| CPU2006 license: | 19 | Test date: | Jul-2008 |
|---|---|---|---|
| Test sponsor: | Fujitsu Limited | Hardware Availability: | Jul-2008 |
| Tested by: | Fujitsu Limited | Software Availability: | Jul-2008 |
| Hardware | |
|---|---|
| CPU Name: | SPARC64 VII |
| CPU Characteristics: | |
| CPU MHz: | 2520 |
| FPU: | Integrated |
| CPU(s) enabled: | 128 cores, 32 chips, 4 cores/chip, 2 threads/core |
| CPU(s) orderable: | 1 to 16 CMUs; each CMU contains 2 or 4 chips |
| Primary Cache: | 64 KB I + 64 KB D on chip per core |
| Secondary Cache: | 6 MB I+D on chip per chip |
| L3 Cache: | None |
| Other Cache: | None |
| Memory: | 512 GB (256 x 2 GB) |
| Disk Subsystem: | OS disk: 1 x 72 GB 10000 RPM disk Seagate Savvio EX.disk: 864 GB RAID 0 Solaris Volume 12 x 72 GB 10000 RPM disk Stripe interlace size 786 blocks |
| Other Hardware: | None |
| Software | |
|---|---|
| Operating System: | Solaris 10 5/08 with Patch 137111-03 |
| Compiler: | Sun Studio 12 with patches 124867-06, 124861-07, 124863-05, 127000-05 (see patch information below) |
| Auto Parallel: | No |
| File System: | ufs |
| 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 | 255 | 3585 | 967 | 3519 | 985 | 3518 | 985 | 255 | 3585 | 967 | 3519 | 985 | 3518 | 985 |
| 416.gamess | 255 | 3241 | 1540 | 3270 | 1530 | 3316 | 1510 | 255 | 3241 | 1540 | 3270 | 1530 | 3316 | 1510 |
| 433.milc | 255 | 4966 | 471 | 4982 | 470 | 4949 | 473 | 255 | 4849 | 483 | 4848 | 483 | 4849 | 483 |
| 434.zeusmp | 255 | 1852 | 1250 | 1854 | 1250 | 1862 | 1250 | 255 | 1852 | 1250 | 1854 | 1250 | 1862 | 1250 |
| 435.gromacs | 255 | 1098 | 1660 | 1111 | 1640 | 1111 | 1640 | 255 | 1012 | 1800 | 997 | 1830 | 1006 | 1810 |
| 436.cactusADM | 255 | 2142 | 1420 | 2151 | 1420 | 2144 | 1420 | 127 | 925 | 1640 | 924 | 1640 | 922 | 1650 |
| 437.leslie3d | 255 | 3679 | 651 | 3703 | 647 | 3684 | 651 | 127 | 1763 | 677 | 1763 | 677 | 1763 | 677 |
| 444.namd | 255 | 1111 | 1840 | 1099 | 1860 | 1077 | 1900 | 255 | 1096 | 1870 | 1095 | 1870 | 1068 | 1920 |
| 447.dealII | 255 | 1463 | 1990 | 1467 | 1990 | 1454 | 2010 | 255 | 1464 | 1990 | 1474 | 1980 | 1454 | 2010 |
| 450.soplex | 255 | 3946 | 539 | 3844 | 553 | 3829 | 555 | 127 | 1965 | 539 | 1882 | 563 | 1888 | 561 |
| 453.povray | 255 | 839 | 1620 | 813 | 1670 | 829 | 1640 | 255 | 591 | 2290 | 598 | 2270 | 585 | 2320 |
| 454.calculix | 255 | 1101 | 1910 | 1121 | 1880 | 1094 | 1920 | 255 | 1107 | 1900 | 1103 | 1910 | 1085 | 1940 |
| 459.GemsFDTD | 255 | 5564 | 486 | 5563 | 486 | 5564 | 486 | 255 | 5564 | 486 | 5563 | 486 | 5564 | 486 |
| 465.tonto | 255 | 1914 | 1310 | 1920 | 1310 | 1925 | 1300 | 191 | 1314 | 1430 | 1315 | 1430 | 1313 | 1430 |
| 470.lbm | 255 | 6339 | 553 | 6338 | 553 | 6339 | 553 | 127 | 3157 | 553 | 3157 | 553 | 3157 | 553 |
| 481.wrf | 255 | 2795 | 1020 | 2786 | 1020 | 2764 | 1030 | 127 | 1341 | 1060 | 1341 | 1060 | 1340 | 1060 |
| 482.sphinx3 | 255 | 5819 | 854 | 5789 | 858 | 5858 | 848 | 255 | 5711 | 870 | 5726 | 868 | 5707 | 871 |
Compiler Invocation Notes
Sun Studio compiler patches are available at
http://developers.sun.com/sunstudio/downloads/patches/ss12_patches.jsp
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
Environment Variable Settings:
LD_PRELOAD=mpss.so.1:madv.so.1
MPSSHEAP=4MB
MPSSSTACK=4MB
Requests system to use 4 MB pages when possible.
MADV access_lwp
access_lwp requests that the next light weight process to touch
the specified address range will access it most heavily.
ulimit -s 131072 was used to limit the space consumed
by the stack (making more space available for the heap)
System Tunables:(/etc/system parameters)
tune_t_fsflushr=4
Controls how many seconds elapse between runs of the
page flush daemon, fsflush.
autoup=1920
Causes pages older than the listed number of seconds to
be written by fsflush.
lpg_alloc_prefer=1
Set lgroup page allocation to strongly prefer local pages.
Other System Settings:
The webconsole service was turned off using
svcadm disable webconsole
Platform Notes
Memory is 8-way interleaved by filling all slots with the same capacity DIMMs. This result is measured on a Fujitsu SPARC Enterprise M9000 Server. Note that the Sun SPARC Enterprise M9000 and Fujitsu SPARC Enterprise M9000 are electrically equivalent.
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 -xipo=2 -xpagesize=4M -xprefetch_level=1 -xalias_level=std -fma=fused -ll2amm |
C++ benchmarks:
| -library=stlport4 -fast -xipo=2 -xpagesize=4M -xprefetch_level=1 -xalias_level=compatible -fma=fused -ll2amm |
Fortran benchmarks:
| -fast -xipo=2 -xpagesize=4M -xprefetch_level=1 -fma=fused -ll2amm |
Benchmarks using both Fortran and C:
| -fast(cc) -fast(f90) -xipo=2 -xpagesize=4M -xprefetch_level=1 -xalias_level=std -fma=fused -ll2amm |
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:
| 433.milc: | -fast -xipo=2 -xpagesize=4M -fma=fused -xalias_level=strong -xprefetch_auto_type=indirect_array_access -ll2amm |
| 470.lbm: | -xprofile=collect:./feedback(pass 1) -xprofile=use:./feedback(pass 2) -fast -xrestrict -xipo=2 -xprefetch_level=2 -xarch=v8plusb -fma=fused -ll2amm |
| 482.sphinx3: | -xprofile=collect:./feedback(pass 1) -xprofile=use:./feedback(pass 2) -fast -xipo=2 -xpagesize=4M -fma=fused -lfast -ll2amm |
C++ benchmarks:
Fortran benchmarks:
| 410.bwaves: | basepeak = yes |
| 416.gamess: | basepeak = yes |
| 434.zeusmp: | basepeak = yes |
| 437.leslie3d: | -fast -xipo=2 -xpagesize=4M -fma=fused -xprefetch=latx:5.0 -ll2amm |
| 459.GemsFDTD: | basepeak = yes |
| 465.tonto: | -fast -xipo=2 -xpagesize=4M -fma=fused -xprefetch_level=2 -xprefetch=latx:3 -lfast -ll2amm |
Benchmarks using both Fortran and C:
| 435.gromacs: | -xprofile=collect:./feedback(pass 1) -xprofile=use:./feedback(pass 2) -fast(cc) -fast(f90) -xipo=2 -xpagesize=4M -fma=fused -xinline= -fsimple=0 -xprefetch=no -xarch=generic -xchip=generic |
| 436.cactusADM: | -fast(cc) -fast(f90) -xipo=2 -xpagesize=4M -fma=fused -ll2amm |
| 454.calculix: | -fast(cc) -fast(f90) -xipo=2 -xpagesize=4M -fma=fused -xvector -xprefetch_level=3 -xprefetch=latx:8.0 -xalias_level=std -ll2amm |
| 481.wrf: | -xprofile=collect:./feedback(pass 1) -xprofile=use:./feedback(pass 2) -fast(cc) -fast(f90) -xipo=2 -xpagesize=4M -fma=fused -xprefetch_level=2 -xprefetch=latx:2 -ll2amm |