SPEC(R) ACCEL(TM) OCL Summary KTNF NVIDIA Tesla A100-PCIE-40GB KR580S2 Test Sponsor: Telecommunications Technology Association Wed Jun 8 04:20:54 2022 ACCEL License: 068A Test date: Jun-2022 Test sponsor: Telecommunications Technology Association Hardware availability: May 2022 Tested by: Telecommunications Technology Association Software availability: Jul 2021 Base Base Base Peak Peak Peak Benchmarks Ref. Run Time Ratio Ref. Run Time Ratio -------------- ------ --------- --------- ------ --------- --------- 101.tpacf 107 6.37 16.8 * 101.tpacf 107 6.39 16.8 S 101.tpacf 107 6.36 16.8 S 103.stencil 125 4.95 25.3 S 103.stencil 125 4.97 25.2 S 103.stencil 125 4.96 25.2 * 104.lbm 112 4.41 25.4 S 104.lbm 112 4.40 25.5 * 104.lbm 112 4.38 25.6 S 110.fft 111 4.59 24.2 S 110.fft 111 4.60 24.1 S 110.fft 111 4.60 24.1 * 112.spmv 147 12.4 11.9 S 112.spmv 147 12.4 11.9 S 112.spmv 147 12.4 11.9 * 114.mriq 109 2.74 39.7 S 114.mriq 109 2.73 39.9 S 114.mriq 109 2.74 39.8 * 116.histo 114 46.9 2.43 S 116.histo 114 46.1 2.47 S 116.histo 114 46.8 2.43 * 117.bfs 117 6.22 18.8 S 117.bfs 117 6.21 18.8 * 117.bfs 117 6.17 18.9 S 118.cutcp 99 3.48 28.4 S 118.cutcp 99 3.49 28.4 S 118.cutcp 99 3.49 28.4 * 120.kmeans 100 41.5 2.41 * 120.kmeans 100 41.5 2.41 S 120.kmeans 100 41.5 2.41 S 121.lavamd 109 4.23 25.8 S 121.lavamd 109 4.24 25.7 * 121.lavamd 109 4.27 25.5 S 122.cfd 126 8.10 15.6 S 122.cfd 126 8.12 15.5 * 122.cfd 126 8.14 15.5 S 123.nw 115 14.4 7.98 S 123.nw 115 14.4 7.98 * 123.nw 115 14.2 8.08 S 124.hotspot 114 5.71 19.9 * 124.hotspot 114 5.71 20.0 S 124.hotspot 114 5.75 19.8 S 125.lud 119 9.43 12.6 * 125.lud 119 9.38 12.7 S 125.lud 119 9.67 12.3 S 126.ge 155 6.27 24.7 * 126.ge 155 6.27 24.7 S 126.ge 155 6.27 24.7 S 127.srad 114 8.16 14.0 * 127.srad 114 8.15 14.0 S 127.srad 114 8.17 14.0 S 128.heartwall 106 8.74 12.1 S 128.heartwall 106 8.75 12.1 S 128.heartwall 106 8.74 12.1 * 140.bplustree 108 6.56 16.5 S 140.bplustree 108 6.56 16.5 * 140.bplustree 108 6.56 16.5 S ============================================================================== 101.tpacf 107 6.37 16.8 * 103.stencil 125 4.96 25.2 * 104.lbm 112 4.40 25.5 * 110.fft 111 4.60 24.1 * 112.spmv 147 12.4 11.9 * 114.mriq 109 2.74 39.8 * 116.histo 114 46.8 2.43 * 117.bfs 117 6.21 18.8 * 118.cutcp 99 3.49 28.4 * 120.kmeans 100 41.5 2.41 * 121.lavamd 109 4.24 25.7 * 122.cfd 126 8.12 15.5 * 123.nw 115 14.4 7.98 * 124.hotspot 114 5.71 19.9 * 125.lud 119 9.43 12.6 * 126.ge 155 6.27 24.7 * 127.srad 114 8.16 14.0 * 128.heartwall 106 8.74 12.1 * 140.bplustree 108 6.56 16.5 * SPECaccel_ocl_base 15.0 SPECaccel_ocl_peak Not Run HARDWARE -------- CPU Name: Intel Xeon Platinum 8330 CPU Characteristics: Intel Turbo Boost Technology up to 3.4 GHz CPU MHz: 2300 CPU MHz Maximum: 3400 FPU: -- CPU(s) enabled: 80 cores, 2 chips, 40 cores/chip, 2 threads/core CPU(s) orderable: 1,2 chips Primary Cache: 32 KB I + 32 KB D on chip per core Secondary Cache: 1024 KB I+D on chip per core L3 Cache: 60 MB I+D on chip per chip Other Cache: None Memory: 512 GB (8 x 64GB 2Rx4 PC4-3200AA-R) Disk Subsystem: 2 x Samsung SSD 860 PRO 512GB (RAID 0) Other Hardware: LSI MegaRAID SAS-3 3108 ACCELERATOR ----------- Accel Model Name: Tesla A100-PCIE-40GB Accel Vendor: NVIDIA Accel Name: NVIDIA Tesla A100-PCIE-40GB Type of Accel: GPU Accel Connection: PCIe 4.0 16x Does Accel Use ECC: Yes Accel Description: NVIDIA Tesla A100-PCIE-40GB Accel Driver: NVIDIA Driver Version 460.106.00 SOFTWARE -------- Operating System: CentOS Linux release 7.9.2009 (Core) 3.10.0-1160.66.1.el7.x86_64 Compiler: GCC version 4.8.5 20150623 File System: xfs System State: Multi-user, run level 3 Other Software: NVIDIA CUDA 11.2 Platform Notes -------------- Sysinfo program /home/tta701/accel-1.4/Docs/sysinfo $Rev: 6965 $ $Date:: 2015-04-21 #$ c05a7f14b1b1765e3fe1df68447e8a35 running on kr580s2 Wed Jun 8 17:20:55 2022 This section contains SUT (System Under Test) info as seen by some common utilities. To remove or add to this section, see: http://www.spec.org/accel/Docs/config.html#sysinfo From /proc/cpuinfo model name : Intel(R) Xeon(R) Platinum 8380 CPU @ 2.30GHz 2 "physical id"s (chips) 160 "processors" cores, siblings (Caution: counting these is hw and system dependent. The following excerpts from /proc/cpuinfo might not be reliable. Use with caution.) cpu cores : 40 siblings : 80 physical 0: cores 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 physical 1: cores 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 cache size : 61440 KB From /proc/meminfo MemTotal: 527825800 kB HugePages_Total: 0 Hugepagesize: 2048 kB From /etc/*release* /etc/*version* centos-release: CentOS Linux release 7.9.2009 (Core) centos-release-upstream: Derived from Red Hat Enterprise Linux 7.8 (Source) os-release: NAME="CentOS Linux" VERSION="7 (Core)" ID="centos" ID_LIKE="rhel fedora" VERSION_ID="7" PRETTY_NAME="CentOS Linux 7 (Core)" ANSI_COLOR="0;31" CPE_NAME="cpe:/o:centos:centos:7" redhat-release: CentOS Linux release 7.9.2009 (Core) system-release: CentOS Linux release 7.9.2009 (Core) system-release-cpe: cpe:/o:centos:centos:7 uname -a: Linux kr580s2 3.10.0-1160.66.1.el7.x86_64 #1 SMP Wed May 18 16:02:34 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux run-level 3 Jan 16 08:46 SPEC is set to: /home/tta701/accel-1.4 Filesystem Type Size Used Avail Use% Mounted on /dev/mapper/centos-home xfs 898G 24G 875G 3% /home Additional information from dmidecode: Warning: Use caution when you interpret this section. The 'dmidecode' program reads system data which is "intended to allow hardware to be accurately determined", but the intent may not be met, as there are frequent changes to hardware, firmware, and the "DMTF SMBIOS" standard. (End of data from sysinfo program) General Notes ------------- ================================================================================================ Spectre and Meltdown Yes: The test sponsor attests, as of date of publication, that CVE-2017-5754 (Meltdown) is mitigated in the system as tested and documented. Yes: The test sponsor attests, as of date of publication, that CVE-2017-5753 (Spectre variant 1) is mitigated in the system as tested and documented. Yes: The test sponsor attests, as of date of publication, that CVE-2017-5715 (Spectre variant 2) is mitigated in the system as tested and documented. Base Runtime Environment ------------------------ C benchmarks: OpenCL Platform: NVIDIA CUDA, OpenCL 1.2 CUDA 11.2.162 OpenCL Device #0: A100-PCIE-40GB, v 460.106.00 C++ benchmarks: OpenCL Platform: NVIDIA CUDA, OpenCL 1.2 CUDA 11.2.162 OpenCL Device #0: A100-PCIE-40GB, v 460.106.00 Base Compiler Invocation ------------------------ C benchmarks: gcc C++ benchmarks: g++ Base Portability Flags ---------------------- 116.histo: -DSPEC_LOCAL_MEMORY_HEADROOM=2 Base Optimization Flags ----------------------- C benchmarks: -O2 -I/usr/local/cuda/include -L/usr/local/cuda/lib64 -lOpenCL C++ benchmarks: -O2 -I/usr/local/cuda/include -L/usr/local/cuda/lib64 -lOpenCL The flags file that was used to format this result can be browsed at https://www.spec.org/accel/flags/gcc_flags.20211028.html You can also download the XML flags source by saving the following link: https://www.spec.org/accel/flags/gcc_flags.20211028.xml SPEC and SPEC ACCEL are registered trademarks or trademarks 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 webmaster@spec.org. Copyright 2015-2022 Standard Performance Evaluation Corporation Tested with SPEC ACCEL v1.4. Report generated on Thu Jun 30 11:44:09 2022 by ACCEL ASCII formatter v1290. Originally published on 29 June 2022.