| CPU2006 license: | 19 | Test date: | Dec-2012 |
|---|---|---|---|
| Test sponsor: | Fujitsu | Hardware Availability: | Mar-2013 |
| Tested by: | Fujitsu | Software Availability: | Mar-2013 |
| Hardware | |
|---|---|
| CPU Name: | SPARC64 X |
| CPU Characteristics: | |
| CPU MHz: | 3000 |
| FPU: | Integrated |
| CPU(s) enabled: | 256 cores, 16 chips, 16 cores/chip, 2 threads/core |
| CPU(s) orderable: | 1 to 16 BBs; each BB contains 2 or 4 CPU chips |
| Primary Cache: | 64 KB I + 64 KB D on chip per core |
| Secondary Cache: | 24 MB I+D on chip per chip |
| L3 Cache: | None |
| Other Cache: | None |
| Memory: | 2 TB (128 x 16 GB 2Rx4 PC3L-12800R-11, ECC, running at 1600 MHz) |
| Disk Subsystem: | 1 x 600 GB SAS, 10025 RPM Toshiba MBF2600RC |
| Other Hardware: | None |
| Software | |
|---|---|
| Operating System: | Oracle Solaris 11.1 |
| Compiler: | C/C++/Fortran: Version 12.3 of Oracle Solaris Studio, 1/13 Platform Specific Enhancement |
| Auto Parallel: | No |
| File System: | zfs and tmpfs |
| System State: | Default |
| Base Pointers: | 32-bit |
| Peak Pointers: | 32-bit |
| Other Software: | None |
Results Table
| Benchmark | Base | Peak | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Copies | Seconds | Ratio | Seconds | Ratio | Seconds | Ratio | Copies | Seconds | Ratio | Seconds | Ratio | Seconds | Ratio | |
| Results appear in the order in which they were run. Bold underlined text indicates a median measurement. | ||||||||||||||
| 410.bwaves | 510 | 995 | 6960 | 993 | 6980 | 992 | 6980 | 510 | 1029 | 6740 | 983 | 7050 | 983 | 7050 |
| 416.gamess | 510 | 2200 | 4540 | 2230 | 4480 | 2197 | 4550 | 510 | 1993 | 5010 | 1996 | 5000 | 2003 | 4990 |
| 433.milc | 510 | 739 | 6330 | 739 | 6340 | 739 | 6330 | 510 | 738 | 6340 | 738 | 6340 | 738 | 6340 |
| 434.zeusmp | 510 | 846 | 5490 | 845 | 5490 | 847 | 5480 | 510 | 846 | 5490 | 845 | 5490 | 847 | 5480 |
| 435.gromacs | 510 | 831 | 4380 | 831 | 4380 | 826 | 4410 | 510 | 816 | 4460 | 809 | 4500 | 807 | 4510 |
| 436.cactusADM | 510 | 867 | 7030 | 867 | 7030 | 875 | 6970 | 510 | 859 | 7090 | 862 | 7070 | 860 | 7090 |
| 437.leslie3d | 510 | 1103 | 4350 | 1101 | 4350 | 1101 | 4360 | 510 | 1079 | 4440 | 1080 | 4440 | 1080 | 4440 |
| 444.namd | 510 | 815 | 5020 | 816 | 5010 | 816 | 5010 | 510 | 756 | 5410 | 753 | 5430 | 752 | 5440 |
| 447.dealII | 510 | 561 | 10400 | 561 | 10400 | 559 | 10400 | 510 | 558 | 10500 | 559 | 10400 | 557 | 10500 |
| 450.soplex | 510 | 1182 | 3600 | 1182 | 3600 | 1182 | 3600 | 510 | 1102 | 3860 | 1101 | 3860 | 1102 | 3860 |
| 453.povray | 510 | 570 | 4760 | 569 | 4770 | 571 | 4750 | 510 | 451 | 6020 | 448 | 6060 | 445 | 6090 |
| 454.calculix | 510 | 570 | 7390 | 571 | 7370 | 568 | 7410 | 510 | 541 | 7770 | 539 | 7810 | 537 | 7830 |
| 459.GemsFDTD | 510 | 1518 | 3560 | 1524 | 3550 | 1524 | 3550 | 510 | 1518 | 3560 | 1524 | 3550 | 1524 | 3550 |
| 465.tonto | 510 | 845 | 5940 | 837 | 5990 | 841 | 5970 | 510 | 714 | 7030 | 716 | 7010 | 717 | 7000 |
| 470.lbm | 510 | 1039 | 6750 | 1040 | 6740 | 1038 | 6750 | 510 | 1039 | 6750 | 1040 | 6740 | 1038 | 6750 |
| 481.wrf | 510 | 1076 | 5290 | 1074 | 5300 | 1071 | 5320 | 510 | 936 | 6090 | 925 | 6160 | 946 | 6020 |
| 482.sphinx3 | 510 | 1732 | 5740 | 1731 | 5740 | 1730 | 5750 | 510 | 1616 | 6150 | 1616 | 6150 | 1612 | 6170 |
Compiler Invocation Notes
The Apache C++ Standard Library V4.2.1 was installed from
http://stdcxx.apache.org/download.html using:
alias gmake=specmake
gmake BUILDTYPE=8d CONFIG=sunpro.config
Submit Notes
Processes were assigned to specific processors using 'pbind' commands. The config file option 'submit' was used, along with a list of processors in the 'BIND' variable, to generate the pbind commands. (For details, please see the config file.)
Operating System Notes
Shell Environments:
ulimit -s 131072 was used to limit the space consumed by the stack
(and therefore make more space available to the heap).
The "webconsole" service was turned off using svcadm disable webconsole.
System Tunables:
(/etc/system parameters)
lpg_alloc_prefer=1
Indicates that extra effort should be taken to ensure that pages are
created in the nearby lgroup (NUMA location).
Platform Notes
Sysinfo program /export/cpu2006-v1.2/config/sysinfo.rev6818 $Rev: 6818 $ $Date:: 2012-07-17 #$ e86d102572650a6e4d596a3cee98f191 running on 4S-710-D0 Sun Dec 23 06:23:44 2012 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/cpu2006/Docs/config.html#sysinfo From /usr/sbin/psrinfo SPARC64-X (chipid 0, clock 3000 MHz) SPARC64-X (chipid 1, clock 3000 MHz) SPARC64-X (chipid 10, clock 3000 MHz) SPARC64-X (chipid 11, clock 3000 MHz) SPARC64-X (chipid 12, clock 3000 MHz) SPARC64-X (chipid 13, clock 3000 MHz) SPARC64-X (chipid 14, clock 3000 MHz) SPARC64-X (chipid 15, clock 3000 MHz) SPARC64-X (chipid 2, clock 3000 MHz) SPARC64-X (chipid 3, clock 3000 MHz) SPARC64-X (chipid 4, clock 3000 MHz) SPARC64-X (chipid 5, clock 3000 MHz) SPARC64-X (chipid 6, clock 3000 MHz) SPARC64-X (chipid 7, clock 3000 MHz) SPARC64-X (chipid 8, clock 3000 MHz) SPARC64-X (chipid 9, clock 3000 MHz) 16 chips 512 threads 3000 MHz From kstat: 256 cores From prtconf: 2091520 Megabytes /etc/release: Oracle Solaris 11.1 SPARC uname -a: SunOS 4S-710-D0 5.11 11.1 sun4v sparc sun4v disk: df -h $SPEC Filesystem Size Used Available Capacity Mounted on rpool/export 547G 4.9G 466G 2% /export (End of data from sysinfo program)
General Notes
output_root was used to put run directories in /tmp/cpu2006 (tmpfs).
Base Compiler Invocation
C benchmarks:
| cc |
C++ benchmarks:
| CC |
Fortran benchmarks:
| f90 |
Benchmarks using both Fortran and C:
| cc f90 |
Base Optimization Flags
C benchmarks:
| -fast -xtarget=sparc64x -fma=fused -xpagesize=4M -xlinkopt -xvector -xipo=2 -xalias_level=std -xprefetch_level=3 -xunroll=4 -lbsdmalloc -M /usr/lib/ld/map.bssalign |
C++ benchmarks:
Fortran benchmarks:
| -fast -xtarget=sparc64x -fma=fused -xpagesize=4M -xlinkopt -xvector -xipo=2 -xprefetch_auto_type=indirect_array_access -xprefetch_level=3 -M /usr/lib/ld/map.bssalign |
Benchmarks using both Fortran and C:
Base Other Flags
C benchmarks:
| -xjobs=16 |
C++ benchmarks:
| -xjobs=16 |
Fortran benchmarks:
| -xjobs=16 |
Benchmarks using both Fortran and C:
| -xjobs=16 |
Peak Compiler Invocation
C benchmarks:
| cc |
C++ benchmarks:
| CC |
Fortran benchmarks:
| f90 |
Benchmarks using both Fortran and C:
| cc f90 |
Peak Optimization Flags
C benchmarks:
C++ benchmarks:
Fortran benchmarks:
| 410.bwaves: | -xprofile=collect:./feedback(pass 1) -xprofile=use:./feedback(pass 2) -fast -xtarget=sparc64x -fma=fused -xpagesize=4M -xipo=2 -xlinkopt -xunroll=4 -xprefetch=no%auto |
| 416.gamess: | -fast -xtarget=sparc64x -fma=fused -xpagesize=4M -xtarget=sparc64vii -xlinkopt -xprefetch=no%auto -xunroll=6 -M /usr/lib/ld/map.bssalign |
| 434.zeusmp: | basepeak = yes |
| 437.leslie3d: | -fast -xtarget=sparc64x -fma=fused -xpagesize=4M -xtarget=sparc64vii -xprefetch=latx:1.2 -Qoption cg -Qms_pipe+alldoall -M /usr/lib/ld/map.bssalign |
| 459.GemsFDTD: | basepeak = yes |
| 465.tonto: | -xprofile=collect:./feedback(pass 1) -xprofile=use:./feedback(pass 2) -fast -xtarget=sparc64x -fma=fused -xpagesize=4M -xipo=1 -xO4 -xunroll=3 -xprefetch=no%auto -lbsdmalloc |