| CPU2006 license: | 11 | Test date: | Mar-2011 |
|---|---|---|---|
| Test sponsor: | IBM Corporation | Hardware Availability: | May-2011 |
| Tested by: | IBM Corporation | Software Availability: | May-2011 |
| Hardware | |
|---|---|
| CPU Name: | POWER7 |
| CPU Characteristics: | Intelligent Energy Optimization enabled, up to 3.86 GHz |
| CPU MHz: | 3612 |
| FPU: | Integrated |
| CPU(s) enabled: | 32 cores, 4 chips, 8 cores/chip, 4 threads/core |
| CPU(s) orderable: | 32 cores |
| Primary Cache: | 32 KB I + 32 KB D on chip per core |
| Secondary Cache: | 256 KB I+D on chip per core |
| L3 Cache: | 4 MB I+D on chip per core |
| Other Cache: | None |
| Memory: | 256 GB (32 x 8 GB) DDR3 1066 MHz |
| Disk Subsystem: | 6 x 146.8 GB Raid0 SAS SFF 15K RPM |
| Other Hardware: | None |
| Software | |
|---|---|
| Operating System: | IBM AIX V7.1 with Service Pack 3 |
| Compiler: | IBM XL C/C++ for AIX, V11.1 Version: 11.01.0000.0005 IBM XL Fortran for AIX, V13.1 Version: 13.01.0000.0005 |
| Auto Parallel: | No |
| File System: | AIX/JFS2 |
| System State: | Multi-user |
| Base Pointers: | 32-bit |
| Peak Pointers: | 32/64-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 | 128 | 2125 | 819 | 2127 | 818 | 2130 | 817 | 128 | 2125 | 819 | 2127 | 818 | 2130 | 817 |
| 416.gamess | 128 | 2478 | 1010 | 2476 | 1010 | 2486 | 1010 | 128 | 2277 | 1100 | 2265 | 1110 | 2273 | 1100 |
| 433.milc | 128 | 1537 | 765 | 1541 | 762 | 1540 | 763 | 32 | 376 | 780 | 376 | 782 | 376 | 782 |
| 434.zeusmp | 128 | 1117 | 1040 | 1104 | 1060 | 1112 | 1050 | 64 | 544 | 1070 | 517 | 1130 | 546 | 1070 |
| 435.gromacs | 128 | 1104 | 828 | 1111 | 823 | 1108 | 824 | 128 | 868 | 1050 | 852 | 1070 | 859 | 1060 |
| 436.cactusADM | 128 | 1174 | 1300 | 1176 | 1300 | 1170 | 1310 | 128 | 1001 | 1530 | 1007 | 1520 | 1002 | 1530 |
| 437.leslie3d | 128 | 2298 | 524 | 2326 | 517 | 2325 | 517 | 32 | 539 | 558 | 541 | 556 | 540 | 557 |
| 444.namd | 128 | 793 | 1300 | 796 | 1290 | 796 | 1290 | 128 | 743 | 1380 | 742 | 1380 | 747 | 1370 |
| 447.dealII | 128 | 689 | 2120 | 688 | 2130 | 684 | 2140 | 128 | 676 | 2160 | 677 | 2160 | 672 | 2180 |
| 450.soplex | 128 | 2083 | 512 | 2075 | 515 | 2056 | 519 | 64 | 1035 | 515 | 1035 | 516 | 1034 | 516 |
| 453.povray | 128 | 599 | 1140 | 596 | 1140 | 599 | 1140 | 128 | 444 | 1530 | 443 | 1540 | 449 | 1520 |
| 454.calculix | 128 | 1095 | 965 | 1087 | 972 | 1102 | 958 | 128 | 1034 | 1020 | 1022 | 1030 | 998 | 1060 |
| 459.GemsFDTD | 128 | 3130 | 434 | 3131 | 434 | 3130 | 434 | 128 | 3130 | 434 | 3131 | 434 | 3130 | 434 |
| 465.tonto | 128 | 1296 | 972 | 1297 | 971 | 1298 | 970 | 128 | 967 | 1300 | 969 | 1300 | 972 | 1300 |
| 470.lbm | 128 | 1520 | 1160 | 1519 | 1160 | 1529 | 1150 | 128 | 1520 | 1160 | 1519 | 1160 | 1529 | 1150 |
| 481.wrf | 128 | 1619 | 883 | 1614 | 886 | 1621 | 882 | 128 | 1619 | 883 | 1614 | 886 | 1621 | 882 |
| 482.sphinx3 | 128 | 3097 | 806 | 3104 | 804 | 3100 | 805 | 128 | 3077 | 811 | 3069 | 813 | 3079 | 810 |
Peak Tuning Notes
fdpr binary optimization tool used for: 433.milc with options: -O4 -nodp -m power7 434.zeusmp with options: -O4 -vrox -nodp -m power7 436.cactusADM 450.soplex with options: -O3 -lu -1 -nodp -sdp 9 -m power7 437.leslie3d with options: -O4 -vrox -m power7 465.tonto with options: -O4 -m power7 482.sphinx3 with options: -O4 -sdp 9 -vrox -m power7
Submit Notes
The config file option 'submit' was used to assign benchmark copy to specific kernel thread using the "bindprocessor" command (see flags file for details).
Operating System Notes
All ulimits set to unlimited. 12800 16M large pages defined with vmo command
General Notes
Environment variables set by runspec before the start of the run: MALLOCOPTIONS = "pool" MEMORY_AFFINITY = "MCM" XLFRTEOPTS = "intrinthds=1" The "IBM Power 750 Express (3.61 GHz)" and the "IBM Power 755 (3.61 GHz)" are electronically equivalent. The results have been measured on the "IBM Power Express 750 (3.61 GHz)".
Base Compiler Invocation
C benchmarks:
| /usr/vac/bin/xlc -qlanglvl=extc99 |
C++ benchmarks:
| /usr/vacpp/bin/xlC |
Fortran benchmarks:
| /usr/bin/xlf95 |
Benchmarks using both Fortran and C:
| /usr/vac/bin/xlc -qlanglvl=extc99 /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: | -DSPEC_CPU_AIX -DNOUNDERSCORE |
| 482.sphinx3: | -qchars=signed |
Base Optimization Flags
C benchmarks:
| -qipa=threads -bmaxdata:0x40000000 -qlargepage -O5 -D_ILS_MACROS -blpdata |
C++ benchmarks:
| -qipa=threads -bmaxdata:0x50000000 -qlargepage -O5 -qsimd -qvecnvol -D_ILS_MACROS -qrtti=all -D__IBM_FAST_VECTOR -D__IBM_FAST_SET_MAP_ITERATOR -blpdata |
Fortran benchmarks:
| -qipa=threads -bmaxdata:0x60000000 -qlargepage -O5 -qsmallstack=dynlenonheap -qalias=nostd -blpdata |
Benchmarks using both Fortran and C:
| -qipa=threads -bmaxdata:0x60000000 -qlargepage -O5 -D_ILS_MACROS -qsmallstack=dynlenonheap -qalias=nostd -blpdata |
Base Other Flags
C benchmarks:
| -qipa=noobject -qsuppress=1500-036 |
C++ benchmarks:
| -qipa=noobject -qsuppress=1500-036 |
Fortran benchmarks:
| -qipa=noobject -qsuppress=1500-010 -qsuppress=cmpmsg -qsuppress=1500-036 |
Benchmarks using both Fortran and C:
| -qipa=noobject -qsuppress=1500-010 -qsuppress=cmpmsg -qsuppress=1500-036 |
Peak Compiler Invocation
C benchmarks:
| /usr/vac/bin/xlc -qlanglvl=extc99 |
C++ benchmarks:
| /usr/vacpp/bin/xlC |
Fortran benchmarks:
| /usr/bin/xlf95 |
Benchmarks using both Fortran and C:
| /usr/vac/bin/xlc -qlanglvl=extc99 /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: | -DSPEC_CPU_AIX -DNOUNDERSCORE |
| 482.sphinx3: | -qchars=signed |
Peak Optimization Flags
C benchmarks:
| 433.milc: | -qipa=threads -bmaxdata:0x40000000 -O5 -qlargepage -D_ILS_MACROS -qrestrict -qprefetch=aggressive -qalign=natural -blpdata -btextpsize:64K |
| 470.lbm: | basepeak = yes |
| 482.sphinx3: | -qipa=threads -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qlargepage -D_ILS_MACROS -blpdata -btextpsize:64K |
C++ benchmarks:
| 444.namd: | -qipa=threads -qpdf1(pass 1) -qpdf2(pass 2) -O5 -q64 -qsimd -qvecnvol -qlargepage -D_ILS_MACROS -blpdata -btextpsize:64K |
| 447.dealII: | -qipa=threads -bmaxdata:0x50000000 -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qsimd -qvecnvol -D_ILS_MACROS -qrtti=all -D__IBM_FAST_VECTOR -D__IBM_FAST_SET_MAP_ITERATOR -blpdata -btextpsize:64K |
| 450.soplex: | -qpdf1(pass 1) -qpdf2(pass 2) -O4 -q64 -qlargepage -D_ILS_MACROS -blpdata -btextpsize:64K |
| 453.povray: | -qipa=threads -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qsimd -qvecnvol -qlargepage -D_ILS_MACROS -qalign=natural -blpdata -btextpsize:64K |
Fortran benchmarks:
| 410.bwaves: | basepeak = yes |
| 416.gamess: | -qipa=threads -bmaxdata:0x40000000 -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qarch=pwr5 -qlargepage -qalias=nostd -blpdata -btextpsize:64K |
| 434.zeusmp: | -bmaxdata:0x40000000 -O3 -qarch=auto -qtune=auto -qlargepage -qxlf90=nosignedzero -blpdata -btextpsize:64K |
| 437.leslie3d: | -O5 -q64 -blpdata -btextpsize:64K |
| 459.GemsFDTD: | basepeak = yes |
| 465.tonto: | -qipa=threads -bmaxdata:0x50000000 -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qsimd -qvecnvol -blpdata -btextpsize:64K |
Benchmarks using both Fortran and C:
| 435.gromacs: | -qipa=threads -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qsimd -qvecnvol -D_ILS_MACROS -blpdata -btextpsize:64K |
| 436.cactusADM: | -qipa=threads -bmaxdata:0x60000000 -O4 -qsimd -qvecnvol -D_ILS_MACROS -qnostrict -blpdata -btextpsize:64K |
| 454.calculix: | -qipa=threads -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qsimd -qvecnvol -qlargepage -D_ILS_MACROS -blpdata -btextpsize:64K |
| 481.wrf: | basepeak = yes |
Peak Other Flags
C benchmarks:
| -qipa=noobject -qsuppress=1500-036 |
C++ benchmarks (except as noted below):
| -qipa=noobject -qsuppress=1500-036 | |
| 450.soplex: | -qsuppress=1500-036 |
Fortran benchmarks (except as noted below):
Benchmarks using both Fortran and C:
| -qipa=noobject -qsuppress=1500-010 -qsuppress=cmpmsg -qsuppress=1500-036 |