|
SGI |
SPECompG_base2012 = 84.5 |
|
SGI UV 300 (Intel Xeon E7-8867 v4, 2.40 GHz) |
SPECompG_peak2012 = 90.0 |
| OMP2012 license: | 14 | Test date: | Jun-2016 |
|---|---|---|---|
| Test sponsor: | SGI | Hardware Availability: | Jun-2016 |
| Tested by: | SGI | Software Availability: | Apr-2016 |
| 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: | 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: | 256 KB I+D on chip per core |
| L3 Cache: | 45 MB I+D on chip per chip |
| Other Cache: | None |
| Memory: | 8 TB (512 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: | 513 |
| Minimum Peak Threads: | 512 |
| 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 | 513 | 31.4 | 147 | 31.4 | 147 | 45.9 | 101 | 576 | 28.3 | 164 | 41.2 | 112 | 28.3 | 164 |
| 351.bwaves | 513 | 33.8 | 134 | 33.7 | 134 | 33.8 | 134 | 517 | 33.7 | 134 | 33.7 | 134 | 33.9 | 134 |
| 352.nab | 513 | 91.1 | 42.7 | 91.2 | 42.7 | 91.2 | 42.7 | 520 | 87.4 | 44.5 | 87.5 | 44.5 | 87.4 | 44.5 |
| 357.bt331 | 513 | 52.4 | 90.5 | 52.3 | 90.6 | 52.4 | 90.4 | 531 | 52.4 | 90.4 | 52.4 | 90.5 | 52.5 | 90.4 |
| 358.botsalgn | 513 | 37.7 | 116 | 37.6 | 116 | 37.5 | 116 | 576 | 34.6 | 126 | 34.7 | 125 | 34.9 | 125 |
| 359.botsspar | 513 | 101 | 51.8 | 102 | 51.7 | 97.5 | 53.9 | 513 | 101 | 51.8 | 102 | 51.7 | 97.5 | 53.9 |
| 360.ilbdc | 513 | 85.7 | 41.5 | 85.7 | 41.6 | 85.5 | 41.6 | 576 | 77.6 | 45.9 | 77.7 | 45.8 | 77.6 | 45.9 |
| 362.fma3d | 513 | 106 | 35.9 | 106 | 36.0 | 106 | 35.9 | 567 | 92.5 | 41.1 | 92.6 | 41.0 | 92.5 | 41.1 |
| 363.swim | 513 | 38.5 | 118 | 38.4 | 118 | 38.4 | 118 | 567 | 35.6 | 127 | 35.7 | 127 | 35.6 | 127 |
| 367.imagick | 513 | 95.6 | 73.5 | 95.7 | 73.4 | 96.1 | 73.2 | 576 | 88.9 | 79.1 | 91.7 | 76.7 | 90.9 | 77.4 |
| 370.mgrid331 | 513 | 43.7 | 101 | 43.7 | 101 | 43.5 | 102 | 512 | 40.8 | 108 | 40.9 | 108 | 40.7 | 108 |
| 371.applu331 | 513 | 98.6 | 61.5 | 97.7 | 62.0 | 97.7 | 62.0 | 531 | 96.1 | 63.0 | 97.3 | 62.3 | 95.7 | 63.3 |
| 372.smithwa | 513 | 20.3 | 264 | 20.0 | 268 | 20.1 | 267 | 576 | 18.4 | 291 | 18.3 | 293 | 18.7 | 287 |
| 376.kdtree | 513 | 47.8 | 94.1 | 47.3 | 95.2 | 45.8 | 98.2 | 576 | 41.0 | 110 | 41.4 | 109 | 43.4 | 104 |
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 set to Disabled
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 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: | Same as 352.nab |
| 359.botsspar: | basepeak = yes |
| 367.imagick: | Same as 352.nab |
| 372.smithwa: | Same as 352.nab |
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 |