| CPU2006 license: | 11 | Test date: | Nov-2007 |
|---|---|---|---|
| Test sponsor: | IBM Corporation | Hardware Availability: | Nov-2007 |
| Tested by: | IBM Corporation | Software Availability: | Nov-2007 |
| Hardware | |
|---|---|
| CPU Name: | POWER6 |
| CPU Characteristics: | |
| CPU MHz: | 4000 |
| FPU: | Integrated |
| CPU(s) enabled: | 4 cores, 2 chips, 2 cores/chip, 2 threads/core |
| CPU(s) orderable: | 4 cores |
| Primary Cache: | 64 KB I + 64 KB D on chip per core |
| Secondary Cache: | 4 MB I+D on chip per core |
| L3 Cache: | None |
| Other Cache: | None |
| Memory: | 16 GB (4x4 GB) DDR2 667 MHz |
| Disk Subsystem: | 1x146 GB SAS 15K RPM |
| Other Hardware: | None |
| Software | |
|---|---|
| Operating System: | IBM AIX 5L V5.3 |
| Compiler: | XL C/C++ Enterprise Edition Version 9.0 for AIX XL Fortran Enterprise Edition Version 11.1 for AIX |
| Auto Parallel: | No |
| File System: | AIX/JFS2 |
| System State: | Multi-user |
| Base Pointers: | 32-bit |
| Peak Pointers: | 32/64-bit |
| Other Software: | IBM Engineering and Scientific Subroutine Library (ESSL) version 4.3.0.2 |
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 | 8 | 1164 | 93.4 | 1161 | 93.7 | 1161 | 93.6 | 8 | 1161 | 93.6 | 1161 | 93.6 | 1161 | 93.6 |
| 416.gamess | 8 | 2089 | 75.0 | 2094 | 74.8 | 2092 | 74.9 | 8 | 2004 | 78.2 | 2002 | 78.2 | 2004 | 78.2 |
| 433.milc | 8 | 1596 | 46.0 | 1595 | 46.0 | 1595 | 46.0 | 8 | 1526 | 48.1 | 1526 | 48.1 | 1526 | 48.1 |
| 434.zeusmp | 8 | 1139 | 63.9 | 1139 | 63.9 | 1116 | 65.2 | 8 | 734 | 99.2 | 714 | 102 | 712 | 102 |
| 435.gromacs | 8 | 928 | 61.6 | 926 | 61.7 | 927 | 61.6 | 8 | 867 | 65.9 | 866 | 65.9 | 866 | 66.0 |
| 436.cactusADM | 8 | 1396 | 68.5 | 1396 | 68.5 | 1395 | 68.5 | 8 | 1396 | 68.5 | 1396 | 68.5 | 1395 | 68.5 |
| 437.leslie3d | 8 | 1325 | 56.8 | 1327 | 56.7 | 1327 | 56.7 | 8 | 1286 | 58.5 | 1289 | 58.3 | 1290 | 58.3 |
| 444.namd | 8 | 648 | 99.0 | 647 | 99.1 | 648 | 99.0 | 8 | 625 | 103 | 625 | 103 | 625 | 103 |
| 447.dealII | 8 | 693 | 132 | 694 | 132 | 692 | 132 | 8 | 693 | 132 | 694 | 132 | 692 | 132 |
| 450.soplex | 8 | 1143 | 58.4 | 1143 | 58.4 | 1141 | 58.5 | 8 | 1143 | 58.4 | 1143 | 58.4 | 1141 | 58.5 |
| 453.povray | 8 | 656 | 64.9 | 656 | 64.9 | 655 | 64.9 | 8 | 501 | 84.9 | 501 | 84.9 | 501 | 85.0 |
| 454.calculix | 8 | 827 | 79.8 | 827 | 79.8 | 825 | 80.0 | 8 | 753 | 87.6 | 752 | 87.8 | 752 | 87.8 |
| 459.GemsFDTD | 8 | 1854 | 45.8 | 1857 | 45.7 | 1864 | 45.5 | 8 | 1854 | 45.8 | 1857 | 45.7 | 1864 | 45.5 |
| 465.tonto | 8 | 1450 | 54.3 | 1440 | 54.7 | 1437 | 54.8 | 8 | 845 | 93.2 | 832 | 94.6 | 831 | 94.7 |
| 470.lbm | 8 | 2296 | 47.9 | 2296 | 47.9 | 2296 | 47.9 | 8 | 2296 | 47.9 | 2296 | 47.9 | 2296 | 47.9 |
| 481.wrf | 8 | 1357 | 65.9 | 1355 | 66.0 | 1357 | 65.9 | 8 | 1098 | 81.4 | 1096 | 81.5 | 1095 | 81.6 |
| 482.sphinx3 | 8 | 1764 | 88.4 | 1756 | 88.8 | 1748 | 89.2 | 8 | 1715 | 90.9 | 1709 | 91.2 | 1769 | 88.1 |
General Notes
AIX 5L V5.3 updated with the 5300-07 Technology Level.
IBM XL Compilers updated to October 2007 PTF Level.
See flags file for details on following settings:
all ulimits set to unlimited
Envionment variables set before executing benchmarks:
MALLOCOPTIONS=pool
MEMORY_AFFINITY=MCM
XLFRTEOPTS=intrinthds=1
768 pages of size 16M defined on systems with vmo command
fdpr binary optimization tool used for peak versions of: 433.milc, 453.povray, 470.lbm, 482.sphinx3
submit used to bind benchmark to a processor using "bindprocessor"
The binaries were compiled on a system with 32 GB of memory.
Base Compiler Invocation
C benchmarks:
| /usr/vac/bin/xlc |
C++ benchmarks:
| /usr/vacpp/bin/xlC |
Fortran benchmarks:
| /usr/bin/xlf95 |
Benchmarks using both Fortran and C:
| /usr/vac/bin/xlc /usr/bin/xlf95 |
Base Portability Flags
| 410.bwaves: | -qfixed |
| 416.gamess: | -qfixed |
| 434.zeusmp: | -qfixed |
| 435.gromacs: | -qfixed -qextname |
| 436.cactusADM: | -qfixed -qextname |
| 437.leslie3d: | -qfixed |
| 454.calculix: | -qfixed -qextname |
| 481.wrf: | -DNOUNDERSCORE -DSPEC_CPU_AIX |
| 482.sphinx3: | -qchars=signed |
Base Optimization Flags
C benchmarks:
| -qlanglvl=extc99 -bmaxdata:0x40000000 -O5 -qlargepage |
C++ benchmarks:
| -bmaxdata:0x50000000 -O5 -qlargepage -qrtti=all -D__IBM_FAST_VECTOR |
Fortran benchmarks:
| -bmaxdata:0x60000000 -O5 -qlargepage -qsmallstack=dynlenonheap -qalias=nostd |
Benchmarks using both Fortran and C:
| -qlanglvl=extc99 -bmaxdata:0x60000000 -O5 -qlargepage -qsmallstack=dynlenonheap -qalias=nostd |
Base Other Flags
C benchmarks:
| -qipa=threads -qsuppress=1500-036 |
C++ benchmarks:
| -qipa=threads -qsuppress=1500-036 |
Fortran benchmarks:
| -qipa=noobject -qsuppress=1500-010 -qsuppress=cmpmsg -qipa=threads -qsuppress=1500-036 |
Benchmarks using both Fortran and C:
| -qipa=noobject -qsuppress=1500-010 -qsuppress=cmpmsg -qipa=threads -qsuppress=1500-036 |
Peak Compiler Invocation
C benchmarks:
| /usr/vac/bin/xlc |
C++ benchmarks:
| /usr/vacpp/bin/xlC |
Fortran benchmarks:
| /usr/bin/xlf95 |
Benchmarks using both Fortran and C:
| /usr/vac/bin/xlc /usr/bin/xlf95 |
Peak Portability Flags
| 410.bwaves: | -qfixed |
| 416.gamess: | -qfixed |
| 434.zeusmp: | -qfixed |
| 435.gromacs: | -qfixed -qextname |
| 436.cactusADM: | -qfixed -qextname |
| 437.leslie3d: | -qfixed |
| 454.calculix: | -qfixed -qextname |
| 481.wrf: | -DNOUNDERSCORE -DSPEC_CPU_AIX |
| 482.sphinx3: | -qchars=signed |
Peak Optimization Flags
C benchmarks:
| 433.milc: | -qlanglvl=extc99 -O5 -qfdpr -qlargepage -bmaxdata:0x40000000 -blpdata |
| 470.lbm: | -qlanglvl=extc99 -O5 -qlargepage -qfdpr -q64 -qenablevmx -qhot=simd -blpdata |
| 482.sphinx3: | -qlanglvl=extc99 -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qlargepage -qfdpr -blpdata |
C++ benchmarks:
| 444.namd: | -qpdf1(pass 1) -qpdf2(pass 2) -O5 |
| 447.dealII: | basepeak = yes |
| 450.soplex: | basepeak = yes |
| 453.povray: | -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qfdpr -qenablevmx -qvecnvol |
Fortran benchmarks:
| 410.bwaves: | -O5 -qlargepage -qsmallstack=dynlenonheap -bmaxdata:0x50000000 -blpdata |
| 416.gamess: | -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qalias=nostd -bmaxdata:0x60000000 |
| 434.zeusmp: | -qpdf1(pass 1) -qpdf2(pass 2) -O3 -qarch=pwr6e -qtune=pwr6 -qxlf90=nosignedzero -qlargepage -qenablevmx -bmaxdata:0x40000000 -blpdata |
| 437.leslie3d: | -O4 -qlargepage -q64 -blpdata |
| 459.GemsFDTD: | basepeak = yes |
| 465.tonto: | -bmaxdata:0x60000000 -blpdata -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qlargepage -lessl |
Benchmarks using both Fortran and C:
| 435.gromacs: | -qlanglvl=extc99 -qpdf1(pass 1) -qpdf2(pass 2) -O5 -bdatapsize:64K -bstackpsize:64K |
| 436.cactusADM: | basepeak = yes |
| 454.calculix: | -qlanglvl=extc99 -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qlargepage -blpdata |
| 481.wrf: | -qlanglvl=extc99 -bmaxdata:0x60000000 -O5 -qlargepage -qsmallstack=dynlenonheap |
Peak Other Flags
C++ benchmarks:
| 447.dealII: | -qipa=threads -qsuppress=1500-036 |
| 450.soplex: | -qipa=threads -qsuppress=1500-036 |
Fortran benchmarks:
| 459.GemsFDTD: | -qipa=noobject -qsuppress=1500-010 -qsuppress=cmpmsg -qipa=threads -qsuppress=1500-036 |
Benchmarks using both Fortran and C:
| 436.cactusADM: | -qipa=noobject -qsuppress=1500-010 -qsuppress=cmpmsg -qipa=threads -qsuppress=1500-036 |
| 481.wrf: | -qipa=noobject -qsuppress=1500-010 -qsuppress=cmpmsg -qipa=threads -qsuppress=1500-036 |