1281 |================================ HP Z6 G5 A - Threadripper PRO 7995WX . 823 |===================== GraphicsMagick 1.3.38 Operation: Enhanced Iterations Per Minute > Higher Is Better GPTshop.ai GH200 ..................... 1761 |================================== Ampere Altra Max M128-30 ............. 1233 |======================== HP Z6 G5 A - Threadripper PRO 7995WX . 1363 |========================== ACES DGEMM 1.0 Sustained Floating-Point Rate GFLOP/s > Higher Is Better GPTshop.ai GH200 ..................... 17.94 |============== Ampere Altra Max M128-30 ............. 18.21 |============== HP Z6 G5 A - Threadripper PRO 7995WX . 43.40 |================================= 7-Zip Compression 22.01 Test: Compression Rating MIPS > Higher Is Better GPTshop.ai GH200 ..................... 345295 |==================== Ampere Altra Max M128-30 ............. 334516 |=================== HP Z6 G5 A - Threadripper PRO 7995WX . 549229 |================================ 7-Zip Compression 22.01 Test: Decompression Rating MIPS > Higher Is Better GPTshop.ai GH200 ..................... 389055 |=================== Ampere Altra Max M128-30 ............. 541268 |========================== HP Z6 G5 A - Threadripper PRO 7995WX . 656839 |================================ Stockfish 15 Total Time Nodes Per Second > Higher Is Better GPTshop.ai GH200 ..................... 153826682 |=============== Ampere Altra Max M128-30 ............. 189937895 |================== HP Z6 G5 A - Threadripper PRO 7995WX . 304253357 |============================= asmFish 2018-07-23 1024 Hash Memory, 26 Depth Nodes/second > Higher Is Better GPTshop.ai GH200 ..................... 150936379 |================== Ampere Altra Max M128-30 ............. 214150509 |========================= HP Z6 G5 A - Threadripper PRO 7995WX . 247720673 |============================= Timed Godot Game Engine Compilation 4.0 Time To Compile Seconds < Lower Is Better GPTshop.ai GH200 ..................... 139.10 |=================== Ampere Altra Max M128-30 ............. 229.97 |================================ HP Z6 G5 A - Threadripper PRO 7995WX . 87.23 |============ Timed LLVM Compilation 16.0 Build System: Ninja Seconds < Lower Is Better GPTshop.ai GH200 ..................... 195.98 |======================= Ampere Altra Max M128-30 ............. 267.10 |================================ HP Z6 G5 A - Threadripper PRO 7995WX . 123.62 |=============== Timed Node.js Compilation 19.8.1 Time To Compile Seconds < Lower Is Better GPTshop.ai GH200 ..................... 173.88 |===================== Ampere Altra Max M128-30 ............. 268.43 |================================ HP Z6 G5 A - Threadripper PRO 7995WX . 108.40 |============= Primesieve 8.0 Length: 1e13 Seconds < Lower Is Better GPTshop.ai GH200 ..................... 35.49 |============================ Ampere Altra Max M128-30 ............. 41.91 |================================= HP Z6 G5 A - Threadripper PRO 7995WX . 23.29 |================== Helsing 1.0-beta Digit Range: 14 digit Seconds < Lower Is Better GPTshop.ai GH200 ..................... 67.61 |================================= Ampere Altra Max M128-30 ............. 57.04 |============================ HP Z6 G5 A - Threadripper PRO 7995WX . 55.91 |=========================== Graph500 3.0 Scale: 26 bfs median_TEPS > Higher Is Better GPTshop.ai GH200 ..................... 1249790000 |============================ Ampere Altra Max M128-30 ............. 976326000 |====================== HP Z6 G5 A - Threadripper PRO 7995WX . 626461000 |============== Graph500 3.0 Scale: 26 bfs max_TEPS > Higher Is Better GPTshop.ai GH200 ..................... 1315650000 |============================ Ampere Altra Max M128-30 ............. 985207000 |===================== HP Z6 G5 A - Threadripper PRO 7995WX . 648393000 |============== Graph500 3.0 Scale: 26 sssp median_TEPS > Higher Is Better GPTshop.ai GH200 ..................... 299027000 |========================== Ampere Altra Max M128-30 ............. 222683000 |=================== HP Z6 G5 A - Threadripper PRO 7995WX . 334139000 |============================= Graph500 3.0 Scale: 26 sssp max_TEPS > Higher Is Better GPTshop.ai GH200 ..................... 467012000 |============================= Ampere Altra Max M128-30 ............. 332248000 |===================== HP Z6 G5 A - Threadripper PRO 7995WX . 415120000 |========================== DuckDB 0.9.1 Benchmark: IMDB Seconds < Lower Is Better GPTshop.ai GH200 ..................... 92.08 |===================== Ampere Altra Max M128-30 ............. 142.91 |================================ HP Z6 G5 A - Threadripper PRO 7995WX . 101.32 |======================= DuckDB 0.9.1 Benchmark: TPC-H Parquet Seconds < Lower Is Better GPTshop.ai GH200 ..................... 148.76 |==================== Ampere Altra Max M128-30 ............. 238.83 |================================ HP Z6 G5 A - Threadripper PRO 7995WX . 113.87 |=============== PostgreSQL 16 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write TPS > Higher Is Better GPTshop.ai GH200 ..................... 54975 |=============================== Ampere Altra Max M128-30 ............. 58226 |================================= HP Z6 G5 A - Threadripper PRO 7995WX . 19313 |=========== PostgreSQL 16 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latency ms < Lower Is Better GPTshop.ai GH200 ..................... 18.23 |============ Ampere Altra Max M128-30 ............. 17.18 |=========== HP Z6 G5 A - Threadripper PRO 7995WX . 51.78 |================================= RawTherapee Total Benchmark Time Seconds < Lower Is Better GPTshop.ai GH200 ..................... 46.72 |======================= Ampere Altra Max M128-30 ............. 66.77 |================================= HP Z6 G5 A - Threadripper PRO 7995WX . 45.13 |====================== Stress-NG 0.16.04 Test: Matrix Math Bogo Ops/s > Higher Is Better GPTshop.ai GH200 ..................... 512759.08 |====================== Ampere Altra Max M128-30 ............. 682631.37 |============================= HP Z6 G5 A - Threadripper PRO 7995WX . 448155.05 |=================== Stress-NG 0.16.04 Test: Matrix 3D Math Bogo Ops/s > Higher Is Better GPTshop.ai GH200 ..................... 17483.02 |============================== Ampere Altra Max M128-30 ............. 5116.05 |========= HP Z6 G5 A - Threadripper PRO 7995WX . 9114.29 |================ Timed Gem5 Compilation 23.0.1 Time To Compile Seconds < Lower Is Better GPTshop.ai GH200 ..................... 180.62 |====================== Ampere Altra Max M128-30 ............. 265.15 |================================ HP Z6 G5 A - Threadripper PRO 7995WX . 149.75 |==================