SGI |
SPECompG_base2012 = 29.9 |
SGI UV1000 (Intel Xeon E7-8837, 2.66GHz) |
SPECompG_peak2012 = Not Run |
OMP2012 license: | 14 | Test date: | Sep-2012 |
---|---|---|---|
Test sponsor: | SGI | Hardware Availability: | Apr-2011 |
Tested by: | SGI | Software Availability: | Aug-2012 |
Hardware | |
---|---|
CPU Name: | Intel Xeon E7-8837 |
CPU Characteristics: | Intel Turbo Boost Technology up to 2.80 GHz |
CPU MHz: | 2667 |
CPU MHz Maximum: | 2800 |
FPU: | Integrated |
CPU(s) enabled: | 256 cores, 32 chips, 8 cores/chip |
CPU(s) orderable: | 2-256 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: | 24 MB I+D on chip per chip |
Other Cache: | None |
Memory: | 1 TB (256 x 4 GB 4Rx8 PC3-8500R-7, ECC) |
Disk Subsystem: | 10 x 1 TB SAS (Seagate Constellation ES, 7200RPM) |
Other Hardware: | Routered quad-plane fat tree topology |
Base Threads Run: | 256 |
Minimum Peak Threads: | -- |
Maximum Peak Threads: | -- |
Software | |
---|---|
Operating System: | SUSE Linux Enterprise Server 11 (x86_64) SP1 2.6.32.46-0.3.1.3592.0.PTF-default |
Compiler: | C/C++/Fortran: Version 13.0 of Intel Composer XE 2013 Build 20120731; |
Auto Parallel: | No |
File System: | xfs |
System State: | Run level 3 ( Multi-user ) |
Base Pointers: | 64-bit |
Peak Pointers: | Not Applicable |
Other Software: | sgi-accelerate-release: SGI Accelerate 1.3, Build 705rp1.sles11-1110302109 sgi-foundation-release: SGI Foundation Software 2.5, Build 705rp1.sles11-1110302109 |
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 | 256 | 129 | 35.8 | 130 | 35.7 | 130 | 35.7 | |||||||
351.bwaves | 256 | 103 | 44.0 | 102 | 44.2 | 103 | 44.2 | |||||||
352.nab | 256 | 227 | 17.2 | 226 | 17.2 | 220 | 17.6 | |||||||
357.bt331 | 256 | 169 | 28.1 | 169 | 28.0 | 168 | 28.1 | |||||||
358.botsalgn | 256 | 109 | 39.9 | 109 | 39.9 | 109 | 39.9 | |||||||
359.botsspar | 256 | 211 | 24.8 | 212 | 24.7 | 213 | 24.7 | |||||||
360.ilbdc | 256 | 211 | 16.9 | 212 | 16.8 | 211 | 16.8 | |||||||
362.fma3d | 256 | 272 | 14.0 | 270 | 14.1 | 270 | 14.1 | |||||||
363.swim | 256 | 90.8 | 49.9 | 89.0 | 50.9 | 89.4 | 50.7 | |||||||
367.imagick | 256 | 412 | 17.1 | 410 | 17.1 | 410 | 17.1 | |||||||
370.mgrid331 | 256 | 109 | 40.4 | 109 | 40.5 | 109 | 40.7 | |||||||
371.applu331 | 256 | 278 | 21.8 | 278 | 21.8 | 279 | 21.8 | |||||||
372.smithwa | 256 | 56.0 | 95.7 | 56.0 | 95.7 | 56.6 | 94.7 | |||||||
376.kdtree | 256 | 128 | 35.3 | 127 | 35.5 | 127 | 35.6 |
The config file option 'submit' was used.
Software Environment: export KMP_AFFINITY=disabled export KMP_STACKSIZE=200M export KMP_SCHEDULE=static,balanced export OMP_DYNAMIC=FALSE limit -s unlimited 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.
350.md: | -free |
367.imagick: | -std=c99 |