| CPU2006 license: | 19 | Test date: | Sep-2009 |
|---|---|---|---|
| Test sponsor: | Fujitsu | Hardware Availability: | Nov-2009 |
| Tested by: | Sun Microsystems | Software Availability: | Oct-2009 |
| Hardware | |
|---|---|
| CPU Name: | SPARC64 VII |
| CPU Characteristics: | |
| CPU MHz: | 2530 |
| FPU: | Integrated |
| CPU(s) enabled: | 16 cores, 4 chips, 4 cores/chip, 2 threads/core |
| CPU(s) orderable: | 1 to 2 CPUMs; each CPUM contains 2 CPU chips |
| Primary Cache: | 64 KB I + 64 KB D on chip per core |
| Secondary Cache: | 5632 KB I+D on chip per chip |
| L3 Cache: | None |
| Other Cache: | None |
| Memory: | 32 GB (32 x 1 GB), 8-way interleaved |
| Disk Subsystem: | 2x 146 GB SAS 10k rpm |
| Other Hardware: | None |
| Software | |
|---|---|
| Operating System: | Solaris 10 10/09 (s10s_u8wos_06) |
| Compiler: | Sun Studio 12 Update 1 plus patches (see notes) |
| Auto Parallel: | No |
| File System: | zfs |
| 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 | 32 | 5992 | 72.6 | 5990 | 72.6 | 5992 | 72.6 | 16 | 2991 | 72.7 | 2991 | 72.7 | 2991 | 72.7 |
| 416.gamess | 32 | 3127 | 200 | 3136 | 200 | 3142 | 199 | 32 | 3051 | 205 | 3041 | 206 | 3043 | 206 |
| 433.milc | 32 | 6592 | 44.6 | 6593 | 44.6 | 6587 | 44.6 | 32 | 6582 | 44.6 | 6577 | 44.7 | 6583 | 44.6 |
| 434.zeusmp | 32 | 2353 | 124 | 2345 | 124 | 2373 | 123 | 32 | 2274 | 128 | 2283 | 128 | 2266 | 129 |
| 435.gromacs | 32 | 1024 | 223 | 1025 | 223 | 1037 | 220 | 32 | 948 | 241 | 950 | 240 | 949 | 241 |
| 436.cactusADM | 32 | 2383 | 160 | 2382 | 161 | 2382 | 161 | 16 | 1152 | 166 | 1154 | 166 | 1156 | 165 |
| 437.leslie3d | 32 | 4895 | 61.5 | 4892 | 61.5 | 4894 | 61.5 | 16 | 2305 | 65.3 | 2304 | 65.3 | 2303 | 65.3 |
| 444.namd | 32 | 1226 | 209 | 1229 | 209 | 1226 | 209 | 32 | 1092 | 235 | 1083 | 237 | 1085 | 237 |
| 447.dealII | 32 | 1507 | 243 | 1490 | 246 | 1508 | 243 | 32 | 1505 | 243 | 1566 | 234 | 1489 | 246 |
| 450.soplex | 32 | 5227 | 51.1 | 5241 | 50.9 | 5239 | 50.9 | 16 | 2613 | 51.1 | 2612 | 51.1 | 2611 | 51.1 |
| 453.povray | 32 | 693 | 246 | 714 | 238 | 687 | 248 | 32 | 566 | 301 | 578 | 295 | 575 | 296 |
| 454.calculix | 32 | 1166 | 227 | 1156 | 228 | 1136 | 232 | 32 | 1136 | 232 | 1115 | 237 | 1136 | 232 |
| 459.GemsFDTD | 32 | 7489 | 45.3 | 7561 | 44.9 | 7573 | 44.8 | 16 | 3563 | 47.6 | 3564 | 47.6 | 3562 | 47.7 |
| 465.tonto | 32 | 2174 | 145 | 2144 | 147 | 2150 | 146 | 32 | 2033 | 155 | 1993 | 158 | 2033 | 155 |
| 470.lbm | 32 | 8545 | 51.5 | 8545 | 51.5 | 8546 | 51.5 | 32 | 8545 | 51.5 | 8545 | 51.5 | 8546 | 51.5 |
| 481.wrf | 32 | 3638 | 98.2 | 3646 | 98.0 | 3641 | 98.2 | 16 | 1716 | 104 | 1712 | 104 | 1727 | 104 |
| 482.sphinx3 | 32 | 9190 | 67.9 | 9191 | 67.9 | 9178 | 68.0 | 32 | 8597 | 72.5 | 8600 | 72.5 | 8627 | 72.3 |
Compiler Invocation Notes
Sun Studio 12 Update 1 was used, plus patch 119963-17 Sun Studio compiler patches are available at http://developers.sun.com/sunstudio/downloads/patches/ss12u1_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
ulimit -s 131072 was used to allow the stack to grow
up to 131072 KB (aka 128 MB). Note that saying "131072"
is preferable to "unlimited", because there is a tradeoff
between space for the stack vs. space for the heap.
System Tunables (/etc/system parameters):
tune_t_fsflushr=10
Controls how many seconds elapse between runs of the
page flush daemon, fsflush.
autoup=600
Causes pages older than the listed number of seconds to
be written by fsflush.
zfs:zfs_arc_max = 0x10000000
Control the amount of memory used by ZFS for caching
Other System Settings:
The webconsole service was turned off using
svcadm disable webconsole
The system had 32 GB of swap space
SPEC CPU2006 used 1 disk, with zfs gzip compression.
Platform Notes
Memory is 8-way interleaved by filling all slots with the same capacity DIMMs. This result is measured on a Sun SPARC Enterprise M4000 Server. The Sun SPARC Enterprise M4000 and the Fujitsu SPARC Enterprise M4000 are electrically equivalent.
General Notes
Environment variables set by runspec before the start of the run: OMP_NUM_THREADS = "32" SUNW_MP_PROCBIND = "true" SUNW_MP_THR_IDLE = "SPIN" (Although these variables were set prior to the run they did not affect performance, since the benchmarks were compiled in serial mode.) 447.dealII (peak): "apache_stdcxx_4_2_1" src.alt was used. 447.dealII (base): "apache_stdcxx_4_2_1" src.alt was used.
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 -fma=fused -xipo=2 -xpagesize=4M -xalias_level=std -xprefetch_auto_type=indirect_array_access -xprefetch_level=3 |
C++ benchmarks:
Fortran benchmarks:
| -fast -fma=fused -xipo=2 -xpagesize=4M -xprefetch_level=2 |
Benchmarks using both Fortran and C:
| -fast(cc) -fast(f90) -fma=fused -xipo=2 -xpagesize=4M -xalias_level=std -xprefetch_auto_type=indirect_array_access -xprefetch_level=3 -xprefetch_level=2 |
Base Other Flags
C benchmarks:
| -xjobs=32 -V -# |
C++ benchmarks:
| -xjobs=32 -verbose=diags,version |
Fortran benchmarks:
| -xjobs=32 -V -v |
Benchmarks using both Fortran and C:
| -xjobs=32 -V -# -v |
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 -xpagesize=4M -fma=fused -xipo=2 -xprefetch_level=2 -fsimple=1 -xprefetch_auto_type=indirect_array_access -W2,-Ainline:rs=400 -xalias_level=std |
| 470.lbm: | basepeak = yes |
| 482.sphinx3: | -xprofile=collect:./feedback(pass 1) -xprofile=use:./feedback(pass 2) -fast -xpagesize=4M -fma=fused -xipo=2 -xinline= -xprefetch=no%auto -xalias_level=strong -lfast -ll2amm |
C++ benchmarks:
Fortran benchmarks:
| 410.bwaves: | -fast -xpagesize=4M -fma=fused -xipo=2 -xprefetch_level=2 |
| 416.gamess: | -xprofile=collect:./feedback(pass 1) -xprofile=use:./feedback(pass 2) -fast -xpagesize=4M -fma=fused -xipo=2 -xprefetch=no%auto |
| 434.zeusmp: | -fast -xpagesize=4M -fma=fused -xipo=2 -xprefetch_level=1 -ll2amm |
| 437.leslie3d: | -fast -xpagesize=4M -xprefetch=no |
| 459.GemsFDTD: | -fast -xpagesize=4M -fma=fused -fsimple=1 -xprefetch=no |
| 465.tonto: | -xprofile=collect:./feedback(pass 1) -xprofile=use:./feedback(pass 2) -fast -xpagesize=4M -xipo=2 -xprefetch=no -lfast -ll2amm |
Benchmarks using both Fortran and C:
| 435.gromacs: | -xprofile=collect:./feedback(pass 1) -xprofile=use:./feedback(pass 2) -fast(cc) -fast(f90) -xpagesize=4M -fma=fused -xipo=2 -xchip=generic -xinline= -fsimple=0 |
| 436.cactusADM: | -fast(cc) -fast(f90) -xpagesize=4M -fma=fused -xipo=2 -xprefetch=latx:0.7 -fsimple=1 |
| 454.calculix: | -fast(cc) -fast(f90) -xpagesize=4M -fma=fused -xipo=2 -xprefetch_level=1 -xalias_level=std -xprefetch_auto_type=indirect_array_access |
| 481.wrf: | -xprofile=collect:./feedback(pass 1) -xprofile=use:./feedback(pass 2) -fast(cc) -fast(f90) -xpagesize=4M -xipo=2 -xprefetch_level=2 |