SPEC® ACCEL™ OCL Result

Copyright 2015-2020 Standard Performance Evaluation Corporation

Lenovo Global Technology

NVIDIA Tesla V100S-PCIE-16GB

ThinkSystem SR860 V2

SPECaccel_ocl_base = 12.2 

SPECaccel_ocl_peak = 13.5 

ACCEL license: 28 Test date: Aug-2020
Test sponsor: Lenovo Global Technology Hardware Availability: Oct-2020
Tested by: Lenovo Global Technology Software Availability: Oct-2020
Benchmark results graph
Hardware
CPU Name: Intel Xeon Platinum 8380H
CPU Characteristics: Intel Turbo Boost Technology up to 4.3 GHz
CPU MHz: 2900
CPU MHz Maximum: 4300
FPU: Integrated
CPU(s) enabled: 112 cores, 4 chips, 28 cores/chip
CPU(s) orderable: 2,4 chips
Primary Cache: 32 KB I + 32 KB D on chip per core
Secondary Cache: 1 MB I+D on chip per core
L3 Cache: 39424 KB I+D on chip per chip
Other Cache: None
Memory: 1536 GB (48 x 32 GB 2Rx8 PC4-3200AA-R)
Disk Subsystem: 1 x 1TB SATA 2.5" SSD
Other Hardware: None
Accelerator
Accel Model Name: Tesla V100S
Accel Vendor: NVIDIA Corporation
Accel Name: NVIDIA Tesla V100S-PCIE-16GB
Type of Accel: GPU
Accel Connection: PCIe 3.0 16x
Does Accel Use ECC: Yes
Accel Description: NVIDIA Tesla V100S-PCIE-16GB
Accel Driver: NVIDIA UNIX x86_64 Kernel Module 450.51.06
Software
Operating System: SUSE Linux Enterprise Server 15 SP2
5.3.18-22-default
Compiler: Nvidia HPC SDK Release 20.5
File System: xfs
System State: Run level 3
Other Software: CUDA 11.0 SDK

Results Table

Benchmark Base Peak
Seconds Ratio Seconds Ratio Seconds Ratio Seconds Ratio Seconds Ratio Seconds Ratio
Results appear in the order in which they were run. Bold underlined text indicates a median measurement.
101.tpacf 5.88  18.2   5.90  18.1   5.87  18.2   5.55  19.3   5.54  19.3   5.52  19.4  
103.stencil 6.43  19.4   6.46  19.4   6.41  19.5   6.43  19.4   6.46  19.4   6.41  19.5  
104.lbm 5.61  20.0   5.62  19.9   5.62  19.9   5.62  19.9   5.64  19.9   5.63  19.9  
110.fft 6.77  16.4   6.77  16.4   6.75  16.4   6.77  16.4   6.77  16.4   6.75  16.4  
112.spmv 15.7   9.34  15.8   9.30  15.8   9.30  15.8   9.31  15.8   9.31  15.8   9.32 
114.mriq 3.14  34.7   3.15  34.6   3.16  34.5   3.14  34.7   3.15  34.6   3.16  34.5  
116.histo 49.4   2.31  48.3   2.36  56.6   2.02  49.4   2.31  48.3   2.36  56.6   2.02 
117.bfs 20.0   5.86  20.0   5.86  19.9   5.87  19.7   5.93  19.8   5.91  19.8   5.92 
118.cutcp 5.11  19.4   5.13  19.3   5.11  19.4   5.11  19.4   5.13  19.3   5.11  19.4  
120.kmeans 36.4   2.74  37.7   2.65  36.8   2.72  36.2   2.76  37.7   2.65  36.4   2.74 
121.lavamd 5.17  21.1   5.16  21.1   5.14  21.2   5.17  21.1   5.16  21.1   5.14  21.2  
122.cfd 9.72  13.0   9.69  13.0   9.68  13.0   9.63  13.1   9.60  13.1   9.62  13.1  
123.nw 13.9   8.28  13.9   8.29  13.9   8.27  13.9   8.28  13.9   8.29  13.9   8.27 
124.hotspot 7.45  15.3   7.48  15.2   7.45  15.3   7.45  15.3   7.48  15.2   7.45  15.3  
125.lud 11.8   10.1   11.8   10.0   11.8   10.1   8.22  14.5   8.25  14.4   8.20  14.5  
126.ge 5.67  27.3   5.68  27.3   5.67  27.3   1.38  112     1.39  112     1.39  112    
127.srad 10.6   10.8   10.6   10.8   10.6   10.8   10.6   10.8   10.6   10.8   10.6   10.8  
128.heartwall 9.49  11.2   9.46  11.2   9.42  11.2   9.49  11.2   9.46  11.2   9.42  11.2  
140.bplustree 7.26  14.9   7.26  14.9   7.25  14.9   7.26  14.9   7.26  14.9   7.25  14.9  

Submit Notes

The config file option 'submit' was used.

Platform Notes

 Sysinfo program /home/ACCEL1.3/Docs/sysinfo
 $Rev: 6965 $ $Date:: 2015-04-21 #$ c05a7f14b1b1765e3fe1df68447e8a35
 running on Narvi152 Tue Aug 18 13:36:52 2020

 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 8380H CPU @ 2.90GHz
       4 "physical id"s (chips)
       112 "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 : 28
       siblings  : 28
       physical 0: cores 0 1 2 3 4 5 6 8 9 10 11 12 13 14 16 17 18 19 20 21 22 24
       25 26 27 28 29 30
       physical 1: cores 0 1 2 3 4 5 6 8 9 10 11 12 13 14 16 17 18 19 20 21 22 24
       25 26 27 28 29 30
       physical 2: cores 0 1 2 3 4 5 6 8 9 10 11 12 13 14 16 17 18 19 20 21 22 24
       25 26 27 28 29 30
       physical 3: cores 0 1 2 3 4 5 6 8 9 10 11 12 13 14 16 17 18 19 20 21 22 24
       25 26 27 28 29 30
    cache size : 39424 KB

 From /proc/meminfo
    MemTotal:       1584974840 kB
    HugePages_Total:       0
    Hugepagesize:       2048 kB

 From /etc/*release* /etc/*version*
    os-release:
       NAME="SLES"
       VERSION="15-SP2"
       VERSION_ID="15.2"
       PRETTY_NAME="SUSE Linux Enterprise Server 15 SP2"
       ID="sles"
       ID_LIKE="suse"
       ANSI_COLOR="0;32"
       CPE_NAME="cpe:/o:suse:sles:15:sp2"

 uname -a:
    Linux Narvi152 5.3.18-22-default #1 SMP Wed Jun 3 12:16:43 UTC 2020 (720aeba)
    x86_64 x86_64 x86_64 GNU/Linux

 run-level 3 Aug 17 15:54

 SPEC is set to: /home/ACCEL1.3
    Filesystem     Type  Size  Used Avail Use% Mounted on
    /dev/sda4      xfs   490G   47G  443G  10% /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.

   BIOS Lenovo M5E103N-1.00 08/07/2020
   Memory:
    48x Samsung M393A4G43AB3-CWE 32 GB 2 rank 3200 MT/s

 (End of data from sysinfo program)

General Notes

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 11.0.197   OpenCL Device #0: Tesla V100S-PCIE-32GB, v 450.51.06 

C++ benchmarks:

 OpenCL Platform: NVIDIA CUDA, OpenCL 1.2 CUDA 11.0.197   OpenCL Device #0: Tesla V100S-PCIE-32GB, v 450.51.06 

Base Compiler Invocation

C benchmarks:

 nvc 

C++ benchmarks:

 nvc++ 

Base Portability Flags

116.histo:  -DSPEC_LOCAL_MEMORY_HEADROOM=1 

Base Optimization Flags

C benchmarks:

 -fast   -Mfprelaxed 

C++ benchmarks:

 -fast   -Mfprelaxed 

Base Other Flags

C benchmarks:

 -I/usr/local/cuda-11.0/targets/x86_64-linux/include   -L/usr/local/cuda-11.0/lib64   -lOpenCL 

C++ benchmarks:

 -I/usr/local/cuda-11.0/targets/x86_64-linux/include   -L/usr/local/cuda-11.0/lib64   -lOpenCL 

Peak Runtime Environment

C benchmarks:

 OpenCL Platform: NVIDIA CUDA, OpenCL 1.2 CUDA 11.0.197   OpenCL Device #0: Tesla V100S-PCIE-32GB, v 450.51.06 

C++ benchmarks:

 OpenCL Platform: NVIDIA CUDA, OpenCL 1.2 CUDA 11.0.197   OpenCL Device #0: Tesla V100S-PCIE-32GB, v 450.51.06 

Peak Compiler Invocation

C benchmarks:

 nvc 

C++ benchmarks:

 nvc++ 

Peak Portability Flags

116.histo:  -DSPEC_LOCAL_MEMORY_HEADROOM=1 

Peak Optimization Flags

C benchmarks:

110.fft:  basepeak = yes 
114.mriq:  basepeak = yes 
116.histo:  basepeak = yes 
117.bfs:  -fast   -Mfprelaxed   -DSPEC_ACCEL_WG_SIZE_0_0=64   -DSPEC_ACCEL_WG_SIZE_1_0=64 
118.cutcp:  basepeak = yes 
121.lavamd:  basepeak = yes 
124.hotspot:  basepeak = yes 
127.srad:  basepeak = yes 
128.heartwall:  basepeak = yes 
140.bplustree:  basepeak = yes 

C++ benchmarks:

101.tpacf:  -fast   -Mfprelaxed   -DSPEC_ACCEL_WG_SIZE_0_0=1024 
103.stencil:  basepeak = yes 
104.lbm:  -fast   -Mfprelaxed   -DSPEC_ACCEL_WG_SIZE_0_0=32   -DSPEC_ACCEL_WG_SIZE_0_1=1   -DSPEC_ACCEL_WG_SIZE_0_2=1 
112.spmv:  -fast   -Mfprelaxed   -DSPEC_ACCEL_WG_SIZE_0_0=96 
120.kmeans:  -fast   -Mfprelaxed   -DSPEC_ACCEL_WG_SIZE_0_0=288 
122.cfd:  -fast   -Mfprelaxed   -DSPEC_ACCEL_WG_SIZE_3_0=288 
123.nw:  basepeak = yes 
125.lud:  -fast   -Mfprelaxed   -DSPEC_ACCEL_WG_SIZE_0_0=32 
126.ge:  -fast   -Mfprelaxed   -DSPEC_ACCEL_WG_SIZE_0_0=512   -DSPEC_ACCEL_WG_SIZE_1_0=1   -DSPEC_ACCEL_WG_SIZE_1_1=512 

Peak Other Flags

C benchmarks:

 -I/usr/local/cuda-11.0/targets/x86_64-linux/include   -L/usr/local/cuda-11.0/lib64   -lOpenCL 

C++ benchmarks:

 -I/usr/local/cuda-11.0/targets/x86_64-linux/include   -L/usr/local/cuda-11.0/lib64   -lOpenCL 

The flags file that was used to format this result can be browsed at
https://www.spec.org/accel/flags/nv2020_flags.html.

You can also download the XML flags source by saving the following link:
https://www.spec.org/accel/flags/nv2020_flags.xml.