SPEC® CFP2006 Result

Copyright 2006-2014 Standard Performance Evaluation Corporation

IBM Corporation

IBM System p 570 (4.7 GHz, 8 core, RHEL)

CPU2006 license: 11 Test date: Oct-2007
Test sponsor: IBM Corporation Hardware Availability: Jun-2007
Tested by: IBM Corporation Software Availability: Oct-2007
Benchmark results graph
Hardware
CPU Name: POWER6
CPU Characteristics:
CPU MHz: 4700
FPU: Integrated
CPU(s) enabled: 8 cores, 4 chips, 2 cores/chip, 2 threads/core
CPU(s) orderable: 2,4,8,12,16 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: 64 GB (32x2 GB) DDR2 667 MHz
Disk Subsystem: 2x73 GB SAS 15K RPM
Other Hardware: None
Software
Operating System: Red Hat Enterprise Linux Advanced Platform 5.1
for IBM POWER
Compiler: IBM XL C/C++ Advanced Edition for Linux, V9.0
IBM XL Fortran Advanced Edition for Linux, V11.1
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 8.1
-IBM Engineering and Scientific Subroutine
Library for Linux on POWER, Version 4.3

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.
410.bwaves 16 912 238 918 237 918 237 16 907 240 907 240 906 240
416.gamess 16 1790 175 1790 175 1790 175 16 1689 185 1690 185 1689 185
433.milc 16 1130 130 1128 130 1128 130 16 1130 130 1128 130 1128 130
434.zeusmp 16 910 160 911 160 906 161 16 558 261 554 263 554 263
435.gromacs 16 802 143 800 143 801 143 16 726 157 726 157 726 157
436.cactusADM 16 1148 167 1147 167 1152 166 16 802 238 807 237 807 237
437.leslie3d 16 846 178 847 178 843 178 8 366 205 366 205 371 203
444.namd 16 668 192 666 193 666 193 16 644 199 645 199 645 199
447.dealII 16 612 299 612 299 612 299 16 597 306 594 308 596 307
450.soplex 16 658 203 629 212 628 212 16 622 215 622 214 622 215
453.povray 16 541 157 541 157 541 157 16 427 199 427 199 427 199
454.calculix 16 688 192 687 192 689 192 16 592 223 593 223 593 223
459.GemsFDTD 16 1213 140 1212 140 1214 140 16 1173 145 1173 145 1171 145
465.tonto 16 1573 100 1575 100 1556 101 16 701 225 703 224 700 225
470.lbm 16 588 374 588 374 588 374 16 588 374 588 374 588 374
481.wrf 16 1223 146 1223 146 1223 146 16 1093 164 1094 163 1097 163
482.sphinx3 16 950 328 977 319 949 329 16 896 348 911 342 896 348

General Notes

 kernel release 2.6.18-52.el5.

 See flags file for details on following settings.

 ulimit -s (stack) set to 262144.


 System set to Enhanced mode when defining partition on HMC

 Large pages reserved as follows by root user:
     echo 1600 > /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 XLFRTEOPTS=intrinthds=1

 fdpr binary optimization tool used for
     435.gromacs 436.cactusADM 482.sphinx3

 Benchmarks bound to a processor using numactl on the submit command.

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   -qnoenablevmx   -lhugetlbfs 

C++ benchmarks:

 -O5   -qrtti   -qnoenablevmx   -qstaticlink 

Fortran benchmarks:

 -O5   -qsmallstack=dynlenonheap   -qalias=nostd   -qnoenablevmx   -B/usr/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-link=BDT 

Benchmarks using both Fortran and C:

 -O5   -qnoenablevmx   -qsmallstack=dynlenonheap   -qalias=nostd   -B/usr/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-link=BDT 

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 
437.leslie3d:  -qfixed 
454.calculix:  -qfixed   -qextname 
481.wrf:  -DNOUNDERSCORE 
482.sphinx3:  -qchars=signed 

Peak Optimization Flags

C benchmarks:

433.milc:  basepeak = yes 
470.lbm:  -O3   -qarch=pwr6e   -qtune=pwr6   -B/usr/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-link=BDT   -q64 
482.sphinx3:  -Wl,-q   -qpdf1(pass 1)   -qpdf2(pass 2)   -O4   -lhugetlbfs 

C++ benchmarks:

444.namd:  -qpdf1(pass 1)   -qpdf2(pass 2)   -O3   -qarch=pwr6e   -qtune=pwr6 
447.dealII:  -O5   -qrtti   -qnoenablevmx   -qstaticlink   -Wl,--whole-archive /usr/lib/libsmartheap.a   -Wl,--no-whole-archive 
450.soplex:  -qpdf1(pass 1)   -qpdf2(pass 2)   -O4   -qstrict   -lhugetlbfs 
453.povray:  -qpdf1(pass 1)   -qpdf2(pass 2)   -O5   -lsmartheap 

Fortran benchmarks:

410.bwaves:  -O5   -qsmallstack=dynlenonheap   -lhugetlbfs 
416.gamess:  -qpdf1(pass 1)   -qpdf2(pass 2)   -O5   -qalias=nostd   -qnoenablevmx 
434.zeusmp:  -qpdf1(pass 1)   -qpdf2(pass 2)   -O3   -qarch=pwr6e   -qtune=pwr6   -qxlf90=nosignedzero   -B/usr/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-link=BDT 
437.leslie3d:  -O3   -qarch=pwr6e   -qtune=pwr6   -B/usr/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-link=BDT   -q64 
459.GemsFDTD:  -qpdf1(pass 1)   -qpdf2(pass 2)   -O5   -B/usr/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-link=BDT   -q64 
465.tonto:  -qpdf1(pass 1)   -qpdf2(pass 2)   -O5   -lessl   -lsmartheap   -lxlf90_r 

Benchmarks using both Fortran and C:

435.gromacs:  -Wl,-q   -O2   -qarch=pwr6e   -qtune=pwr6   -lhugetlbfs 
436.cactusADM:  -Wl,-q   -qpdf1(pass 1)   -qpdf2(pass 2)   -O2   -qarch=pwr6e   -qtune=pwr6   -lhugetlbfs 
454.calculix:  -qpdf1(pass 1)   -qpdf2(pass 2)   -O4   -B/usr/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-link=BDT 
481.wrf:  -O5   -qnoenablevmx   -qalias=nostd   -lhugetlbfs 

Peak 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 

The flags file that was used to format this result can be browsed at
http://www.spec.org/cpu2006/flags/lop-xl-flags.20090714.html.

You can also download the XML flags source by saving the following link:
http://www.spec.org/cpu2006/flags/lop-xl-flags.20090714.xml.