| CPU2006 license: | 6 | Test date: | Apr-2007 |
|---|---|---|---|
| Test sponsor: | Sun Microsystems | Hardware Availability: | Apr-2007 |
| Tested by: | Fujitsu Limited | Software Availability: | May-2007 |
| Hardware | |
|---|---|
| CPU Name: | SPARC64 VI |
| CPU Characteristics: | |
| CPU MHz: | 2400 |
| FPU: | Integrated |
| CPU(s) enabled: | 64 cores, 32 chips, 2 cores/chip, 2 threads/core |
| CPU(s) orderable: | 1 to 8 CMUs; each CMU contains 2 or 4 chips |
| Primary Cache: | 128 KB I + 128 KB D on chip per core |
| Secondary Cache: | 6 MB I+D on chip per chip |
| L3 Cache: | None |
| Other Cache: | None |
| Memory: | 256 GB (256 x 1 GB) |
| Disk Subsystem: | 1095 GB RAID 0 using 15 x 73GB 10,000 RPM Fujitsu ETERNUS4000 Model 80 |
| Other Hardware: | None |
| Software | |
|---|---|
| Operating System: | Solaris 10 11/06 |
| Compiler: | Sun Studio 12 (Early Access) |
| Auto Parallel: | No |
| File System: | ufs |
| System State: | Default |
| Base Pointers: | 32-bit |
| Peak Pointers: | 32-bit |
| Other Software: | None |
| 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 | 127 | 2106 | 820 | 2071 | 833 | 2068 | 834 | 127 | 2106 | 820 | 2071 | 833 | 2068 | 834 |
| 416.gamess | 127 | 3779 | 658 | 3782 | 658 | 3786 | 657 | 127 | 3779 | 658 | 3782 | 658 | 3786 | 657 |
| 433.milc | 127 | 3181 | 367 | 3186 | 366 | 3245 | 359 | 127 | 2840 | 411 | 2869 | 406 | 2812 | 415 |
| 434.zeusmp | 127 | 2388 | 484 | 2391 | 483 | 2381 | 485 | 127 | 2389 | 484 | 2383 | 485 | 2382 | 485 |
| 435.gromacs | 127 | 1375 | 659 | 1375 | 660 | 1375 | 660 | 63 | 627 | 718 | 627 | 718 | 627 | 718 |
| 436.cactusADM | 127 | 1666 | 911 | 1666 | 911 | 1685 | 901 | 127 | 1661 | 913 | 1679 | 904 | 1661 | 914 |
| 437.leslie3d | 127 | 2183 | 547 | 2183 | 547 | 2185 | 546 | 127 | 2027 | 589 | 2011 | 594 | 2021 | 591 |
| 444.namd | 127 | 1584 | 643 | 1579 | 645 | 1579 | 645 | 63 | 769 | 657 | 766 | 660 | 765 | 660 |
| 447.dealII | 127 | 1863 | 780 | 1855 | 783 | 1835 | 792 | 127 | 1622 | 896 | 1602 | 907 | 1613 | 901 |
| 450.soplex | 127 | 2845 | 372 | 2565 | 413 | 2581 | 410 | 127 | 2845 | 372 | 2565 | 413 | 2581 | 410 |
| 453.povray | 127 | 1235 | 547 | 1227 | 551 | 1230 | 549 | 63 | 374 | 896 | 374 | 897 | 373 | 900 |
| 454.calculix | 127 | 1389 | 754 | 1389 | 755 | 1391 | 753 | 127 | 1350 | 776 | 1354 | 774 | 1362 | 769 |
| 459.GemsFDTD | 127 | 3329 | 405 | 3349 | 402 | 3334 | 404 | 127 | 3329 | 405 | 3349 | 402 | 3334 | 404 |
| 465.tonto | 127 | 1893 | 660 | 1911 | 654 | 1892 | 661 | 63 | 826 | 751 | 825 | 751 | 826 | 751 |
| 470.lbm | 127 | 3424 | 510 | 3419 | 510 | 3427 | 509 | 127 | 3299 | 529 | 3301 | 529 | 3297 | 529 |
| 481.wrf | 127 | 2061 | 688 | 2050 | 692 | 1989 | 713 | 127 | 2093 | 678 | 1953 | 726 | 1966 | 722 |
| 482.sphinx3 | 127 | 5341 | 463 | 5312 | 466 | 5283 | 469 | 127 | 4544 | 545 | 4502 | 550 | 4476 | 553 |
Processes were bound to cores using "submit" and "pbind".
These shell commands request use of local 4MB pages:
MPSSHEAP=4MB
MPSSSTACK=4MB
MADV=access_lwp
LD_PRELOAD=mpss.so.1:madv.so.1
'access_lwp' means that the next light weight
process to touch the specified address range
will access it the most heavily.
Stack size set to unlimited via "ulimit -s unlimited"
System Tunables:
(/etc/system parameters)
maxphys=4194304
Defines the maximum size of I/O requests, in bytes.
maxpgio=1024
Defines the maximum number of page I/O requests that can
be queued by the paging system.
tune_t_fsflushr=30
Controls how many seconds elapse between runs of the
page flush daemon, fsflush.
autoup=300
Causes pages older than the listed number of seconds to
be written by fsflush.
bufhwm=3000
Memory byte limit for caching I/O buffers
segmap_percent=1
Set maximum percent memory for file system cache
"CMU" = CPU/Memory Unit; each holds 2 or 4 CPU chips. 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 Fujitsu SPARC Enterprise M9000 and Sun SPARC Enterprise M9000 are electrically equivalent.
| /opt/SUNWspro12_EA070303/bin/cc |
| /opt/SUNWspro12_EA070303/bin/CC |
| /opt/SUNWspro12_EA070303/bin/f90 |
| /opt/SUNWspro12_EA070303/bin/cc /opt/SUNWspro12_EA070303/bin/f90 |
| -fast -xipo=2 -xtarget=sparc64vi -xcache=128/64/2:6144/256/12 -xarch=sparcfmaf -fma=fused -Wc,-fma=fused -xprefetch_level=2 |
| -library=stlport4 -fast -xipo=2 -xtarget=sparc64vi -xcache=128/64/2:6144/256/12 -xarch=sparcfmaf -fma=fused -Qoption cg -fma=fused |
| -fast -xipo=2 -xtarget=sparc64vi -xcache=128/64/2:6144/256/12 -xarch=sparcfmaf -fma=fused -Qoption cg -fma=fused -xprefetch_level=2 |
| -fast(cc) -fast(f90) -xipo=2 -xtarget=sparc64vi -xcache=128/64/2:6144/256/12 -xarch=sparcfmaf -fma=fused -Wc,-fma=fused -xprefetch_level=2 -Qoption cg -fma=fused |
| /opt/SUNWspro12_EA070303/bin/cc |
| /opt/SUNWspro12_EA070303/bin/CC |
| /opt/SUNWspro12_EA070303/bin/f90 |
| /opt/SUNWspro12_EA070303/bin/cc /opt/SUNWspro12_EA070303/bin/f90 |
| 410.bwaves: | basepeak = yes |
| 416.gamess: | basepeak = yes |
| 434.zeusmp: | -fast -xipo=2 -xtarget=sparc64vi -xcache=128/64/2:6144/256/12 -xarch=sparcfmaf -fma=fused -Qoption cg -fma=fused -lmopt |
| 437.leslie3d: | -fast -xipo=2 -xtarget=sparc64vi -xcache=128/64/2:6144/256/12 -xarch=sparcfmaf -fma=fused -Qoption cg -fma=fused -xprefetch_level=2 -xprefetch=latx:8.0 |
| 459.GemsFDTD: | basepeak = yes |
| 465.tonto: | -xprofile=collect:./feedback(pass 1) -xprofile=use:./feedback(pass 2) -fast -xipo=2 -xtarget=sparc64vi -xcache=128/64/2:6144/256/12 -xarch=v8plusa -fma=fused -Qoption cg -fma=fused -lfast |
| 435.gromacs: | -xprofile=collect:./feedback(pass 1) -xprofile=use:./feedback(pass 2) -fast(cc) -fast(f90) -xipo=2 -xtarget=sparc64vi -xcache=128/64/2:6144/256/12 -xarch=sparcfmaf -fma=fused -Wc,-fma=fused -Qoption cg -fma=fused |
| 436.cactusADM: | -fast(cc) -fast(f90) -xipo=2 -xtarget=sparc64vi -xcache=128/64/2:6144/256/12 -xarch=sparcfmaf -fma=fused -Wc,-fma=fused -Qoption cg -fma=fused |
| 454.calculix: | Same as 436.cactusADM |
| 481.wrf: | -xprofile=collect:./feedback(pass 1) -xprofile=use:./feedback(pass 2) -fast(cc) -fast(f90) -xipo=2 -xtarget=sparc64vi -xcache=128/64/2:6144/256/12 -xarch=sparcfmaf -fma=fused -Wc,-fma=fused -Qoption cg -fma=fused -xprefetch_level=2 |