SPECaccel(R)2023 Result NVIDIA Corporation Tesla H100 96GB MGX-GH200 accel2023 License: 9045 Test date: Oct-2023 Test sponsor: NVIDIA Corporation Hardware availability: Sep-2023 Tested by: NVIDIA Corporation Software availability: Nov-2023 Base Base Base Base Peak Peak Peak Peak Benchmarks Model Ref. Run Time Ratio Model Ref. Run Time Ratio -------------- ------ ------ --------- --------- ------ ------ --------- --------- 403.stencil LOP 440 146 3.01 S LOP 440 146 3.01 S 403.stencil LOP 440 146 3.01 * LOP 440 146 3.01 * 404.lbm LOP 455 60.6 7.51 * LOP 455 60.6 7.51 * 404.lbm LOP 455 60.5 7.52 S LOP 455 60.5 7.52 S 450.md LOP 600 190 3.16 * LOP 600 190 3.16 * 450.md LOP 600 190 3.16 S LOP 600 190 3.16 S 452.ep LOP 415 89.5 4.64 * LOP 415 89.5 4.64 * 452.ep LOP 415 89.4 4.64 S LOP 415 89.4 4.64 S 453.clvrleaf LOP 1000 91.1 11.0 * LOP 1000 91.1 11.0 * 453.clvrleaf LOP 1000 91.1 11.0 S LOP 1000 91.1 11.0 S 455.seismic LOP 780 136 5.74 * LOP 780 136 5.74 * 455.seismic LOP 780 136 5.74 S LOP 780 136 5.74 S 456.spF LOP 475 78.6 6.04 * LOP 475 78.6 6.04 * 456.spF LOP 475 78.5 6.05 S LOP 475 78.5 6.05 S 457.spC LOP 540 110 4.89 S LOP 540 110 4.89 S 457.spC LOP 540 111 4.89 * LOP 540 111 4.89 * 459.miniGhost LOP 590 197 2.99 S LOP 590 197 2.99 S 459.miniGhost LOP 590 198 2.98 * LOP 590 198 2.98 * 460.ilbdc LOP 555 101 5.48 S LOP 555 101 5.48 S 460.ilbdc LOP 555 101 5.48 * LOP 555 101 5.48 * 463.swim LOP 440 32.1 13.7 S LOP 440 32.1 13.7 S 463.swim LOP 440 32.2 13.7 * LOP 440 32.2 13.7 * 470.bt LOP 1055 114 9.29 S LOP 1055 114 9.29 S 470.bt LOP 1055 114 9.27 * LOP 1055 114 9.27 * ============================================================================================ 403.stencil LOP 440 146 3.01 * LOP 440 146 3.01 * 404.lbm LOP 455 60.6 7.51 * LOP 455 60.6 7.51 * 450.md LOP 600 190 3.16 * LOP 600 190 3.16 * 452.ep LOP 415 89.5 4.64 * LOP 415 89.5 4.64 * 453.clvrleaf LOP 1000 91.1 11.0 * LOP 1000 91.1 11.0 * 455.seismic LOP 780 136 5.74 * LOP 780 136 5.74 * 456.spF LOP 475 78.6 6.04 * LOP 475 78.6 6.04 * 457.spC LOP 540 111 4.89 * LOP 540 111 4.89 * 459.miniGhost LOP 590 198 2.98 * LOP 590 198 2.98 * 460.ilbdc LOP 555 101 5.48 * LOP 555 101 5.48 * 463.swim LOP 440 32.2 13.7 * LOP 440 32.2 13.7 * 470.bt LOP 1055 114 9.27 * LOP 1055 114 9.27 * SPECaccel 2023_base 5.74 SPECaccel 2023_peak 5.74 HARDWARE -------- CPU Name: Grace Max MHz.: 3100 Nominal: 2900 Enabled: 72 cores, 1 chip Orderable: 1 chip Cache L1: 64 KB I + 64 KB D on chip per core L2: 1 MB I+D on chip per core L3: 117 MB I+D on chip per chip Other: None Memory: 480 GB (1 x 480 GB LPDDR5) Storage: 1.8 TB NVMe Other: None Base Threads Run: 1 Min. Peak Threads: 1 Max. Peak Threads: 1 ACCELERATOR ----------- Accel Model Name: H100 96GB Accel Vendor: NVIDIA Corporation Accel Name: Tesla H100 96GB Type of Accel: GPU Accel Connection: NVLink-C2C Does Accel Use ECC: Yes Accel Description: Grace Hopper Superchip w/ 96GB device memory Accel Driver: NVIDIA UNIX Open Kernel Module for aarch64 525.105.17 SOFTWARE -------- OS: Ubuntu 22.04.3 LTS 6.2.0-1010-nvidia-64k Compiler: C/Fortran: Version 23.11 NVHPC SDK Firmware: NVIDIA 00010001 09/21/2023 File System: ext4 System State: Run level 3 (multi-user) Other: None Base Parallel Model: LOP Base Threads Run: 1 Peak Parallel Models: LOP Max. Peak Threads: 1 Min. Peak Threads: 1 Operating System Notes ---------------------- Shell stacksize set to unlimited via "limit stacksize unlimited" General Notes ------------- Environment variables set by runaccel before the start of the run: LD_LIBRARY_PATH = "/var/data0/sandbox/nvuser/SPECACCEL/nv239_libs" Set to the location of the NVHPC compiler runtime libraries. Platform Notes -------------- Sysinfo program /var/data0/sandbox/nvuser/mcolgrove/ACCELv2/bin/sysinfo Rev: r6622 of 2021-04-07 b1a7d5f8f71be5aff70a755cad7211a0 running on LegoCG1-96GB-QS-102 Tue Oct 17 01:53:46 2023 SUT (System Under Test) info as seen by some common utilities. For more information on this section, see https://www.spec.org/cpu2017/Docs/config.html#sysinfo From /proc/cpuinfo * * Did not identify cpu model. If you would * like to write your own sysinfo program, see * www.spec.org/cpu2017/config.html#sysinfo * * * 0 "physical id" tags found. Perhaps this is an older system, * or a virtualized system. Not attempting to guess how to * count chips/cores for this system. * 72 "processors" cores, siblings (Caution: counting these is hw and system dependent. The following excerpts from /proc/cpuinfo might not be reliable. Use with caution.) From lscpu from util-linux 2.37.2: Architecture: aarch64 CPU op-mode(s): 64-bit Byte Order: Little Endian CPU(s): 72 On-line CPU(s) list: 0-71 Vendor ID: ARM Model: 0 Thread(s) per core: 1 Core(s) per socket: 72 Socket(s): 1 Stepping: r0p0 Frequency boost: disabled CPU max MHz: 3429.0000 CPU min MHz: 81.0000 BogoMIPS: 2000.00 Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm ssbs sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh bti L1d cache: 4.5 MiB (72 instances) L1i cache: 4.5 MiB (72 instances) L2 cache: 72 MiB (72 instances) L3 cache: 114 MiB (1 instance) NUMA node(s): 9 NUMA node0 CPU(s): 0-71 NUMA node1 CPU(s): NUMA node2 CPU(s): NUMA node3 CPU(s): NUMA node4 CPU(s): NUMA node5 CPU(s): NUMA node6 CPU(s): NUMA node7 CPU(s): NUMA node8 CPU(s): Vulnerability Gather data sampling: Not affected Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Not affected Vulnerability Meltdown: Not affected Vulnerability Mmio stale data: Not affected Vulnerability Retbleed: Not affected Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl Vulnerability Spectre v1: Mitigation; __user pointer sanitization Vulnerability Spectre v2: Not affected Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Not affected From lscpu --cache: NAME ONE-SIZE ALL-SIZE WAYS TYPE LEVEL SETS PHY-LINE COHERENCY-SIZE L1d 64K 4.5M 4 Data 1 256 64 L1i 64K 4.5M 4 Instruction 1 256 64 L2 1M 72M 8 Unified 2 2048 64 L3 114M 114M 12 Unified 3 155648 64 From numactl --hardware WARNING: a numactl 'node' might or might not correspond to a physical chip. available: 9 nodes (0-8) node 0 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 node 0 size: 490308 MB node 0 free: 76379 MB node 1 cpus: node 1 size: 97280 MB node 1 free: 96966 MB node 2 cpus: node 2 size: 0 MB node 2 free: 0 MB node 3 cpus: node 3 size: 0 MB node 3 free: 0 MB node 4 cpus: node 4 size: 0 MB node 4 free: 0 MB node 5 cpus: node 5 size: 0 MB node 5 free: 0 MB node 6 cpus: node 6 size: 0 MB node 6 free: 0 MB node 7 cpus: node 7 size: 0 MB node 7 free: 0 MB node 8 cpus: node 8 size: 0 MB node 8 free: 0 MB node distances: node 0 1 2 3 4 5 6 7 8 0: 10 80 80 80 80 80 80 80 80 1: 80 10 255 255 255 255 255 255 255 2: 80 255 10 255 255 255 255 255 255 3: 80 255 255 10 255 255 255 255 255 4: 80 255 255 255 10 255 255 255 255 5: 80 255 255 255 255 10 255 255 255 6: 80 255 255 255 255 255 10 255 255 7: 80 255 255 255 255 255 255 10 255 8: 80 255 255 255 255 255 255 255 10 From /proc/meminfo MemTotal: 601690240 kB HugePages_Total: 0 Hugepagesize: 524288 kB /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor has performance /usr/bin/lsb_release -d Ubuntu 22.04.3 LTS From /etc/*release* /etc/*version* debian_version: bookworm/sid l4t-release: L4T_NAME="L4T Server" L4T_PRETTY_NAME="NVIDIA L4T Server" L4T_SWBUILD_DATE="2023-09-12-19-18-30" L4T_SWBUILD_VERSION="6.0.0-nvidia" L4T_COMMIT_ID="3244539" L4T_PLATFORM="" L4T_SERIAL_NUMBER="" os-release: PRETTY_NAME="Ubuntu 22.04.3 LTS" NAME="Ubuntu" VERSION_ID="22.04" VERSION="22.04.3 LTS (Jammy Jellyfish)" VERSION_CODENAME=jammy ID=ubuntu ID_LIKE=debian HOME_URL="https://www.ubuntu.com/" uname -a: Linux LegoCG1-96GB-QS-102 6.2.0-1010-nvidia-64k #10-Ubuntu SMP PREEMPT_DYNAMIC Wed Aug 30 06:23:50 UTC 2023 aarch64 aarch64 aarch64 GNU/Linux Kernel self-reported vulnerability status: gather_data_sampling: Not affected CVE-2018-12207 (iTLB Multihit): Not affected CVE-2018-3620 (L1 Terminal Fault): Not affected Microarchitectural Data Sampling: Not affected CVE-2017-5754 (Meltdown): Not affected mmio_stale_data: Not affected retbleed: Not affected CVE-2018-3639 (Speculative Store Bypass): Mitigation: Speculative Store Bypass disabled via prctl CVE-2017-5753 (Spectre variant 1): Mitigation: __user pointer sanitization CVE-2017-5715 (Spectre variant 2): Not affected CVE-2020-0543 (Special Register Buffer Data Sampling): Not affected CVE-2019-11135 (TSX Asynchronous Abort): Not affected run-level 3 Oct 14 00:07 SPEC is set to: /var/data0/sandbox/nvuser/mcolgrove/ACCELv2 Filesystem Type Size Used Avail Use% Mounted on /dev/nvme0n1p2 ext4 1.8T 1.5T 200G 89% / Cannot run dmidecode; consider saying (as root) chmod +s /usr/sbin/dmidecode BIOS: BIOS Vendor: NVIDIA BIOS Version: 00010001 BIOS Date: 20230921 (End of data from sysinfo program) Compiler Version Notes ---------------------- ============================================================================== C | 403.stencil(base) 404.lbm(base) 452.ep(base) 470.bt(base) ------------------------------------------------------------------------------ nvc Rel Dev-r238862 linuxarm64 target on aarch64 Linux -tp neoverse-v2 NVIDIA Compilers and Tools Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved. ------------------------------------------------------------------------------ ============================================================================== C | 457.spC(base) ------------------------------------------------------------------------------ nvc-Warning-Only small, tiny and large code models are allowed on AArch64. The code model large will be used. /usr/bin/ld: /usr/lib/aarch64-linux-gnu/crt1.o: in function `__wrap_main': (.text+0x38): undefined reference to `main' pgacclnk: child process exit status 1: /usr/bin/ld nvc Rel Dev-r238862 linuxarm64 target on aarch64 Linux -tp neoverse-v2 NVIDIA Compilers and Tools Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved. ------------------------------------------------------------------------------ ============================================================================== C | 403.stencil(base) 404.lbm(base) 452.ep(base) 470.bt(base) ------------------------------------------------------------------------------ nvc Rel Dev-r238862 linuxarm64 target on aarch64 Linux -tp neoverse-v2 NVIDIA Compilers and Tools Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved. ------------------------------------------------------------------------------ ============================================================================== C | 457.spC(base) ------------------------------------------------------------------------------ nvc-Warning-Only small, tiny and large code models are allowed on AArch64. The code model large will be used. /usr/bin/ld: /usr/lib/aarch64-linux-gnu/crt1.o: in function `__wrap_main': (.text+0x38): undefined reference to `main' pgacclnk: child process exit status 1: /usr/bin/ld nvc Rel Dev-r238862 linuxarm64 target on aarch64 Linux -tp neoverse-v2 NVIDIA Compilers and Tools Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved. ------------------------------------------------------------------------------ ============================================================================== Fortran | 450.md(base) 455.seismic(base) 456.spF(base) 460.ilbdc(base) | 463.swim(base) ------------------------------------------------------------------------------ nvfortran Rel Dev-r238862 linuxarm64 target on aarch64 Linux -tp neoverse-v2 NVIDIA Compilers and Tools Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved. ------------------------------------------------------------------------------ ============================================================================== Fortran, C | 453.clvrleaf(base) 459.miniGhost(base) ------------------------------------------------------------------------------ nvfortran Rel Dev-r238862 linuxarm64 target on aarch64 Linux -tp neoverse-v2 NVIDIA Compilers and Tools Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved. nvc Rel Dev-r238862 linuxarm64 target on aarch64 Linux -tp neoverse-v2 NVIDIA Compilers and Tools Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved. ------------------------------------------------------------------------------ Base Compiler Invocation ------------------------ C benchmarks: nvc Fortran benchmarks: nvfortran Benchmarks using both Fortran and C: nvfortran nvc Base Portability Flags ---------------------- 403.stencil: -DSPEC_NO_NOTHING 457.spC: -mcmodel=medium -Wl,--no-relax Base Optimization Flags ----------------------- C benchmarks: -Ofast -mp=gpu -Mfprelaxed -Mstack_arrays -static-nvidia -march=neoverse-v2 Fortran benchmarks: -Ofast -mp=gpu -Mfprelaxed -Mstack_arrays -static-nvidia -march=neoverse-v2 Benchmarks using both Fortran and C: 453.clvrleaf: -Ofast -mp=gpu -Mfprelaxed -Mstack_arrays -static-nvidia -march=neoverse-v2 459.miniGhost: -Mnomain -Ofast -mp=gpu -Mfprelaxed -Mstack_arrays -static-nvidia -march=neoverse-v2 Peak Optimization Flags ----------------------- C benchmarks: 403.stencil: basepeak = yes 404.lbm: basepeak = yes 452.ep: basepeak = yes 457.spC: basepeak = yes 470.bt: basepeak = yes Fortran benchmarks: 450.md: basepeak = yes 455.seismic: basepeak = yes 456.spF: basepeak = yes 460.ilbdc: basepeak = yes 463.swim: basepeak = yes Benchmarks using both Fortran and C: 453.clvrleaf: basepeak = yes 459.miniGhost: basepeak = yes The flags file that was used to format this result can be browsed at http://www.spec.org/accel2023/flags/nv2023_flags_v2_arm.html You can also download the XML flags source by saving the following link: http://www.spec.org/accel2023/flags/nv2023_flags_v2_arm.xml SPECaccel is a registered trademark of the Standard Performance Evaluation Corporation. All other brand and product names appearing in this result are trademarks or registered trademarks of their respective holders. ------------------------------------------------------------------------------------------------------------ For questions about this result, please contact the tester. For other inquiries, please contact info@spec.org. Copyright 2023 Standard Performance Evaluation Corporation Tested with SPECaccel2023 v2.0.17 on 2023-10-16 21:53:46-0400. Report generated on 2023-12-06 13:07:19 by accel2023 ASCII formatter v112. Originally published on 2023-11-08.