SPEC® OMPG2012 Result

Copyright 2012-2023 Standard Performance Evaluation Corporation

Lenovo Global Technology

SPECompG_base2012 = 52.7

SPECompG_energy_base2012 = 26.4

ThinkSystem SR655V3 (AMD EPYC 9754, 2.25GHz)

SPECompG_peak2012 = 53.4

SPECompG_energy_peak2012 = 27.0

OMP2012 license: 28 Test date: Jul-2023
Test sponsor: Lenovo Global Technology Hardware Availability: Aug-2023
Tested by: Lenovo Global Technology Software Availability: Aug-2023
Benchmark results graph
Hardware
CPU Name: AMD EPYC 9754
CPU Characteristics: Max Boost Clock up to 3.1 GHz
CPU MHz: 2250
CPU MHz Maximum: 3100
FPU: Integrated
CPU(s) enabled: 128 cores, 1 chip, 128 cores/chip, 2 threads/core
CPU(s) orderable: 1 chip
Primary Cache: 32 KB I + 32 KB D on chip per core
Secondary Cache: 1 MB I+D on chip per core
L3 Cache: 256 MB I+D on chip per chip, 32 MB shared 8 cores
Other Cache: None
Memory: 768 GB ( 12 x 64 GB 2Rx4 PC5-4800B-R)
Disk Subsystem: 1 x 1 TB SATA Hard Drive
Other Hardware: None
Base Threads Run: 256
Minimum Peak Threads: 128
Maximum Peak Threads: 256
Software
Operating System: SUSE Linux Enterprise for High-Performance
Computing 15
SP4 (x86_64), Kernel 5.14.21-150400.22-default
Compiler: C/C++/Fortran: Version 2022.2.0.191 of Intel
oneAPI oneAPI DPC/C++
Auto Parallel: No
File System: xfs
System State: Multi-user, run level 3
Base Pointers: 64-bit
Peak Pointers: 64-bit
Other Software: None
Power
Power Supply: 1 x 2400 W
Power Supply Details: Lenovo P/N:SP57A14720
Max. Power (W): 817.6
Idle Power (W): 212.03
Min. Temperature (C): 20.19
Power Analyzer
Power Analyzer: Power Analyzer
Hardware Vendor: Yokogawa
Model: WT310E
Serial Number: C3TE30007E
Input Connection: USB
Metrology Institute: ETC (Taiwan Testing and Certification Center)
Calibration By: ETC
Calibration Label: 23-05-BAC-245-01L
Calibration Date: 05.16.2023
PTDaemon Version: 1.8.0 (3505e615; 2016-01-20)
Setup Description: connected to the single power supply
Current Ranges Used: 10.0A
Voltage Range Used: 300V
Temperature Meter
Temperature Meter: Temperature Meter
Hardware Vendor: Digi
Model: Watchport/H
Serial Number: w94638774
Input Connection: USB
PTDaemon Version: 1.8.0 (3505e615; 2016-01-20)
Setup Description: positioned in front of intake fan

Base Results Table

Benchmark Threads Seconds Ratio Energy
(kJ)
Maximum
Power
Average
Power
Energy
Ratio
Seconds Ratio Energy
(kJ)
Maximum
Power
Average
Power
Energy
Ratio
Seconds Ratio Energy
(kJ)
Maximum
Power
Average
Power
Energy
Ratio
Results appear in the order in which they were run. Bold underlined text indicates a median measurement.
350.md 256 44.3  105    28.5  709   643   44.1  45.5  102    29.7  672   654   42.2  45.4  102    29.1  661   641   43.1 
351.bwaves 256 130    34.8  76.8  612   589   17.5  130    34.7  76.8  614   589   17.5  130    34.9  76.6  615   591   17.5 
352.nab 256 83.4  46.6  48.7  652   584   22.1  83.4  46.7  49.1  659   589   21.9  83.4  46.6  49.8  728   597   21.6 
357.bt331 256 75.5  62.7  48.1  740   637   30.4  75.4  62.9  48.0  740   637   30.5  75.5  62.8  47.7  651   631   30.7 
358.botsalgn 256 81.2  53.5  42.4  533   522   30.2  81.2  53.6  42.5  536   524   30.1  81.1  53.6  42.8  536   528   29.9 
359.botsspar 256 139    37.8  92.8  694   669   17.0  139    37.9  93.0  699   671   16.9  139    37.8  93.2  698   672   16.9 
360.ilbdc 256 131    27.1  72.0  603   549   15.3  131    27.1  72.3  602   551   15.3  131    27.2  72.1  603   551   15.3 
362.fma3d 256 106    35.9  59.1  596   558   19.1  106    35.7  59.4  600   559   19.0  106    35.9  58.9  595   556   19.2 
363.swim 256 171    26.4  85.3  503   497   16.2  172    26.4  85.4  503   498   16.2  172    26.4  85.5  500   498   16.2 
367.imagick 256 74.8  94.0  51.4  778   687   41.9  74.5  94.4  51.0  751   685   42.2  74.8  94.0  51.9  794   694   41.4 
370.mgrid331 256 183    24.1  98.5  579   538   13.9  183    24.1  98.8  579   539   13.8  183    24.1  98.7  579   538   13.8 
371.applu331 256 53.2  114    33.0  695   621   56.3  53.4  114    33.2  704   622   56.0  53.1  114    33.7  703   635   55.2 
372.smithwa 256 40.5  132    26.1  689   645   70.2  40.5  132    26.6  723   657   68.9  40.5  132    26.8  714   661   68.5 
376.kdtree 256 66.0  68.2  46.7  793   708   29.7  65.6  68.6  46.1  798   702   30.1  65.8  68.4  46.2  795   702   30.0 

Peak Results Table

Benchmark Threads Seconds Ratio Energy
(kJ)
Maximum
Power
Average
Power
Energy
Ratio
Seconds Ratio Energy
(kJ)
Maximum
Power
Average
Power
Energy
Ratio
Seconds Ratio Energy
(kJ)
Maximum
Power
Average
Power
Energy
Ratio
Results appear in the order in which they were run. Bold underlined text indicates a median measurement.
350.md 256 45.6  102    29.3  663   644   42.8  45.4  102    29.1  660   642   43.1  44.7  103    28.8  662   643   43.6 
351.bwaves 128 128    35.3  72.6  595   566   18.5  129    35.0  73.1  593   565   18.3  129    35.1  73.5  609   569   18.2 
352.nab 256 83.4  46.7  48.8  684   585   22.1  83.4  46.6  49.3  710   591   21.9  83.4  46.7  48.9  685   587   22.0 
357.bt331 256 74.6  63.6  47.9  751   642   30.6  74.7  63.4  48.2  763   645   30.4  75.2  63.0  48.7  747   647   30.1 
358.botsalgn 256 81.2  53.5  43.2  542   532   29.6  81.2  53.5  43.2  542   532   29.6  81.2  53.5  42.8  535   527   29.9 
359.botsspar 256 139    37.7  93.4  702   670   16.9  139    37.7  93.6  699   672   16.8  139    37.8  93.6  698   673   16.8 
360.ilbdc 128 131    27.2  69.7  589   534   15.8  131    27.3  69.7  586   534   15.8  131    27.3  69.6  587   533   15.9 
362.fma3d 256 106    35.9  59.0  599   557   19.1  105    36.1  58.8  600   558   19.2  107    35.6  59.1  595   553   19.1 
363.swim 128 169    26.8  82.5  496   488   16.7  169    26.8  82.5  489   488   16.7  169    26.8  82.7  504   489   16.7 
367.imagick 128 68.3  103    46.3  818   679   46.4  68.5  103    46.4  814   677   46.4  68.6  103    46.3  812   675   46.5 
370.mgrid331 128 173    25.6  90.0  540   520   15.2  173    25.6  90.0  562   521   15.2  173    25.6  90.1  561   521   15.2 
371.applu331 256 52.4  116    34.3  754   654   54.2  52.4  116    33.3  731   636   55.8  52.6  115    34.0  740   645   54.8 
372.smithwa 256 40.5  132    26.1  692   644   70.3  40.5  132    25.7  660   635   71.4  40.5  132    25.5  666   631   71.8 
376.kdtree 256 65.9  68.3  46.1  783   700   30.0  65.8  68.4  45.6  771   693   30.4  65.9  68.3  46.6  797   708   29.7 

Platform Notes

 Sysinfo program /home/omp2012/Docs/sysinfo
 Revision 563 of 2016-06-10 (097295389cf6073d8c3b03fa376740a5)
 running on bergamoA21P Sat Jul  8 10:57:21 2023

 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/omp2012/Docs/config.html#sysinfo

 From /proc/cpuinfo
    model name : AMD EPYC 9754 128-Core Processor
       1 "physical id"s (chips)
       256 "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 : 128
       siblings  : 256
       physical 0: cores 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 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 72 73 74 75 76 77 78 79 80 81 82 83
       84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105
       106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123
       124 125 126 127
    cache size : 1024 KB

 From /proc/meminfo
    MemTotal:       792101408 kB
    HugePages_Total:       0
    Hugepagesize:       2048 kB

 From /etc/*release* /etc/*version*
    os-release:
       NAME="SLE_HPC"
       VERSION="15-SP4"
       VERSION_ID="15.4"
       PRETTY_NAME="SUSE Linux Enterprise High Performance Computing 15 SP4"
       ID="sle_hpc"
       ID_LIKE="suse"
       ANSI_COLOR="0;32"
       CPE_NAME="cpe:/o:suse:sle_hpc:15:sp4"

 uname -a:
    Linux bergamoA21P 5.14.21-150400.22-default #1 SMP PREEMPT_DYNAMIC Wed May 11
    06:57:18 UTC 2022 (49db222) x86_64 x86_64 x86_64 GNU/Linux

 run-level 3 Apr 29 20:00

 SPEC is set to: /home/omp2012
    Filesystem     Type  Size  Used Avail Use% Mounted on
    /dev/sda3      xfs   413G   54G  359G  13% /var/tmp
 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 KAE111I-2.10 05/06/2023
   Memory:
    12x Samsung M321R8GA0BB0-CQKEG 64 GB 2 rank 4800 MT/s

 (End of data from sysinfo program)

General Notes

===============================================================================
 General OMP Library Settings
   OMP_DYNAMIC		= FALSE
   OMP_THREADS		= 256
   KMP_SCHEDULE	= static
   KMP_LIBRARY		= turnaround
   KMP_STACKSIZE	= 768M
   KMP_BLOCKTIME	= infinite
   KMP_AFFINITY	= granularity=fine,proclist=[0-7,8-15,16-23,24-31,32-39,
   40-47,48-55,56-63,64-71,72-79,80-87,88-95,96-103,104-111,112-119,120-127,
   128-135,136-143,144-151,152-159,160-167,168-175,176-183,184-191,192-199,
   200-207,208-215,216-223,224-231,232-239,240-247,248-255],explicit
===============================================================================
 uEFI Setting notes:
  Choose "Maximum Performance" operating mode and changed to "Custom" operating mode.
  Below items also configured:
   -
   - NUMA Nodes per Socket = NPS2
   - CPPC  = Disabled
   - DRAM Scrub Time = Disabled
   -
   -
   -

==============================================
	Yes: The test sponsor attests, as of date of publication, the CVE-2017-5754 (Meltdown)
	is mitigated in the system as tested and documented.

	Yes: The test sponsor attests, as of date of publication, the 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-2017-5754 (Spectre variant 2)
	is mitigated in the system as tested and documented.
=============================================
 OS tunning:
	ulimit -s unlimited

Base Compiler Invocation

C benchmarks:

 icx 

C++ benchmarks:

 icpx 

Fortran benchmarks:

 ifx 

Base Portability Flags

350.md:  -FR 
357.bt331:  -mcmodel=medium 
363.swim:  -mcmodel=medium 
367.imagick:  -std=c99 

Base Optimization Flags

C benchmarks:

 -Ofast   -fopenmp   -march=core-avx2   -fma   -ipo    -ansi-alias   -fp-model fast=2   -qno-opt-multiple-gather-scatter-by-shuffles   -qopt-zmm-usage=high   -ffast-math   -fstrict-enums   -fstrict-vtable-pointers   -fvirtual-function-elimination 

C++ benchmarks:

 -Ofast   -fopenmp   -march=core-avx2   -fma   -ipo    -ansi-alias   -fp-model fast=2   -qno-opt-multiple-gather-scatter-by-shuffles   -qopt-zmm-usage=high   -ffast-math   -fstrict-enums   -fstrict-vtable-pointers 

Fortran benchmarks:

 -Ofast   -fopenmp   -march=core-avx2   -fma   -ipo    -ansi-alias   -fp-model fast=2   -qno-opt-multiple-gather-scatter-by-shuffles   -qopt-zmm-usage=high   -align array128byte   -ffinite-math-only   -fno-omit-frame-pointer   -m64   -ipo1   -foptimize-sibling-calls   -vec 

Peak Compiler Invocation

C benchmarks:

 icx 

C++ benchmarks:

 icpx 

Fortran benchmarks:

 ifx 

Peak Portability Flags

350.md:  -FR 
357.bt331:  -mcmodel=medium 
363.swim:  -mcmodel=medium 
367.imagick:  -std=c99 

Peak Optimization Flags

C benchmarks:

 -Ofast   -fopenmp   -march=core-avx2   -fma   -ipo    -ansi-alias   -fp-model fast=2   -qno-opt-multiple-gather-scatter-by-shuffles   -qopt-zmm-usage=high   -ffast-math   -fstrict-enums   -fstrict-vtable-pointers   -fvirtual-function-elimination 

C++ benchmarks:

 -Ofast   -fopenmp   -march=core-avx2   -fma   -ipo    -ansi-alias   -fp-model fast=2   -qno-opt-multiple-gather-scatter-by-shuffles   -qopt-zmm-usage=high   -ffast-math   -fstrict-enums   -fstrict-vtable-pointers 

Fortran benchmarks:

 -Ofast   -fopenmp   -march=core-avx2   -fma   -ipo    -ansi-alias   -fp-model fast=2   -qno-opt-multiple-gather-scatter-by-shuffles   -qopt-zmm-usage=high   -align array128byte   -ffinite-math-only   -fno-omit-frame-pointer   -m64   -ipo1   -foptimize-sibling-calls   -vec 

The flags file that was used to format this result can be browsed at
http://www.spec.org/omp2012/flags/lenovo-omp2012-oneAPI.20230816.00.html.

You can also download the XML flags source by saving the following link:
http://www.spec.org/omp2012/flags/lenovo-omp2012-oneAPI.20230816.00.xml.