| CPU2006 license: | 11 | Test date: | Aug-2010 |
|---|---|---|---|
| Test sponsor: | IBM Corporation | Hardware Availability: | Sep-2010 |
| Tested by: | IBM Corporation | Software Availability: | Aug-2010 |
| Hardware | |
|---|---|
| CPU Name: | POWER7 |
| CPU Characteristics: | Intelligent Energy Optimization enabled, up to 4.14 GHz |
| CPU MHz: | 4004 |
| FPU: | Integrated |
| CPU(s) enabled: | 256 cores, 32 chips, 8 cores/chip, 4 threads/core |
| CPU(s) orderable: | 32,64,96,128,160,192,224,256 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: | 2 TB (256x8 GB) DDR3 1066 MHz |
| Disk Subsystem: | 42x146 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: | IBM XL C/C++ for Linux, V11.1 |
| Auto Parallel: | No |
| File System: | xfs |
| System State: | Run level 5 (multi-user) |
| Base Pointers: | 32-bit |
| Peak Pointers: | 32/64-bit |
| Other Software: | -Post-Link Optimization for Linux on POWER, Version 5.5.0-3 -MicroQuill Smartheap 9 |
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. | ||||||||||||||
| 400.perlbench | 1000 | 1289 | 7580 | 1289 | 7580 | 1281 | 7630 | 1024 | 1114 | 8980 | 1080 | 9270 | 1086 | 9210 |
| 401.bzip2 | 1000 | 1322 | 7300 | 1311 | 7360 | 1321 | 7300 | 1008 | 1193 | 8150 | 1212 | 8030 | 1201 | 8100 |
| 403.gcc | 1000 | 920 | 8750 | 923 | 8720 | 926 | 8690 | 1008 | 903 | 8990 | 906 | 8960 | 903 | 8990 |
| 429.mcf | 1000 | 687 | 13300 | 682 | 13400 | 690 | 13200 | 1008 | 692 | 13300 | 702 | 13100 | 677 | 13600 |
| 445.gobmk | 1000 | 1006 | 10400 | 1008 | 10400 | 1004 | 10400 | 1024 | 853 | 12600 | 852 | 12600 | 856 | 12600 |
| 456.hmmer | 1000 | 1061 | 8800 | 1061 | 8800 | 1060 | 8800 | 1024 | 591 | 16200 | 598 | 16000 | 590 | 16200 |
| 458.sjeng | 1000 | 1197 | 10100 | 1200 | 10100 | 1199 | 10100 | 1024 | 1105 | 11200 | 1106 | 11200 | 1106 | 11200 |
| 462.libquantum | 1000 | 1110 | 18700 | 1113 | 18600 | 1113 | 18600 | 512 | 523 | 20300 | 523 | 20300 | 524 | 20200 |
| 464.h264ref | 1000 | 1737 | 12700 | 1746 | 12700 | 1746 | 12700 | 1024 | 1617 | 14000 | 1639 | 13800 | 1658 | 13700 |
| 471.omnetpp | 1000 | 1094 | 5710 | 1094 | 5710 | 1093 | 5720 | 992 | 1069 | 5800 | 1069 | 5800 | 1068 | 5800 |
| 473.astar | 1000 | 883 | 7950 | 878 | 8000 | 883 | 7950 | 1024 | 777 | 9250 | 780 | 9220 | 781 | 9210 |
| 483.xalancbmk | 1000 | 908 | 7600 | 920 | 7500 | 908 | 7600 | 992 | 751 | 9120 | 749 | 9140 | 751 | 9120 |
Peak Tuning Notes
fdpr binary optimization tool used for:
400.perlbench
with options -O4 -omullX for optimization phase,
and -imullX for instrumentation phase
401.bzip2
with options -O4 -vrox
403.gcc
with options -O4 -nodp -rtb
429.mcf 445.gobmk 458.sjeng 473.astar
with options -O3
456.hmmer
with options -O4 -nodp -m power7
462.libquantum
with options -O4 -vrox -nodp
464.h264ref
with options -O4 -vrox -nodp -rtb
471.omnetpp
with options -O3 -lu -1 -nodp -sdp 9
483.xalancbmk
with options -O3 -m power7
Submit Notes
The config file option 'submit' was used. Benchmarks bound to a processor using numactl on the submit command.
Operating System Notes
ulimit -s (stack) set to 1048576.
Large pages reserved as follows by root user:
echo 56320 > /proc/sys/vm/nr_overcommit_hugepages
The following environment varibles were set before the runspec command:
export HUGETLB_VERBOSE=0
export HUGETLB_MORECORE=yes
export XLFRTEOPTS=intrinthds=1
Base Portability Flags
| 400.perlbench: | -DSPEC_CPU_LINUX_PPC |
| 462.libquantum: | -DSPEC_CPU_LINUX |
| 464.h264ref: | -qchars=signed |
| 483.xalancbmk: | -DSPEC_CPU_LINUX |
Base Optimization Flags
C benchmarks:
| -O5 -qarch=pwr7 -qtune=pwr7 -qalias=noansi -qalloca -lhugetlbfs |
C++ benchmarks:
| -O5 -qarch=pwr7 -qtune=pwr7 -qrtti -lsmartheap |
Base Other Flags
C benchmarks:
| -qipa=noobject -qipa=threads |
C++ benchmarks:
| -qipa=noobject -qipa=threads |
Peak Portability Flags
| 400.perlbench: | -DSPEC_CPU_LINUX_PPC |
| 462.libquantum: | -DSPEC_CPU_LINUX |
| 464.h264ref: | -qchars=signed |
| 483.xalancbmk: | -DSPEC_CPU_LINUX |
Peak Optimization Flags
C benchmarks:
| 400.perlbench: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qarch=pwr7 -qtune=pwr7 -qalias=noansi -qipa=level=2 -lsmartheap |
| 401.bzip2: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O3 -qarch=pwr7 -qtune=pwr7 -lhugetlbfs |
| 403.gcc: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qarch=pwr7 -qtune=pwr7 -qalloca -lhugetlbfs |
| 429.mcf: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qarch=pwr7 -qtune=pwr7 -lhugetlbfs |
| 445.gobmk: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qarch=pwr7 -qtune=pwr7 -lhugetlbfs |
| 456.hmmer: | -Wl,-q -O5 -qarch=pwr7 -qtune=pwr7 -qsimd -qassert=refalign -qipa=inline=threshold=2888 -qipa=inline=limit=11880 -lhugetlbfs |
| 458.sjeng: | Same as 429.mcf |
| 462.libquantum: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qarch=pwr7 -qtune=pwr7 -q64 -lhugetlbfs |
| 464.h264ref: | Same as 429.mcf |
C++ benchmarks:
| 471.omnetpp: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qarch=pwr7 -qtune=pwr7 -qrtti -lsmartheap |
| 473.astar: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qarch=pwr7 -qtune=pwr7 -qipa=inline=threshold=2468 -qipa=inline=limit=11060 -qipa=partition=large -lhugetlbfs -lsmartheap |
| 483.xalancbmk: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qarch=pwr5 -qtune=pwr5 -qipa=inline=threshold=2468 -qipa=inline=limit=11060 -qipa=partition=large -lsmartheap |