| CPU2006 license: | 11 | Test date: | Mar-2009 |
|---|---|---|---|
| Test sponsor: | IBM Corporation | Hardware Availability: | Jun-2008 |
| Tested by: | IBM Corporation | Software Availability: | Mar-2009 |
| Hardware | |
|---|---|
| CPU Name: | POWER6 |
| CPU Characteristics: | |
| CPU MHz: | 5000 |
| FPU: | Integrated |
| CPU(s) enabled: | 64 cores, 32 chips, 2 cores/chip, 2 threads/core |
| CPU(s) orderable: | 8,16,24,32,40,48,56,64 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: | 512 GB (256x2 GB) DDR2 667 MHz |
| Disk Subsystem: | 4x146 GB SCSI 15K RPM |
| Other Hardware: | None |
| Software | |
|---|---|
| Operating System: | SUSE Linux Enterprise Server 11 |
| Compiler: | IBM XL C/C++ for Linux, V10.1 Updated with the Mar2009 PTF. |
| Auto Parallel: | No |
| File System: | ext3 |
| 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 5.4.0-21 -MicroQuill SmartHeap 8.1 |
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. | ||||||||||||||
| 400.perlbench | 128 | 1006 | 1240 | 1007 | 1240 | 1003 | 1250 | 128 | 833 | 1500 | 834 | 1500 | 834 | 1500 |
| 401.bzip2 | 128 | 824 | 1500 | 824 | 1500 | 825 | 1500 | 128 | 734 | 1680 | 734 | 1680 | 734 | 1680 |
| 403.gcc | 128 | 571 | 1800 | 573 | 1800 | 570 | 1810 | 128 | 511 | 2020 | 511 | 2020 | 510 | 2020 |
| 429.mcf | 128 | 404 | 2890 | 402 | 2910 | 401 | 2910 | 128 | 397 | 2940 | 396 | 2950 | 397 | 2940 |
| 445.gobmk | 128 | 813 | 1650 | 812 | 1650 | 813 | 1650 | 128 | 647 | 2080 | 647 | 2070 | 646 | 2080 |
| 456.hmmer | 128 | 825 | 1450 | 824 | 1450 | 824 | 1450 | 128 | 571 | 2090 | 572 | 2090 | 572 | 2090 |
| 458.sjeng | 128 | 986 | 1570 | 985 | 1570 | 985 | 1570 | 128 | 889 | 1740 | 891 | 1740 | 890 | 1740 |
| 462.libquantum | 128 | 542 | 4890 | 543 | 4890 | 543 | 4880 | 64 | 201 | 6580 | 201 | 6590 | 201 | 6590 |
| 464.h264ref | 128 | 977 | 2900 | 977 | 2900 | 977 | 2900 | 128 | 926 | 3060 | 927 | 3060 | 927 | 3060 |
| 471.omnetpp | 128 | 633 | 1260 | 630 | 1270 | 614 | 1300 | 128 | 592 | 1350 | 592 | 1350 | 592 | 1350 |
| 473.astar | 128 | 700 | 1280 | 700 | 1280 | 701 | 1280 | 128 | 611 | 1470 | 610 | 1470 | 611 | 1470 |
| 483.xalancbmk | 128 | 422 | 2090 | 418 | 2110 | 422 | 2090 | 128 | 399 | 2210 | 397 | 2220 | 400 | 2210 |
Submit Notes
The config file option 'submit' was used. Benchmarks bound to a processor using taskset on the submit command.
General Notes
kernel release 2.6.27.19-5-ppc64.
See flags file for details on following settings.
ulimit -s (stack) set to 1048576.
System configured with libhugetlbfs library for application access to large pages
Large pages reserved as follows by root user:
echo 7680 > /proc/sys/vm/nr_hugepages
Environment variables set before executing benchmarks.
export HUGETLB_VERBOSE=0
export HUGETLB_MORECORE=yes
export XLFRTEOPTS=intrinthds=1
IBM Post-Link Optimization tool was used for these benchmarks, with options:
400.perlbench : "-imullX" (instrumentation phase), "-O4 -omullX" (optimization phase)
401.bzip2 : same as 400.perlbench
403.gcc : same as 400.perlbench
456.hmmer : same as 400.perlbench
458.sjeng : same as 400.perlbench
483.xalancbmk : same as 400.perlbench
429.mcf : "-imullX" (instrumentation phase), "-bf -dp -hr -las -pca -RC -RD
-rmte -si -tlo -A 64 -isf 104 -lu 8 -rt 0.16
-hrf 0.18 -ihf 40 -sdp 6 -sdpms 128 -shci 65 -si -sidf 45 -omullX" (optimization phase)
445.gobmk : "-imullX" (instrumentation phase), "-q -O3 -A 32 -omullX" (optimization phase)
462.libquantum : "-imullX" (instrumentation phase), "-bf -dp -lro -nop -RC -RD -tb -tlo -vro -A 4
-isf 88 -lu 8 -hrf 0.10 -sdp 4 -lun 27 -omullX" (optimization phase)
473.astar : "-imullX" (instrumentation phase), "-O4 -omullX -see 1" (optimization phase)
464.h264ref : "-O4" (optimization phase)
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=pwr6 -qtune=pwr6 -qalias=noansi -qalloca -lhugetlbfs |
C++ benchmarks:
| -O5 -qarch=pwr6 -qtune=pwr6 -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 |
| 403.gcc: | -DSPEC_CPU_LP64 |
| 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=pwr6 -qtune=pwr6 -qalias=noansi -lsmartheap |
| 401.bzip2: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O3 -qarch=pwr6 -qtune=pwr6 -lhugetlbfs |
| 403.gcc: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qarch=pwr6 -qtune=pwr6 -qalloca -q64 -lhugetlbfs |
| 429.mcf: | -Wl,-q -O5 -qarch=pwr6 -qtune=pwr6 -qnoenablevmx -lhugetlbfs |
| 445.gobmk: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qarch=pwr6 -qtune=pwr6 -qnoenablevmx -lhugetlbfs |
| 456.hmmer: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qarch=pwr6 -qtune=pwr6 -lhugetlbfs |
| 458.sjeng: | -Wl,-q -O5 -qarch=pwr6 -qtune=pwr6 -lhugetlbfs |
| 462.libquantum: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qarch=pwr6 -qtune=pwr6 -qnoenablevmx -q64 -lhugetlbfs |
| 464.h264ref: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qarch=pwr6 -qtune=pwr6 -q64 -lhugetlbfs |
C++ benchmarks:
| 471.omnetpp: | -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qarch=pwr6 -qtune=pwr6 -qrtti -lsmartheap |
| 473.astar: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qarch=pwr6 -qtune=pwr6 -qnoenablevmx -lsmartheap |
| 483.xalancbmk: | -Wl,-q -O5 -qarch=pwr6 -qtune=pwr6 -lsmartheap |