| CPU2006 license: | 11 | Test date: | Mar-2008 |
|---|---|---|---|
| Test sponsor: | IBM Corporation | Hardware Availability: | May-2008 |
| Tested by: | IBM Corporation | Software Availability: | May-2008 |
| Hardware | |
|---|---|
| CPU Name: | POWER6 |
| CPU Characteristics: | |
| CPU MHz: | 4700 |
| FPU: | Integrated |
| CPU(s) enabled: | 32 cores, 16 chips, 2 cores/chip, 2 threads/core |
| CPU(s) orderable: | 32 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: | 32 MB I+D off chip per chip |
| Other Cache: | None |
| Memory: | 128 GB (64x2 GB) DDR2 533 MHz |
| Disk Subsystem: | 2x146 GB SFF SAS 10K RPM |
| Other Hardware: | None |
| Software | |
|---|---|
| Operating System: | IBM AIX V5.3 with the 5300-08 Technology Level |
| Compiler: | XL C/C++ Enterprise Edition V9 for AIX Updated with the Oct2007 PTF. XL Fortran Enterprise Edition V11.1 for AIX Updated with the Oct2007 PTF. |
| Auto Parallel: | No |
| File System: | AIX/JFS2 |
| System State: | Multi-user |
| Base Pointers: | 32-bit |
| Peak Pointers: | 32/64-bit |
| Other Software: | -- |
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 | 64 | 747 | 1160 | 746 | 1170 | 748 | 1160 | 64 | 747 | 1160 | 746 | 1170 | 748 | 1160 |
| 416.gamess | 64 | 1809 | 693 | 1794 | 699 | 1808 | 693 | 64 | 1702 | 736 | 1716 | 730 | 1708 | 734 |
| 433.milc | 64 | 1403 | 419 | 1402 | 419 | 1403 | 419 | 64 | 1360 | 432 | 1359 | 432 | 1359 | 432 |
| 434.zeusmp | 64 | 1034 | 563 | 1004 | 580 | 979 | 595 | 64 | 616 | 946 | 622 | 936 | 625 | 932 |
| 435.gromacs | 64 | 792 | 577 | 792 | 577 | 792 | 577 | 64 | 732 | 624 | 733 | 624 | 731 | 625 |
| 436.cactusADM | 64 | 1128 | 678 | 1132 | 676 | 1128 | 678 | 64 | 739 | 1030 | 739 | 1040 | 740 | 1030 |
| 437.leslie3d | 64 | 881 | 683 | 881 | 683 | 887 | 678 | 32 | 400 | 753 | 400 | 751 | 401 | 751 |
| 444.namd | 64 | 554 | 927 | 553 | 928 | 553 | 927 | 64 | 534 | 962 | 534 | 960 | 534 | 961 |
| 447.dealII | 64 | 626 | 1170 | 628 | 1170 | 628 | 1170 | 64 | 630 | 1160 | 625 | 1170 | 627 | 1170 |
| 450.soplex | 64 | 984 | 542 | 914 | 584 | 915 | 583 | 64 | 984 | 542 | 914 | 584 | 915 | 583 |
| 453.povray | 64 | 537 | 634 | 537 | 634 | 537 | 634 | 64 | 413 | 825 | 413 | 824 | 413 | 825 |
| 454.calculix | 64 | 675 | 783 | 673 | 784 | 675 | 783 | 64 | 614 | 859 | 607 | 870 | 610 | 866 |
| 459.GemsFDTD | 64 | 1390 | 489 | 1387 | 490 | 1409 | 482 | 64 | 1390 | 489 | 1387 | 490 | 1409 | 482 |
| 465.tonto | 64 | 1179 | 534 | 1148 | 549 | 1146 | 549 | 64 | 661 | 953 | 636 | 990 | 641 | 982 |
| 470.lbm | 64 | 674 | 1300 | 674 | 1310 | 674 | 1310 | 64 | 674 | 1310 | 673 | 1310 | 673 | 1310 |
| 481.wrf | 64 | 1098 | 651 | 1095 | 653 | 1103 | 648 | 64 | 910 | 785 | 928 | 770 | 921 | 776 |
| 482.sphinx3 | 64 | 990 | 1260 | 983 | 1270 | 999 | 1250 | 64 | 926 | 1350 | 932 | 1340 | 932 | 1340 |
General Notes
See flags file of details on following settings. all ulimits set to unlimited. Envionment variables set before executing benchmarks: MALLOCOPTIONS=pool MEMORY_AFFINITY=MCM XLFRTEOPTS=intrinthds=1 System set to "Enhanced" mode when defining partition on HMC. System set to "Chip affinity" mode using the HMC command chsyscfg ... -i "addr_broadcast_perf_policy=chip_affinity" bindprocessor command used on submit to bind each copy to a unique processor. 6400 16M large pages defined with vmo command Remote console disabled in /etc/inittab. fdpr binary optimization tool used for: 410.bwaves 433.milc 435.gromacs 436.cactusADM 453.povray 470.lbm 482.sphinx3
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:
| -bmaxdata:0x40000000 -O5 -qlargepage -D_ILS_MACROS -blpdata |
C++ benchmarks:
| -bmaxdata:0x50000000 -O5 -qlargepage -D_ILS_MACROS -qrtti=all -D__IBM_FAST_VECTOR -blpdata |
Fortran benchmarks:
| -bmaxdata:0x60000000 -O5 -qlargepage -qsmallstack=dynlenonheap -qalias=nostd -blpdata |
Benchmarks using both Fortran and C:
| -bmaxdata:0x60000000 -O5 -qlargepage -D_ILS_MACROS -qsmallstack=dynlenonheap -qalias=nostd -blpdata |
Base Other Flags
C benchmarks:
| -qipa=noobject -qipa=threads -qsuppress=1500-036 |
C++ benchmarks:
| -qipa=noobject -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 -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: | -bmaxdata:0x40000000 -O5 -qlargepage -D_ILS_MACROS -qalign=natural -qfdpr -blpdata |
| 470.lbm: | -O5 -qlargepage -D_ILS_MACROS -qfdpr -q64 -blpdata |
| 482.sphinx3: | -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qlargepage -qenablevmx -qvecnvol -D_ILS_MACROS -qfdpr -blpdata |
C++ benchmarks:
| 444.namd: | -qpdf1(pass 1) -qpdf2(pass 2) -O5 -D_ILS_MACROS |
| 447.dealII: | -bmaxdata:0x50000000 -O5 -qlargepage -D_ILS_MACROS -qrtti=all -D__IBM_FAST_VECTOR -blpdata |
| 450.soplex: | basepeak = yes |
| 453.povray: | -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qlargepage -qenablevmx -qvecnvol -D_ILS_MACROS -qalign=natural -qfdpr -blpdata |
Fortran benchmarks:
| 410.bwaves: | basepeak = yes |
| 416.gamess: | -bmaxdata:0x40000000 -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qalias=nostd |
| 434.zeusmp: | -bmaxdata:0x40000000 -qpdf1(pass 1) -qpdf2(pass 2) -O3 -qarch=auto -qtune=auto -qlargepage -qenablevmx -qvecnvol -qxlf90=nosignedzero -blpdata |
| 437.leslie3d: | -O4 -qlargepage -q64 -blpdata |
| 459.GemsFDTD: | basepeak = yes |
| 465.tonto: | -bmaxdata:0x20000000 -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qlargepage -blpdata |
Benchmarks using both Fortran and C:
| 435.gromacs: | -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qlargepage -qenablevmx -qvecnvol -qfdpr -D_ILS_MACROS -blpdata |
| 436.cactusADM: | -bmaxdata:0x60000000 -qpdf1(pass 1) -qpdf2(pass 2) -O2 -qarch=auto -qtune=auto -qlargepage -qenablevmx -qvecnvol -qfdpr -qnostrict -D_ILS_MACROS -blpdata |
| 454.calculix: | -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qlargepage -D_ILS_MACROS -blpdata |
| 481.wrf: | -bmaxdata:0x30000000 -O5 -qlargepage -D_ILS_MACROS -blpdata |
Peak Other Flags
C benchmarks:
| -qipa=noobject -qipa=threads -qsuppress=1500-036 |
C++ benchmarks:
| -qipa=noobject -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 |