| CPU2006 license: | 11 | Test date: | Oct-2007 |
|---|---|---|---|
| Test sponsor: | IBM Corporation | Hardware Availability: | Nov-2007 |
| Tested by: | IBM Corporation | Software Availability: | Sep-2007 |
| Hardware | |
|---|---|
| CPU Name: | POWER6 |
| CPU Characteristics: | |
| CPU MHz: | 4000 |
| FPU: | Integrated |
| CPU(s) enabled: | 4 cores, 2 chips, 2 cores/chip, 2 threads/core |
| CPU(s) orderable: | 4 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: | None |
| Other Cache: | None |
| Memory: | 16 GB (4x4 GB) DDR2 667 MHz |
| Disk Subsystem: | 1x73 GB SAS 15K RPM |
| Other Hardware: | None |
| Software | |
|---|---|
| Operating System: | SUSE Linux Enterprise 10 SP1 |
| Compiler: | IBM XL C/C++ Advanced Edition for Linux, V9.0 |
| Auto Parallel: | No |
| File System: | ext3 |
| System State: | Multi-User |
| Base Pointers: | 32-bit |
| Peak Pointers: | 32/64-bit |
| Other Software: | -IBM Post-Link Optimization for Linux on POWER, Version 5.4.0-10 -MicroQuill SmartHeap 7.3 |
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 | 8 | 1306 | 59.8 | 1304 | 59.9 | 1304 | 59.9 | 8 | 1058 | 73.9 | 1057 | 74.0 | 1052 | 74.3 |
| 401.bzip2 | 8 | 1068 | 72.3 | 1118 | 69.0 | 1080 | 71.5 | 8 | 974 | 79.3 | 988 | 78.2 | 971 | 79.5 |
| 403.gcc | 8 | 883 | 72.9 | 880 | 73.2 | 881 | 73.1 | 8 | 834 | 77.2 | 832 | 77.4 | 832 | 77.4 |
| 429.mcf | 8 | 581 | 126 | 581 | 126 | 581 | 126 | 8 | 580 | 126 | 579 | 126 | 580 | 126 |
| 445.gobmk | 8 | 1009 | 83.2 | 1007 | 83.3 | 1006 | 83.5 | 8 | 890 | 94.3 | 886 | 94.8 | 887 | 94.6 |
| 456.hmmer | 8 | 982 | 76.0 | 982 | 76.0 | 982 | 76.0 | 8 | 908 | 82.2 | 907 | 82.3 | 907 | 82.3 |
| 458.sjeng | 8 | 1282 | 75.5 | 1281 | 75.6 | 1279 | 75.7 | 8 | 1163 | 83.2 | 1163 | 83.2 | 1163 | 83.2 |
| 462.libquantum | 8 | 1309 | 127 | 1307 | 127 | 1308 | 127 | 8 | 1296 | 128 | 1299 | 128 | 1296 | 128 |
| 464.h264ref | 8 | 1430 | 124 | 1431 | 124 | 1431 | 124 | 8 | 1195 | 148 | 1186 | 149 | 1187 | 149 |
| 471.omnetpp | 8 | 1269 | 39.4 | 1268 | 39.4 | 1269 | 39.4 | 8 | 1229 | 40.7 | 1231 | 40.6 | 1232 | 40.6 |
| 473.astar | 8 | 1102 | 51.0 | 1103 | 50.9 | 1105 | 50.8 | 8 | 964 | 58.2 | 970 | 57.9 | 967 | 58.1 |
| 483.xalancbmk | 8 | 728 | 75.8 | 729 | 75.7 | 729 | 75.8 | 8 | 678 | 81.4 | 679 | 81.2 | 678 | 81.4 |
General Notes
kernel release 2.6.16.53-0.8-ppc64.
See flags file for details on following settings.
ulimit -s (stack) set to unlimited.
The binaries were compiled on a system with 32 GB of memory.
Large pages reserved as follows by root user:
echo 530 > /proc/sys/vm/nr_hugepages
System configured with libhugetlbfs library for application access to large pages
Environment variables set before executing benchmarks.
export HUGETLB_VERBOSE=0
export HUGETLB_MORECORE=yes
export HUGETLB_MORECORE_HEAPBASE=0x50000000
export XLFRTEOPTS=intrinthds=1
fdpr binary optimization tool used for
400.perlbench 401.bzip2 403.gcc 429.mcf 456.hmmer 458.sjeng
462.libquantum 464.h264ref 473.astar 483.xalancbmk
Benchmarks bound to a processor using numactl on the submit command.
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 -qalias=noansi -qalloca -lhugetlbfs |
C++ benchmarks:
| -O5 -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 -qalias=noansi -lsmartheap |
| 401.bzip2: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O4 -lhugetlbfs |
| 403.gcc: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qalloca -q64 -lhugetlbfs |
| 429.mcf: | -Wl,-q -O5 -qnoenablevmx -lhugetlbfs |
| 445.gobmk: | -qpdf1(pass 1) -qpdf2(pass 2) -O4 -qnoenablevmx -lhugetlbfs |
| 456.hmmer: | Same as 401.bzip2 |
| 458.sjeng: | Same as 401.bzip2 |
| 462.libquantum: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qnoenablevmx -q64 -lhugetlbfs |
| 464.h264ref: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O5 -q64 -lhugetlbfs |
C++ benchmarks:
| 471.omnetpp: | -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qrtti -lsmartheap |
| 473.astar: | -Wl,-q -qpdf1(pass 1) -qpdf2(pass 2) -O5 -qnoenablevmx -lsmartheap |
| 483.xalancbmk: | -Wl,-q -O4 -lsmartheap |