SPEC® CFP2006 Result

IBM Corporation

IBM PowerLinux 7R4 (4.0 GHz, 32 core, RHEL,
GCC)

SPECfp®_rate2006 = 1110

SPECfp_rate_base2006 = 983

CPU2006 license:	11	Test date:	May-2013
Test sponsor:	IBM Corporation	Hardware Availability:	Aug-2013
Tested by:	IBM Corporation	Software Availability:	May-2013

Hardware
CPU Name:	POWER7+
CPU Characteristics:	Intelligent Energy Optimization enabled, up to 4.431 GHz
CPU MHz:	4060
FPU:	Integrated
CPU(s) enabled:	32 cores, 8 chips, 4 cores/chip, 4 threads/core
CPU(s) orderable:	16, 32 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:	10 MB I+D on chip per core
Other Cache:	None
Memory:	256 GB (64 x 4 GB) DDR3 1066 MHz
Disk Subsystem:	1 x 300 GB SAS SFF 15K RPM
Other Hardware:	None

Software
Operating System:	Red Hat Enterprise Linux Server release 6.4 (ppc64) kernel 2.6.32-358.6.1.el6.ppc64
Compiler:	C/C++/Fortran: Version 4.7.3 of IBM Advance Toolchain 6.0-4 gcc/g++/gfortran compiler
Auto Parallel:	No
File System:	ext4
System State:	Run level 3 (multi-user)
Base Pointers:	32-bit
Peak Pointers:	32/64-bit
Other Software:	-IBM Advance Toolchain 6.0-4 -IBM Mathematical Acceleration Subsystem (MASS) libraries 7.1.0.2 -Post-Link Optimization for Linux on POWER, version 5.6.2-1

Results Table

Benchmark	Base							Peak
Benchmark	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	128	1874	928	1870	930	1872	929	32	441	986	441	987	441	985
416.gamess	128	2346	1070	2341	1070	2347	1070	128	2284	1100	2244	1120	2251	1110
433.milc	128	3739	314	3730	315	3738	314	32	902	326	903	325	901	326
434.zeusmp	128	1019	1140	1008	1160	1015	1150	128	977	1190	978	1190	977	1190
435.gromacs	128	717	1280	714	1280	715	1280	128	698	1310	701	1300	706	1290
436.cactusADM	128	901	1700	907	1690	902	1700	64	389	1970	388	1970	388	1970
437.leslie3d	128	1964	613	1966	612	1968	611	32	417	721	416	723	417	721
444.namd	128	783	1310	780	1320	782	1310	128	783	1310	780	1320	782	1310
447.dealII	128	606	2420	611	2400	601	2440	128	610	2400	599	2440	600	2440
450.soplex	128	2009	531	1545	691	1550	689	128	2009	531	1545	691	1550	689
453.povray	128	490	1390	493	1380	493	1380	128	417	1630	418	1630	418	1630
454.calculix	128	778	1360	779	1360	782	1350	128	762	1390	764	1380	762	1390
459.GemsFDTD	128	2892	470	2895	469	2896	469	64	1391	488	1390	488	1391	488
465.tonto	128	971	1300	969	1300	967	1300	128	781	1610	784	1610	782	1610
470.lbm	128	3314	531	3314	531	3310	531	32	345	1270	347	1270	345	1270
481.wrf	128	1348	1060	1330	1080	1342	1070	96	997	1080	998	1070	1001	1070
482.sphinx3	128	2244	1110	2287	1090	2256	1110	64	858	1450	861	1450	866	1440

Compiler Invocation Notes

For more information about IBM Advance Toolchain, including support, see
ftp://ftp.unicamp.br/pub/linuxpatch/toolchain/at/redhat/RHEL6/at6.0/release_notes.at6.0-6.0-4.html

Peak Tuning Notes

 Post-Link optimization tool used for:
 410.bwaves
      with options -O4 -omullX for optimization phase,
      and -imullX for instrumentation phase
 416.gamess
      with options -O4 -omullX for optimization phase,
      and -imullX for instrumentation phase
 429.mcf
      with options -O4 -omullX for optimization phase,
      and -imullX for instrumentation phase
 433.milc
      with options -O4 -omullX for optimization phase,
      and -imullX for instrumentation phase
 434.zeusmp
      with options -O4 -omullX for optimization phase,
      and -imullX for instrumentation phase
 437.leslie3d
      with options -O4 -omullX for optimization phase,
      and -imullX for instrumentation phase
 453.povray
      with options -O4 -omullX for optimization phase,
      and -imullX for instrumentation phase
 454.calculix
      with options -O4 -omullX for optimization phase,
      and -imullX for instrumentation phase
 459.GemsFDTD
      with options -O4 -omullX for optimization phase,
      and -imullX for instrumentation phase
 465.tonto
      with options -O4 -omullX for optimization phase,
      and -imullX for instrumentation phase
 470.lbm
      with options -O4 -omullX for optimization phase,
      and -imullX for instrumentation phase
 481.wrf
      with options -O4 -omullX for optimization phase,
      and -imullX for instrumentation phase

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 1048576.

  Large pages reserved as follows by root user:
  echo 8448 > /proc/sys/vm/nr_hugepages

  The Mathematical Acceleration Subsystem libraries
  are shipped with IBM XL C/C++ version 12.1 and
  IBM XL Fortran version 14.1 compiler products.

  crashkernel was set to 256 MB in /etc/yaboot.conf file.

General Notes

Environment variables set by runspec before the start of the run:
HUGETLB_ELFMAP = "RW"
HUGETLB_MORECORE = "yes"
HUGETLB_VERBOSE = "0"
TCMALLOC_MEMFS_MALLOC_PATH = "/libhugetlbfs"
XLFRTEOPTS = "intrinthds=1"

Base Compiler Invocation

Base Portability Flags

447.dealII:	-DSPEC_CPU_LINUX
481.wrf:	-DSPEC_CPU_CASE_FLAG -DSPEC_CPU_LINUX_PPC
482.sphinx3:	-fsigned-char

Base Optimization Flags

C benchmarks:

C++ benchmarks:

Fortran benchmarks:

Benchmarks using both Fortran and C:

Peak Compiler Invocation

C benchmarks:

/opt/at6.0/bin/gcc

C++ benchmarks:

/opt/at6.0/bin/g++

Fortran benchmarks:

/opt/at6.0/bin/gfortran

Benchmarks using both Fortran and C:

/opt/at6.0/bin/gcc /opt/at6.0/bin/gfortran

Peak Portability Flags

436.cactusADM:	-DSPEC_CPU_LP64
447.dealII:	-DSPEC_CPU_LINUX
459.GemsFDTD:	-DSPEC_CPU_LP64
470.lbm:	-DSPEC_CPU_LP64
481.wrf:	-DSPEC_CPU_CASE_FLAG -DSPEC_CPU_LINUX_PPC -DSPEC_CPU_LP64
482.sphinx3:	-fsigned-char

Peak Optimization Flags

C benchmarks:

433.milc:	-fprofile-generate(pass 1) -fprofile-use(pass 2) -ffast-math -O3 -mcpu=power7 -mtune=power7 -mrecip=rsqrt -fpeel-loops -funroll-loops -mpopcntd -m32 -fvect-cost-model -mveclibabi=mass -flto -fwhole-program -fuse-linker-plugin -Wl,-q -Wl,-Map=link.map,--cref -L /opt/at6.0/lib -L /opt/ibmcmp/xlmass/7.1/lib -Wl,-rpath,/opt/at6.0/lib -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib -lhugetlbfs -lmassvp7 -lmass_simdp7 -lmass
470.lbm:	-fprofile-generate(pass 1) -fprofile-use(pass 2) -ffast-math -O3 -mcpu=power7 -mtune=power7 -mrecip=rsqrt -fpeel-loops -funroll-loops -mpopcntd -m64 -mcmodel=medium -fvect-cost-model -mveclibabi=mass -flto -fwhole-program -fuse-linker-plugin -Wl,-q -Wl,-Map=link.map,--cref -L /opt/at6.0/lib64 -L /opt/ibmcmp/xlmass/7.1/lib64 -Wl,-rpath,/opt/at6.0/lib64 -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib64 -Wl,-dynamic-linker,/opt/at6.0/lib64/ld64.so.1 -lhugetlbfs -lmassvp7_64 -lmass_simdp7_64 -lmass_64
482.sphinx3:	-fprofile-generate(pass 1) -fprofile-use(pass 2) -ffast-math -O3 -mcpu=power7 -mtune=power7 -mrecip=rsqrt -fpeel-loops -funroll-loops -mpopcntd -m32 -fvect-cost-model -mveclibabi=mass -Wl,-q -Wl,-Map=link.map,--cref -L /opt/at6.0/lib -L /opt/ibmcmp/xlmass/7.1/lib -Wl,-rpath,/opt/at6.0/lib -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib -lhugetlbfs -lmassvp7 -lmass_simdp7 -lmass

C++ benchmarks:

444.namd:	basepeak = yes
447.dealII:	-ffast-math -O3 -mcpu=power7 -mtune=power7 -mrecip=rsqrt -fpeel-loops -funroll-loops -mpopcntd -m32 -fvect-cost-model -mveclibabi=mass -flto -fwhole-program -fuse-linker-plugin --param max-inline-insns-auto=200 -fno-associative-math -Wl,-q -Wl,-Map=link.map,--cref -L /opt/at6.0/lib -L /opt/ibmcmp/xlmass/7.1/lib -Wl,-rpath,/opt/at6.0/lib -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib -lhugetlbfs -lmassvp7 -lmass_simdp7 -lmass -ltcmalloc -lstdc++ -lpthread
450.soplex:	basepeak = yes
453.povray:	-fprofile-generate(pass 1) -fprofile-use(pass 2) -ffast-math -O3 -mcpu=power7 -mtune=power7 -mrecip=rsqrt -fpeel-loops -funroll-loops -mpopcntd -m32 -fvect-cost-model -mveclibabi=mass -flto -fwhole-program -fuse-linker-plugin --param max-inline-insns-auto=200 -fno-associative-math -Wl,-q -Wl,-Map=link.map,--cref -L /opt/at6.0/lib -L /opt/ibmcmp/xlmass/7.1/lib -Wl,-rpath,/opt/at6.0/lib -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib -lhugetlbfs -lmassvp7 -lmass_simdp7 -lmass -ltcmalloc -lstdc++ -lpthread

Fortran benchmarks:

410.bwaves:	-fprofile-generate(pass 1) -fprofile-use(pass 2) -ffast-math -O3 -mcpu=power7 -mtune=power7 -mrecip=rsqrt -fpeel-loops -funroll-loops -mpopcntd -m64 -mcmodel=medium -fvect-cost-model -mveclibabi=mass -flto -fwhole-program -fuse-linker-plugin -Wl,-q -Wl,-Map=link.map,--cref -L /opt/at6.0/lib64 -L /opt/ibmcmp/xlmass/7.1/lib64 -Wl,-rpath,/opt/at6.0/lib64 -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib64 -Wl,-dynamic-linker,/opt/at6.0/lib64/ld64.so.1 -lhugetlbfs -lmassvp7_64 -lmass_simdp7_64 -lmass_64
416.gamess:	-fprofile-generate(pass 1) -fprofile-use(pass 2) -ffast-math -O3 -mcpu=power7 -mtune=power7 -mrecip=rsqrt -fpeel-loops -funroll-loops -mpopcntd -m32 -fvect-cost-model -mveclibabi=mass -Wl,-q -Wl,-Map=link.map,--cref -L /opt/at6.0/lib -L /opt/ibmcmp/xlmass/7.1/lib -Wl,-rpath,/opt/at6.0/lib -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib -lhugetlbfs -lmassvp7 -lmass_simdp7 -lmass -ltcmalloc -lstdc++ -lpthread
434.zeusmp:	-fprofile-generate(pass 1) -fprofile-use(pass 2) -ffast-math -O3 -mcpu=power7 -mtune=power7 -mrecip=rsqrt -fpeel-loops -funroll-loops -mpopcntd -m32 -fvect-cost-model -mveclibabi=mass -flto -fwhole-program -fuse-linker-plugin -Wl,-q -Wl,-Map=link.map,--cref -L /opt/at6.0/lib -L /opt/ibmcmp/xlmass/7.1/lib -Wl,-rpath,/opt/at6.0/lib -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib -lhugetlbfs -lmassvp7 -lmass_simdp7 -lmass -ltcmalloc -lstdc++ -lpthread
437.leslie3d:	-fprofile-generate(pass 1) -fprofile-use(pass 2) -ffast-math -O3 -mcpu=power7 -mtune=power7 -mrecip=rsqrt -fpeel-loops -funroll-loops -mpopcntd -m32 -fvect-cost-model -mveclibabi=mass -flto -fwhole-program -fuse-linker-plugin -Wl,-q -Wl,-Map=link.map,--cref -L /opt/at6.0/lib -L /opt/ibmcmp/xlmass/7.1/lib -Wl,-rpath,/opt/at6.0/lib -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib -lhugetlbfs -lmassvp7 -lmass_simdp7 -lmass
459.GemsFDTD:	-fprofile-generate(pass 1) -fprofile-use(pass 2) -ffast-math -O3 -mcpu=power7 -mtune=power7 -mrecip=rsqrt -fpeel-loops -funroll-loops -mpopcntd -m64 -mcmodel=medium -fvect-cost-model -mveclibabi=mass -Wl,-q -Wl,-Map=link.map,--cref -L /opt/at6.0/lib64 -L /opt/ibmcmp/xlmass/7.1/lib64 -Wl,-rpath,/opt/at6.0/lib64 -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib64 -Wl,-dynamic-linker,/opt/at6.0/lib64/ld64.so.1 -lhugetlbfs -lmassvp7_64 -lmass_simdp7_64 -lmass_64
465.tonto:	Same as 416.gamess

Benchmarks using both Fortran and C:

435.gromacs:	-fprofile-generate(pass 1) -fprofile-use(pass 2) -ffast-math -O3 -mcpu=power7 -mtune=power7 -mrecip=rsqrt -fpeel-loops -funroll-loops -mpopcntd -m32 -fvect-cost-model -mveclibabi=mass -Wl,-q -Wl,-Map=link.map,--cref -L /opt/at6.0/lib -L /opt/ibmcmp/xlmass/7.1/lib -Wl,-rpath,/opt/at6.0/lib -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib -lhugetlbfs -lmassvp7 -lmass_simdp7 -lmass
436.cactusADM:	-fprofile-generate(pass 1) -fprofile-use(pass 2) -ffast-math -O3 -mcpu=power7 -mtune=power7 -mrecip=rsqrt -fpeel-loops -funroll-loops -mpopcntd -m64 -mcmodel=medium -fvect-cost-model -mveclibabi=mass -Wl,-q -Wl,-Map=link.map,--cref -L /opt/at6.0/lib64 -L /opt/ibmcmp/xlmass/7.1/lib64 -Wl,-rpath,/opt/at6.0/lib64 -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib64 -Wl,-dynamic-linker,/opt/at6.0/lib64/ld64.so.1 -lhugetlbfs -lmassvp7_64 -lmass_simdp7_64 -lmass_64
454.calculix:	-ffast-math -O3 -mcpu=power7 -mtune=power7 -mrecip=rsqrt -fpeel-loops -funroll-loops -mpopcntd -m32 -fvect-cost-model -mveclibabi=mass -flto -fwhole-program -fuse-linker-plugin -Wl,-q -Wl,-Map=link.map,--cref -L /opt/at6.0/lib -L /opt/ibmcmp/xlmass/7.1/lib -Wl,-rpath,/opt/at6.0/lib -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib -lhugetlbfs -lmassvp7 -lmass_simdp7 -lmass -ltcmalloc -lstdc++ -lpthread
481.wrf:	-ffast-math -O3 -mcpu=power7 -mtune=power7 -mrecip=rsqrt -fpeel-loops -funroll-loops -mpopcntd -m64 -mcmodel=medium -fvect-cost-model -mveclibabi=mass -Wl,-q -Wl,-Map=link.map,--cref -L /opt/at6.0/lib64 -L /opt/ibmcmp/xlmass/7.1/lib64 -Wl,-rpath,/opt/at6.0/lib64 -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib64 -Wl,-dynamic-linker,/opt/at6.0/lib64/ld64.so.1 -lhugetlbfs -lmassvp7_64 -lmass_simdp7_64 -lmass_64

The flags file that was used to format this result can be browsed at
http://www.spec.org/cpu2006/flags/IBM-Linux-AT.20130813.html.

You can also download the XML flags source by saving the following link:
http://www.spec.org/cpu2006/flags/IBM-Linux-AT.20130813.xml.