SPEC® OMPG2012 Result

Copyright 2012-2016 Standard Performance Evaluation Corporation

SGI

SPECompG_base2012 = 57.0 

SGI UV 300 (Intel Xeon E7-8867 v4, 2.40 GHz)

SPECompG_peak2012 = 62.0 

OMP2012 license: 14 Test date: Jun-2016
Test sponsor: SGI Hardware Availability: Jun-2016
Tested by: SGI Software Availability: Apr-2016
Benchmark results graph
Hardware
CPU Name: Intel Xeon E7-8867 v4
CPU Characteristics: Intel Turbo Boost Technology up to 3.30 GHz
CPU MHz: 2400
CPU MHz Maximum: 3300
FPU: Integrated
CPU(s) enabled: 288 cores, 16 chips, 18 cores/chip, 2 threads/core
CPU(s) orderable: 4-32 chips
Primary Cache: 32 KB I + 32 KB D on chip per core
Secondary Cache: 256 KB I+D on chip per core
L3 Cache: 45 MB I+D on chip per chip
Other Cache: None
Memory: 4 TB (256 x 16 GB 2Rx4 PC4-2133P-R,
running at 1600 MHz)
Disk Subsystem: 1 x 400 GB SSD (Intel SSD 3500 Series,
SATA II)
Other Hardware: None
Base Threads Run: 512
Minimum Peak Threads: 256
Maximum Peak Threads: 576
Software
Operating System: SUSE Linux Enterprise Server 12 (x86_64) SP1
Kernel 3.12.57-60.35-default
Compiler: C/C++/Fortran: Version 16.0.3.210 of Intel
Composer XE for Linux, Build 20160415
Auto Parallel: No
File System: ext3
System State: Multi-user, run level 3
Base Pointers: 64-bit
Peak Pointers: Not Applicable
Other Software: SGI Accelerate 1.12
(Build 714r28.sles12sp1-1604201900),
SGI Foundation Software 2.14
(Build 714r28.sles12sp1-1604201900)

Results Table

Benchmark Base Peak
Threads Seconds Ratio Seconds Ratio Seconds Ratio Threads Seconds Ratio Seconds Ratio Seconds Ratio
Results appear in the order in which they were run. Bold underlined text indicates a median measurement.
350.md 512 45.2  102    45.1  103    45.2  103    576 40.9  113    43.0  108    41.0  113   
351.bwaves 512 61.9  73.2  62.0  73.1  61.9  73.2  576 59.4  76.2  59.4  76.2  59.5  76.2 
352.nab 512 103    37.9  103    37.9  103    37.8  567 101    38.6  101    38.6  101    38.6 
357.bt331 512 74.5  63.7  74.5  63.6  74.4  63.8  540 74.3  63.8  74.1  64.0  73.9  64.2 
358.botsalgn 512 60.0  72.5  60.0  72.5  60.0  72.5  512 60.0  72.5  60.0  72.5  60.0  72.5 
359.botsspar 512 198    26.5  217    24.2  201    26.2  288 124    42.4  123    42.6  125    42.0 
360.ilbdc 512 99.0  36.0  99.0  36.0  99.0  35.9  576 89.6  39.7  89.5  39.8  89.5  39.8 
362.fma3d 512 166    22.9  166    22.9  166    22.9  567 144    26.4  144    26.4  144    26.4 
363.swim 512 71.5  63.3  71.4  63.4  71.8  63.1  288 61.6  73.6  61.5  73.6  61.5  73.6 
367.imagick 512 111    63.6  111    63.3  111    63.2  512 111    63.6  111    63.3  111    63.2 
370.mgrid331 512 75.8  58.3  75.9  58.2  75.7  58.4  256 73.7  60.0  73.5  60.1  73.6  60.0 
371.applu331 512 108    56.2  108    56.0  108    56.0  512 108    56.2  108    56.0  108    56.0 
372.smithwa 512 33.0  163    33.1  162    33.2  161    576 29.8  180    29.3  183    29.6  181   
376.kdtree 512 70.2  64.1  70.3  64.0  70.8  63.5  549 69.3  64.9  70.3  64.0  69.8  64.5 

Submit Notes

The config file option 'submit' was used.
For all benchmarks threads were bound to cores using
the following submit command:
    dplace -x2 $command
This binds threads in order of creation, beginning
with the master thread on logical cpu 0, the first slave
thread on logical cpu 1, and so on.  The -x2 flag instructs
dplace to skip placement of the lightweight OpenMP monitor
thread, which is created prior to the slave threads.

Operating System Notes

Transparent Hugepages :
    Transparent Hugepages are disabled by
    echo never > /sys/kernel/mm/transparent_hugepage/enabled

Software Environment:
    export KMP_AFFINITY=disabled
    export KMP_STACKSIZE=200M
    export KMP_SCHEDULE=static,balanced
    export OMP_DYNAMIC=FALSE
    ulimit -s unlimited

Platform Notes

Intel Hyperthreading Enabled

Base Compiler Invocation

C benchmarks:

 icc 

C++ benchmarks:

 icpc 

Fortran benchmarks:

 ifort 

Base Portability Flags

350.md:  -free 
367.imagick:  -std=c99 

Base Optimization Flags

C benchmarks:

 -O3   -xCORE-AVX2   -ipo1   -openmp   -ansi-alias   -mcmodel=medium   -shared-intel 

C++ benchmarks:

 -O3   -xCORE-AVX2   -ipo1   -openmp   -ansi-alias   -mcmodel=medium   -shared-intel 

Fortran benchmarks:

 -O3   -xCORE-AVX2   -ipo1   -openmp   -mcmodel=medium   -shared-intel   -align array64byte 

Peak Compiler Invocation

C benchmarks:

 icc 

C++ benchmarks:

 icpc 

Fortran benchmarks:

 ifort 

Peak Portability Flags

350.md:  -free 
367.imagick:  -std=c99 

Peak Optimization Flags

C benchmarks:

352.nab:  -O3   -xCORE-AVX2   -ipo1   -openmp   -ansi-alias   -mcmodel=medium   -shared-intel 
358.botsalgn:  basepeak = yes 
359.botsspar:  Same as 352.nab 
367.imagick:  basepeak = yes 
372.smithwa:  Same as 352.nab 

C++ benchmarks:

 -O3   -xCORE-AVX2   -ipo1   -openmp   -ansi-alias   -mcmodel=medium   -shared-intel 

Fortran benchmarks:

350.md:  -O3   -xCORE-AVX2   -ipo1   -openmp   -mcmodel=medium   -shared-intel   -align array64byte 
351.bwaves:  Same as 350.md 
357.bt331:  Same as 350.md 
360.ilbdc:  Same as 350.md 
362.fma3d:  Same as 350.md 
363.swim:  Same as 350.md 
370.mgrid331:  Same as 350.md 
371.applu331:  basepeak = yes 

The flags files that were used to format this result can be browsed at
http://www.spec.org/omp2012/flags/SGI-OMP2012-ic16.20160706.html,
http://www.spec.org/omp2012/flags/SGI-UV300-RevB.20160706.html.

You can also download the XML flags sources by saving the following links:
http://www.spec.org/omp2012/flags/SGI-OMP2012-ic16.20160706.xml,
http://www.spec.org/omp2012/flags/SGI-UV300-RevB.20160706.xml.