| CPU2006 license: | 11 | Test date: | Aug-2011 |
|---|---|---|---|
| Test sponsor: | IBM Corporation | Hardware Availability: | Oct-2011 |
| Tested by: | IBM Corporation | Software Availability: | Jul-2011 |
| Hardware | |
|---|---|
| CPU Name: | POWER7 |
| CPU Characteristics: | Intelligent Energy Optimization enabled, up to 3.780 GHz |
| CPU MHz: | 3444 |
| FPU: | Integrated |
| CPU(s) enabled: | 96 cores, 16 chips, 6 cores/chip, 4 threads/core |
| CPU(s) orderable: | 24,48,72,96 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: | 1 TB (64 x 16 GB) DDR3 1066 MHz |
| Disk Subsystem: | 10 x 146.8 GB Raid0 SAS SFF 15K RPM |
| Other Hardware: | None |
| Software | |
|---|---|
| Operating System: | SUSE Linux Enterprise Server 11 SP1 (ppc64), Kernel 2.6.32.12-0.7-ppc64 |
| Compiler: | C/C++: Version 11.1 of IBM XL C/C++ for Linux; Fortran: Version 13.1 of IBM XL Fortran for Linux |
| Auto Parallel: | No |
| File System: | ext2 |
| System State: | Run level 3 (multi-user) |
| Base Pointers: | 32-bit |
| Peak Pointers: | 32/64-bit |
| Other Software: | -IBM Post-Link Optimization for Linux on POWER, version 5.6.0-4 -MicroQuill SmartHeap 9 -Apache C++ Standard Library V4.2.1 |
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 | 384 | 1958 | 2670 | 1969 | 2650 | 1970 | 2650 | 384 | 1958 | 2670 | 1958 | 2670 | 1956 | 2670 |
| 416.gamess | 384 | 2534 | 2970 | 2538 | 2960 | 2544 | 2960 | 384 | 2381 | 3160 | 2383 | 3160 | 2382 | 3160 |
| 433.milc | 384 | 1259 | 2800 | 1240 | 2840 | 1245 | 2830 | 96 | 299 | 2950 | 296 | 2970 | 297 | 2960 |
| 434.zeusmp | 384 | 1303 | 2680 | 1301 | 2690 | 1304 | 2680 | 384 | 1100 | 3180 | 1098 | 3180 | 1100 | 3180 |
| 435.gromacs | 384 | 1135 | 2420 | 1137 | 2410 | 1129 | 2430 | 384 | 881 | 3110 | 880 | 3120 | 884 | 3100 |
| 436.cactusADM | 384 | 1245 | 3690 | 1247 | 3680 | 1244 | 3690 | 192 | 525 | 4370 | 523 | 4390 | 525 | 4370 |
| 437.leslie3d | 384 | 1912 | 1890 | 1911 | 1890 | 1920 | 1880 | 96 | 435 | 2070 | 435 | 2080 | 434 | 2080 |
| 444.namd | 384 | 757 | 4070 | 755 | 4080 | 754 | 4090 | 384 | 749 | 4110 | 750 | 4110 | 744 | 4140 |
| 447.dealII | 384 | 623 | 7050 | 633 | 6940 | 633 | 6940 | 384 | 632 | 6950 | 639 | 6880 | 632 | 6960 |
| 450.soplex | 384 | 1914 | 1670 | 1911 | 1680 | 1919 | 1670 | 384 | 1847 | 1730 | 1846 | 1740 | 1847 | 1730 |
| 453.povray | 384 | 597 | 3420 | 602 | 3390 | 597 | 3420 | 384 | 494 | 4140 | 490 | 4170 | 496 | 4120 |
| 454.calculix | 384 | 1115 | 2840 | 1126 | 2810 | 1119 | 2830 | 384 | 1109 | 2860 | 1097 | 2890 | 1090 | 2910 |
| 459.GemsFDTD | 384 | 3131 | 1300 | 3134 | 1300 | 3136 | 1300 | 384 | 3131 | 1300 | 3134 | 1300 | 3136 | 1300 |
| 465.tonto | 384 | 1461 | 2590 | 1460 | 2590 | 1458 | 2590 | 384 | 1185 | 3190 | 1187 | 3180 | 1182 | 3200 |
| 470.lbm | 384 | 1209 | 4360 | 1215 | 4340 | 1214 | 4350 | 384 | 1209 | 4360 | 1215 | 4340 | 1214 | 4350 |
| 481.wrf | 384 | 1474 | 2910 | 1494 | 2870 | 1495 | 2870 | 384 | 1441 | 2980 | 1444 | 2970 | 1443 | 2970 |
| 482.sphinx3 | 384 | 2658 | 2820 | 2689 | 2780 | 2683 | 2790 | 384 | 2647 | 2830 | 2646 | 2830 | 2639 | 2840 |
Compiler Invocation Notes
C/C++ compiler updated to July2011 PTF
Version 11.01.0000.0003
Fortran compiler updated to July2011 PTF
Version 13.01.0000.0003
Peak Tuning Notes
IBM Post-Link optimization tool used for:
433.milc 435.gromacs 436.cactusADM 450.soplex 482.sphinx3
with options -O4 -nodp
444.namd
with options -O3 -lu -1 -nodp -sdp 9
465.tonto
with options -O4
470.lbm
with options -kr -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 "numactl" command (see flags file for details).
Operating System Notes
ulimit -s (stack) set to 2097152
Large pages reserved as follows by root user:
echo 25728 > /proc/sys/vm/nr_hugepages
The following environment varibles were set before the runspec command:
export XLFRTEOPTS=intrinthds=1
export HUGETLB_VERBOSE=0
export HUGETLB_MORECORE=yes
export HUGETLB_ELFMAP=RW
General Notes
447.dealII (peak): "apache_stdcxx_4_2_1" src.alt was used. 447.dealII (base): "apache_stdcxx_4_2_1" src.alt was used. The Apache C++ Standard Library V4.2.1 was installed from http://stdcxx.apache.org/download.html using: gmake BUILDTYPE=8d CONFIG=gcc.config IBM Post-Link optimization tool can be downloaded from http://www-304.ibm.com/webapp/set2/sas/f/lopdiags/sdkdownload.html
Base Compiler Invocation
C benchmarks:
| xlc -qlanglvl=extc99 |
C++ benchmarks:
| xlC |
Fortran benchmarks:
| xlf95 |
Benchmarks using both Fortran and C:
| xlc -qlanglvl=extc99 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 |
| 482.sphinx3: | -qchars=signed |
Base Optimization Flags
C benchmarks:
| -O5 -lhugetlbfs |
C++ benchmarks:
| -O4 -qrtti -qcpp_stdinc=/root/stdcxx421/include/ansi:/root/stdcxx421/include:/opt/ibmcmp/vacpp/11.1/include -lhugetlbfs -L/root/stdcxx421/lib -R/root/stdcxx421/lib -lstd8d |
Fortran benchmarks:
| -O5 -qalias=nostd -lhugetlbfs |
Benchmarks using both Fortran and C:
| -O5 -qalias=nostd -lhugetlbfs |
Base Other Flags
C benchmarks:
| -qipa=noobject -qipa=threads |
C++ benchmarks:
| -qipa=noobject -qipa=threads |
Fortran benchmarks:
| -qipa=noobject -qipa=threads |
Benchmarks using both Fortran and C:
| -qipa=noobject -qipa=threads |
Peak Compiler Invocation
C benchmarks:
| xlc -qlanglvl=extc99 |
C++ benchmarks:
| xlC |
Fortran benchmarks:
| xlf95 |
Benchmarks using both Fortran and C:
| xlc -qlanglvl=extc99 xlf95 |
Peak Portability Flags
| 410.bwaves: | -qfixed |
| 416.gamess: | -qfixed |
| 434.zeusmp: | -qfixed |
| 435.gromacs: | -qfixed -qextname |
| 436.cactusADM: | -qfixed -qextname -DSPEC_CPU_LP64 |
| 437.leslie3d: | -qfixed |
| 453.povray: | -DSPEC_CPU_LP64 |
| 454.calculix: | -qfixed -qextname |
| 481.wrf: | -DNOUNDERSCORE |
| 482.sphinx3: | -qchars=signed |
Peak Optimization Flags
C benchmarks:
| 433.milc: | -Wl,-q -O5 -lhugetlbfs |
| 470.lbm: | basepeak = yes |
| 482.sphinx3: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O4 -lhugetlbfs |
C++ benchmarks:
| 444.namd: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O5 -lhugetlbfs |
| 447.dealII: | -O4 -qrtti -qcpp_stdinc=/root/stdcxx421/include/ansi:/root/stdcxx421/include:/opt/ibmcmp/vacpp/11.1/include -lsmartheap -lhugetlbfs -L/root/stdcxx421/lib -R/root/stdcxx421/lib -lstd8d |
| 450.soplex: | -Wl,-q -O3 -qarch=auto -qtune=auto -lhugetlbfs |
| 453.povray: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qsimd -q64 -lsmartheap64 |
Fortran benchmarks:
| 410.bwaves: | -qpdf1(pass 1) -qpdf2(pass 2) -O5 -q64 -lhugetlbfs |
| 416.gamess: | -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qalias=nostd -lhugetlbfs |
| 434.zeusmp: | -O5 -qsmallstack=dynlenonheap -qalias=nostd -B/usr/share/libhugetlbfs/ -tl -Wl,--hugetlbfs-align |
| 437.leslie3d: | -O5 -lhugetlbfs |
| 459.GemsFDTD: | basepeak = yes |
| 465.tonto: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qsimd -lhugetlbfs |
Benchmarks using both Fortran and C:
| 435.gromacs: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qsimd -lhugetlbfs |
| 436.cactusADM: | -Wl,-q -O4 -q64 -qsimd -qnostrict -qsmallstack=dynlenonheap -qalias=nostd -lhugetlbfs |
| 454.calculix: | -qpdf1(pass 1) -qpdf2(pass 2) -O5 -lhugetlbfs |
| 481.wrf: | -O3 -qarch=auto -qtune=auto -q64 -lhugetlbfs |