|
Hewlett Packard Enterprise (Test Sponsor: HPE) |
SPECompG_base2012 = 128 |
|
Superdome Flex (Intel Xeon Gold 6154, 3.00 GHz) |
SPECompG_peak2012 = 134 |
| OMP2012 license: | 1 | Test date: | Dec-2017 |
|---|---|---|---|
| Test sponsor: | HPE | Hardware Availability: | Mar-2018 |
| Tested by: | HPE | Software Availability: | Mar-2018 |
| Hardware | |
|---|---|
| CPU Name: | Intel Xeon Gold 6154 |
| CPU Characteristics: | Intel Turbo Boost Technology up to 3.70 GHz |
| CPU MHz: | 3000 |
| CPU MHz Maximum: | 3700 |
| FPU: | Integrated |
| CPU(s) enabled: | 576 cores, 32 chips, 18 cores/chip |
| CPU(s) orderable: | 4-32 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: | 24.75 MB I+D on chip per chip |
| Other Cache: | None |
| Memory: | 12 TB (384 x 32 GB 2Rx4 PC4-2666V-R) |
| Disk Subsystem: | tmpfs |
| Other Hardware: | None |
| Base Threads Run: | 513 |
| Minimum Peak Threads: | 512 |
| Maximum Peak Threads: | 576 |
| Software | |
|---|---|
| Operating System: | SUSE Linux Enterprise Server 12 SP2 Kernel 4.4.74-92.38-default |
| Compiler: | C/C++/Fortran: Version 18.0.0.128 of Intel Composer XE for Linux, Build 20170811 |
| Auto Parallel: | No |
| File System: | tmpfs |
| System State: | Multi-user, run level 3 |
| Base Pointers: | 64-bit |
| Peak Pointers: | 64-bit |
| Other Software: | HPE Foundation Software 1.0, Build 717a270.sles12sp2-1709012000 |
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 | 513 | 5.65 | 819 | 5.60 | 827 | 5.59 | 828 | 576 | 5.33 | 869 | 5.52 | 839 | 5.31 | 872 |
| 351.bwaves | 513 | 26.9 | 168 | 26.9 | 168 | 27.0 | 168 | 517 | 27.1 | 167 | 26.9 | 168 | 26.9 | 169 |
| 352.nab | 513 | 69.2 | 56.2 | 69.2 | 56.2 | 69.2 | 56.2 | 513 | 69.2 | 56.2 | 69.2 | 56.2 | 69.2 | 56.2 |
| 357.bt331 | 513 | 47.7 | 99.3 | 47.8 | 99.3 | 47.7 | 99.4 | 531 | 47.5 | 99.7 | 48.0 | 98.8 | 47.6 | 99.7 |
| 358.botsalgn | 513 | 29.5 | 147 | 29.5 | 148 | 29.5 | 147 | 576 | 26.7 | 163 | 26.7 | 163 | 26.7 | 163 |
| 359.botsspar | 513 | 69.1 | 76.0 | 70.0 | 75.0 | 70.1 | 74.9 | 513 | 69.1 | 76.0 | 70.0 | 75.0 | 70.1 | 74.9 |
| 360.ilbdc | 513 | 36.6 | 97.3 | 36.5 | 97.5 | 36.6 | 97.3 | 576 | 33.9 | 105 | 33.8 | 105 | 33.8 | 105 |
| 362.fma3d | 513 | 78.8 | 48.2 | 78.7 | 48.3 | 78.8 | 48.2 | 567 | 73.7 | 51.6 | 73.8 | 51.5 | 73.6 | 51.6 |
| 363.swim | 513 | 28.4 | 160 | 28.4 | 159 | 28.3 | 160 | 567 | 26.6 | 170 | 26.4 | 171 | 26.5 | 171 |
| 367.imagick | 513 | 54.4 | 129 | 55.3 | 127 | 54.9 | 128 | 576 | 54.1 | 130 | 54.0 | 130 | 53.6 | 131 |
| 370.mgrid331 | 513 | 34.0 | 130 | 34.1 | 130 | 34.1 | 130 | 512 | 30.1 | 147 | 30.2 | 146 | 30.2 | 146 |
| 371.applu331 | 513 | 84.4 | 71.8 | 85.1 | 71.2 | 84.8 | 71.4 | 513 | 84.4 | 71.8 | 85.1 | 71.2 | 84.8 | 71.4 |
| 372.smithwa | 513 | 14.1 | 381 | 14.1 | 380 | 16.0 | 334 | 576 | 12.8 | 418 | 12.9 | 417 | 12.8 | 419 |
| 376.kdtree | 513 | 40.8 | 110 | 40.4 | 111 | 41.1 | 109 | 576 | 38.2 | 118 | 38.1 | 118 | 38.0 | 118 |
Compiler Invocation Notes
COPTIMIZE=-O3 -qopt-zmm-usage=high -xCORE-AVX512 -ipo1 -qopenmp -ansi-alias -mcmodel=medium -shared-intel CXXOPTIMIZE=-O3 -qopt-zmm-usage=high -xCORE-AVX512 -ipo1 -qopenmp -ansi-alias -mcmodel=medium -shared-intel FOPTIMIZE=-O3 -qopt-zmm-usage=high -xCORE-AVX512 -ipo1 -qopenmp -mcmodel=medium -shared-intel
Submit Notes
The config file option 'submit' was used.
For all benchmarks threads were bound to cores using
the following submit command:
dplace $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.
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
The tmpfs filesystem was set up with:
mount -t tmpfs -o rw,remount,mode=1777,mpol=interleave tmpfs /dev/shm
Platform Notes
Rack Management Controller settings: modify npar pnum=0 ras=hpc modify npar pnum=0 hthread=off
Base Portability Flags
| 350.md: | -free |
| 367.imagick: | -std=c99 |
Base Optimization Flags
C benchmarks:
| -O3 -qopt-zmm-usage=high -xCORE-AVX512 -ipo1 -qopenmp -ansi-alias -mcmodel=medium -shared-intel |
C++ benchmarks:
| -O3 -qopt-zmm-usage=high -xCORE-AVX512 -ipo1 -qopenmp -ansi-alias -mcmodel=medium -shared-intel |
Fortran benchmarks:
| -O3 -qopt-zmm-usage=high -xCORE-AVX512 -ipo1 -qopenmp -mcmodel=medium -shared-intel |
Peak Portability Flags
| 350.md: | -free |
| 367.imagick: | -std=c99 |
Peak Optimization Flags
C benchmarks:
| 352.nab: | basepeak = yes |
| 358.botsalgn: | -O3 -qopt-zmm-usage=high -xCORE-AVX512 -ipo1 -qopenmp -ansi-alias -mcmodel=medium -shared-intel |
| 359.botsspar: | basepeak = yes |
| 367.imagick: | Same as 358.botsalgn |
| 372.smithwa: | Same as 358.botsalgn |
C++ benchmarks:
| -O3 -qopt-zmm-usage=high -xCORE-AVX512 -ipo1 -qopenmp -ansi-alias -mcmodel=medium -shared-intel |
Fortran benchmarks:
| 350.md: | -O3 -qopt-zmm-usage=high -xCORE-AVX512 -ipo1 -qopenmp -mcmodel=medium -shared-intel |
| 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 |