SPEC® CFP2006 Result

Copyright 2006-2014 Standard Performance Evaluation Corporation

IBM Corporation

IBM Power 780 (3.7 GHz, 128 core, RHEL)

CPU2006 license: 11 Test date: Sep-2012
Test sponsor: IBM Corporation Hardware Availability: Oct-2012
Tested by: IBM Corporation Software Availability: Dec-2012
Benchmark results graph
Hardware
CPU Name: POWER7+
CPU Characteristics: Intelligent Energy Optimization enabled,
up to 4.144 GHz
CPU MHz: 3724
FPU: Integrated
CPU(s) enabled: 128 cores, 16 chips, 8 cores/chip, 4 threads/core
CPU(s) orderable: 32,64,96,128 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: 1 TB (64 x 16 GB) DDR3 1066 MHz
Disk Subsystem: 12x146.8 GB SAS SFF 15K RPM
Other Hardware: None
Software
Operating System: Red Hat Enterprise Linux Server release 6.3
(ppc64) kernel 2.6.32-279.el6.ppc64
Compiler: C/C++: Version 12.1 of IBM XL C/C++ for Linux;
Fortran: Version 14.1 of IBM XL Fortran for Linux
Auto Parallel: No
File System: ext4
System State: Run level 3 (multi-user)
Base Pointers: 32-bit
Peak Pointers: 32/64-bit
Other Software: -Post-Link Optimization for Linux on
POWER, version 5.6.1-7
-MicroQuill SmartHeap 9
-Apache C++ Standard Library V4.2.1

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.
410.bwaves 512 2420 2880 2421 2870 2423 2870 128 547 3180 545 3190 545 3190
416.gamess 512 2331 4300 2340 4290 2332 4300 512 2313 4330 2304 4350 2271 4410
433.milc 512 1639 2870 1632 2880 1624 2890 128 390 3010 388 3030 388 3030
434.zeusmp 512 987 4720 986 4730 987 4720 512 987 4720 986 4730 987 4720
435.gromacs 512 1001 3650 980 3730 977 3740 512 960 3810 939 3890 942 3880
436.cactusADM 512 1383 4430 1379 4440 1372 4460 256 467 6560 471 6500 463 6600
437.leslie3d 512 2439 1970 2439 1970 2438 1970 128 505 2380 505 2380 505 2380
444.namd 512 782 5250 734 5590 724 5670 512 767 5350 710 5780 710 5780
447.dealII 512 761 7690 755 7760 757 7740 512 643 9110 648 9030 626 9360
450.soplex 512 3079 1390 1853 2300 1795 2380 256 1836 1160 763 2800 764 2800
453.povray 512 617 4410 612 4450 612 4450 512 448 6080 443 6150 446 6110
454.calculix 512 898 4710 898 4700 903 4680 512 880 4800 882 4790 883 4790
459.GemsFDTD 512 3617 1500 3616 1500 3606 1510 128 865 1570 863 1570 867 1570
465.tonto 512 1093 4610 1097 4590 1097 4590 512 1046 4820 1045 4820 1043 4830
470.lbm 512 1624 4330 1620 4340 1621 4340 512 1619 4350 1624 4330 1620 4340
481.wrf 512 1695 3370 1589 3600 1592 3590 512 1707 3350 1589 3600 1591 3590
482.sphinx3 512 2963 3370 2883 3460 2875 3470 128 467 5340 424 5880 417 5990

Compiler Invocation Notes

 C/C++ compiler updated to December 2012 PTF
 Version: 12.01.0000.0002
 Fortran compiler updated to December 2012 PTF
 Version: 14.01.0000.0002

Peak Tuning Notes

 Post-Link optimization tool used for:
 433.milc 435.gromacs 450.soplex 482.sphinx3
      with options -O4 -nodp
 434.zeusmp
      with options -O4 -vrox -nodp
  437.leslie3d
       with options -O3 -lu -1 -nodp -sdp 9
  444.namd
       with options -O3 -lu -1 -nodp -sdp 9
  450.soplex
      with options -O4 -nodp
 465.tonto
    with options -O4
 482.sphinx3
     with options -O4 -nodp

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.
 ulimit -s (open files) set to 500000.
 Filesystem cache changed as follows by root user:
 echo 150 > /proc/sys/vm/vfs_cache_pressure
 Large pages reserved as follows by root user:
 echo 6000 > /proc/sys/vm/nr_hugepages
 echo 30000 > /proc/sys/vm/nr_overcommit_hugepages
 The Apache C++ Standard Library V4.2.1 was installed from
 http://stdcxx.apache.org/download.html using:
     gmake BUILDTYPE=8d CONFIG=gcc.config

Platform Notes

 Service processor memory mirroring property disabled.
 Hardware Page Table (HPT) ratio changed as follows by root user on Hardware Management Console (HMC):
 chsyscfg -m <CEC NAME> -r prof -i "name=<PROFILE NAME>,lpar_name=<PARTITION NAME>,hpt_ratio=1:512"

General Notes

 The following environment varibles were set before the runspec command:
   export HUGETLB_VERBOSE=0
   export HUGETLB_MORECORE=yes
   export XLFRTEOPTS=intrinthds=1
   export HUGETLB_ELFMAP=RW

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   -qarch=pwr7   -qtune=pwr7   -q32   -qipa=threads   -B/usr/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-align 

C++ benchmarks:

 -O5   -qarch=pwr7   -qtune=pwr7   -q32   -qipa=threads   -qrtti   -B/usr/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-align 

Fortran benchmarks:

 -O5   -qarch=pwr7   -qtune=pwr7   -q32   -qipa=threads   -qalias=nostd   -B/usr/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-align 

Benchmarks using both Fortran and C:

 -O5   -qarch=pwr7   -qtune=pwr7   -q32   -qipa=threads   -B/usr/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-align   -qalias=nostd 

Base Other Flags

C benchmarks:

C++ benchmarks:

Fortran benchmarks:

Benchmarks using both Fortran and C:

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:  -DSPEC_CPU_LP64   -qfixed   -qextname 
437.leslie3d:  -qfixed 
453.povray:  -DSPEC_CPU_LP64 
454.calculix:  -qfixed   -qextname 
481.wrf:  -DNOUNDERSCORE 
482.sphinx3:  -qchars=signed 

Peak Optimization Flags

C benchmarks:

433.milc:  -Wl,-q   -O5   -qarch=pwr7   -qtune=pwr7   -qipa=threads   -lhugetlbfs 
470.lbm:  -qpdf1(pass 1)   -qpdf2(pass 2)   -O5   -qarch=pwr7   -qtune=pwr7   -qipa=threads   -B/usr/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-align 
482.sphinx3:  -Wl,-q   -qpdf1(pass 1)   -qpdf2(pass 2)   -O4   -qarch=pwr7   -qtune=pwr7   -qipa=threads   -lhugetlbfs 

C++ benchmarks:

444.namd:  -Wl,-q   -qpdf1(pass 1)   -qpdf2(pass 2)   -O5   -qarch=pwr7   -qtune=pwr7   -qipa=threads   -lhugetlbfs 
447.dealII:  -O4   -qarch=pwr7   -qtune=pwr7   -qipa=threads   -qrtti   -qcpp_stdinc=/autobench/sources/speccpu2006/stdcxx421/include/ansi:/autobench/sources/speccpu2006/stdcxx421/include:/opt/ibmcmp/vacpp/12.1/include   -lsmartheap   -L/autobench/sources/speccpu2006/stdcxx421/lib   -R/autobench/sources/speccpu2006/stdcxx421/lib   -lstd8d 
450.soplex:  -Wl,-q   -qpdf1(pass 1)   -qpdf2(pass 2)   -O3   -qarch=pwr7   -qtune=pwr7   -q64   -lhugetlbfs 
453.povray:  -Wl,-q   -qpdf1(pass 1)   -qpdf2(pass 2)   -O4   -qarch=pwr7   -qtune=pwr7   -qipa=threads   -qsimd   -q64   -lsmartheap64 

Fortran benchmarks:

410.bwaves:  -qpdf1(pass 1)   -qpdf2(pass 2)   -O4   -qarch=pwr7   -qtune=pwr7   -qipa=threads   -qsmallstack=dynlenonheap   -q64   -lhugetlbfs 
416.gamess:  -qpdf1(pass 1)   -qpdf2(pass 2)   -O5   -qarch=pwr7   -qtune=pwr7   -qipa=threads   -qalias=nostd   -lhugetlbfs 
434.zeusmp:  basepeak = yes 
437.leslie3d:  -Wl,-q   -O5   -qarch=pwr7   -qtune=pwr7   -qipa=threads   -q64   -B/usr/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-align 
459.GemsFDTD:  -O4   -qarch=pwr7   -qtune=pwr7   -qipa=threads   -qsimd   -B/usr/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-align 
465.tonto:  -Wl,-q   -qpdf1(pass 1)   -qpdf2(pass 2)   -O5   -qarch=pwr7   -qtune=pwr7   -qipa=threads   -qsimd   -lhugetlbfs 

Benchmarks using both Fortran and C:

435.gromacs:  -Wl,-q   -qpdf1(pass 1)   -qpdf2(pass 2)   -O4   -qarch=pwr7   -qtune=pwr7   -qipa=threads   -qsimd   -lhugetlbfs 
436.cactusADM:  -O4   -qarch=pwr7   -qtune=pwr7   -qipa=threads   -qsimd   -qnostrict   -q64   -lhugetlbfs 
454.calculix:  -qpdf1(pass 1)   -qpdf2(pass 2)   -O5   -qarch=pwr7   -qtune=pwr7   -qipa=threads   -B/usr/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-align 
481.wrf:  -O3   -qarch=pwr7   -qtune=pwr7   -q64   -lhugetlbfs 

Peak Other Flags

C benchmarks:

C++ benchmarks:

Fortran benchmarks:

Benchmarks using both Fortran and C:


The flags files that were used to format this result can be browsed at
http://www.spec.org/cpu2006/flags/IBM-Linux-XL.20121024.html,
http://www.spec.org/cpu2006/flags/IBM-Power.html.

You can also download the XML flags sources by saving the following links:
http://www.spec.org/cpu2006/flags/IBM-Linux-XL.20121024.xml,
http://www.spec.org/cpu2006/flags/IBM-Power.xml.