SPEC® CFP2006 Result

Copyright 2006-2017 Standard Performance Evaluation Corporation

IBM Corporation

IBM Power S822LC (2.92 GHz, 20 core, Ubuntu)

CPU2006 license: 11 Test date: Sep-2015
Test sponsor: IBM Corporation Hardware Availability: Oct-2015
Tested by: IBM Corporation Software Availability: Dec-2015
Benchmark results graph
Hardware
CPU Name: POWER8
CPU Characteristics: Intelligent Energy Optimization enabled,
up to 3.49 GHz
CPU MHz: 2926
FPU: Integrated
CPU(s) enabled: 20 cores, 2 chips, 10 cores/chip, 8 threads/core
CPU(s) orderable: 2 Modules
Primary Cache: 32 KB I + 64 KB D on chip per core
Secondary Cache: 512 KB I+D on chip per core
L3 Cache: 8 MB I+D on chip per core
Other Cache: 16 MB I+D off chip per 4 DIMMs
Memory: 256 GB (32 x 8 GB DIMMs) DDR3 1333 MHz
Disk Subsystem: 2 x 500 GB 15K RPM SAS SFF-2 Raid5
Other Hardware: None
Software
Operating System: Ubuntu 14.04 LTS Updated to 14.04.3
(ppc64le) kernel <3.16.0-46-generic>
Compiler: C/C++: Version 13.1.3 of IBM XL C/C++ for Linux;
Fortran: Version 15.1.3 of IBM XL Fortran for
Linux
Auto Parallel: No
File System: ext4
System State: Run level 3 (multi-user)
Base Pointers: 64-bit
Peak Pointers: 64-bit
Other Software: Post-Link Optimization for Linux on
POWER, version 5.6.2-6f
IBM Advance Toolchain 8.0-3

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 80 1408 772 1414 769 1413 769 20 312 871 313 869 313 869
416.gamess 80 2426 646 2427 645 2431 644 80 2426 646 2427 645 2431 644
433.milc 80 1178 623 1179 623 1179 623 20 241 763 241 763 240 763
434.zeusmp 80 987 738 1009 722 988 737 80 919 792 913 797 920 792
435.gromacs 80 851 671 861 663 860 664 80 686 833 696 820 698 818
436.cactusADM 80 949 1010 956 1000 952 1000 40 326 1470 326 1470 325 1470
437.leslie3d 80 1300 579 1303 577 1301 578 20 227 827 229 821 229 820
444.namd 80 790 812 769 834 769 834 80 737 870 716 896 716 896
447.dealII 80 640 1430 638 1430 638 1430 140 995 1610 993 1610 990 1620
450.soplex 80 1642 406 1007 662 1008 662 60 1159 432 746 671 749 668
453.povray 80 493 864 496 858 494 862 120 570 1120 570 1120 570 1120
454.calculix 80 1057 624 1058 624 1058 624 80 1056 625 1053 627 1054 626
459.GemsFDTD 80 1831 463 1829 464 1829 464 40 879 483 879 483 880 482
465.tonto 80 856 920 855 920 857 919 80 777 1010 778 1010 777 1010
470.lbm 80 2106 522 2104 522 2106 522 20 216 1270 216 1270 216 1270
481.wrf 80 945 945 933 958 933 958 80 917 975 904 989 901 992
482.sphinx3 80 1932 807 1876 831 1875 832 40 858 908 825 945 822 948

Peak Tuning Notes

 410.bwaves fdpr options: -O4 -m power8 -A 2 -rcl 2 -sls -dir -vrox
 416.gamess fdpr options: -O4 -m power8 -A 2 -rcl 2 -sls -dir -vrox
 434.zeusmp fdpr options: -O4 -m power8 -A 2 -rcl 2 -sls -dir -vrox
 435.gromacs fdpr options: -O4 -m power8 -A 2 -rcl 2 -sls -dir -vrox
 436.cactusADM fdpr options: -O4 -m power8 -A 2 -sls -dir -vrox
 437.leslie3d fdpr options: -O4 -m power8 -A 2 -rcl 2 -sls -dir -vrox
 444.namd fdpr options: -O4 -m power8 -A 2 -rcl 2 -sls -dir -vrox
 447.dealII fdpr options: -O4 -m power8 -A 2 -rcl 2 -sls -dir -vrox
 454.calculix fdpr options: -O4 -m power8 -A 2 -rcl 2 -sls -dir -vrox
 459.GemsFDTD fdpr options: -O4 -m power8 -A 2 -sls -dir -vrox
 465.tonto fdpr options: -O4 -m power8 -A 2 -sls -dir -vrox
 470.lbm fdpr options: -O4 -m power8 -A 2 -rcl 2 -sls -dir -vrox
 481.wrf fdpr options: -O4 -m power8 -A 2 -rcl 2 -sls -dir -vrox
 482.sphinx3 fdpr options: -O4 -m power8 -A 2 -rcl 2 -sls -dir -vrox

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 unlimited

 8000 16M large pages defined
 Transparent huge page disabled with
 echo never > /sys/kernel/mm/transparent_hugepage/enabled
 sysctl vm.nr_hugepages=N and reboot to set large page pool

General Notes

Environment variables set by runspec before the start of the run:
HUGETLB_MORECORE = "yes"
HUGETLB_VERBOSE = "0"
LD_LIBRARY_PATH = "$LD_LIBRARY_PATH:/opt/ibm/fdprpro/lib"
TCMALLOC_MEMFS_MALLOC_PATH = "/dev/hugepages/"
XLFRTEOPTS = "intrinthds=1"

This result uses the post_setup and/or bench_post_setup to drop caches. SPEC has
determined that although the effect may have been negligible for this run, future
submissions will not be considered rule compliant if the post_setup actions drop
caches (e.g. : "echo 3 > /proc/sys/vm/drop_caches").

Base Compiler Invocation

C benchmarks:

 /opt/ibm/xlC/13.1.2/bin/xlc_at   -qlanglvl=extc99 

C++ benchmarks:

 /opt/ibm/xlC/13.1.2/bin/xlC_at 

Fortran benchmarks:

 /opt/ibm/xlf/15.1.2/bin/xlf95_at 

Benchmarks using both Fortran and C:

 /opt/ibm/xlC/13.1.2/bin/xlc_at   -qlanglvl=extc99   /opt/ibm/xlf/15.1.2/bin/xlf95_at 

Base Portability Flags

410.bwaves:  -qfixed 
416.gamess:  -qfixed 
433.milc:  -DSPEC_CPU_LP64 
434.zeusmp:  -qfixed 
435.gromacs:  -DSPEC_CPU_LP64   -qfixed   -qextname 
436.cactusADM:  -DSPEC_CPU_LP64   -qfixed   -qextname 
437.leslie3d:  -qfixed 
444.namd:  -DSPEC_CPU_LP64 
447.dealII:  -DSPEC_CPU_LP64 
450.soplex:  -DSPEC_CPU_LP64 
453.povray:  -DSPEC_CPU_LP64 
454.calculix:  -DSPEC_CPU_LP64   -qfixed   -qextname 
470.lbm:  -DSPEC_CPU_LP64 
481.wrf:  -DNOUNDERSCORE   -DSPEC_CPU_LINUX   -DSPEC_CPU_WORDS_LITTLEENDIAN 
482.sphinx3:  -DSPEC_CPU_LP64   -qchars=signed 

Base Optimization Flags

C benchmarks:

 -qinline=40   -qipa=threads   -O5   -qsimd=noauto   -q64   -lhugetlbfs 

C++ benchmarks:

 -qinline=40   -qipa=threads   -O5   -q64   -qrtti   -qnoxlcompatmacros   -D__extern_always_inline=inline   -lhugetlbfs 

Fortran benchmarks:

 -qipa=threads   -O5   -q64   -qalias=nostd   -lhugetlbfs 

Benchmarks using both Fortran and C:

 -qinline=40   -qipa=threads   -O5   -qsimd=noauto   -q64   -qalias=nostd   -lhugetlbfs 

Base Other Flags

C benchmarks:

 -qipa=noobject   -qsuppress=1500-036 

C++ benchmarks:

 -qipa=noobject   -qsuppress=1500-036 

Fortran benchmarks:

 -qipa=noobject   -qsuppress=1500-010   -qsuppress=cmpmsg   -qsuppress=1500-036 

Benchmarks using both Fortran and C:

 -qipa=noobject   -qsuppress=1500-010   -qsuppress=cmpmsg   -qsuppress=1500-036 

Peak Compiler Invocation

C benchmarks:

 /opt/ibm/xlC/13.1.2/bin/xlc_at   -qlanglvl=extc99 

C++ benchmarks:

 /opt/ibm/xlC/13.1.2/bin/xlC_at 

Fortran benchmarks:

 /opt/ibm/xlf/15.1.2/bin/xlf95_at 

Benchmarks using both Fortran and C:

 /opt/ibm/xlC/13.1.2/bin/xlc_at   -qlanglvl=extc99   /opt/ibm/xlf/15.1.2/bin/xlf95_at 

Peak Portability Flags

410.bwaves:  -qfixed 
416.gamess:  -qfixed 
433.milc:  -DSPEC_CPU_LP64 
434.zeusmp:  -qfixed 
435.gromacs:  -DSPEC_CPU_LP64   -qfixed   -qextname 
436.cactusADM:  -DSPEC_CPU_LP64   -qfixed   -qextname 
437.leslie3d:  -qfixed 
444.namd:  -DSPEC_CPU_LP64 
447.dealII:  -DSPEC_CPU_LP64 
450.soplex:  -DSPEC_CPU_LP64 
453.povray:  -DSPEC_CPU_LP64 
454.calculix:  -DSPEC_CPU_LP64   -qfixed   -qextname 
470.lbm:  -DSPEC_CPU_LP64 
481.wrf:  -DNOUNDERSCORE   -DSPEC_CPU_LINUX   -DSPEC_CPU_WORDS_LITTLEENDIAN 
482.sphinx3:  -DSPEC_CPU_LP64   -qchars=signed 

Peak Optimization Flags

C benchmarks:

433.milc:  -qinline=40   -qipa=threads   -O5   -q64   -qprefetch=dscr=0x93   -lhugetlbfs 
470.lbm:  -qinline=40   -qipa=threads   -qpdf1(pass 1)   -qpdf2(pass 2)   -O5   -qsimd=noauto   -q64   -qfdpr   -lhugetlbfs   -Wl,-q 
482.sphinx3:  -qinline=40   -O5   -qsimd=noauto   -q64   -qfdpr   -lhugetlbfs   -Wl,-q 

C++ benchmarks:

444.namd:  -qinline=40   -qipa=threads   -O4   -q64   -qfdpr   -D__extern_always_inline=inline   -lhugetlbfs   -Wl,-q 
447.dealII:  -qinline=40   -qipa=threads   -qpdf1(pass 1)   -qpdf2(pass 2)   -O4   -q64   -qfdpr   -qrtti   -qnoxlcompatmacros   -D__extern_always_inline=inline   -lhugetlbfs   -Wl,-q 
450.soplex:  -qinline=40   -qipa=threads   -qpdf1(pass 1)   -qpdf2(pass 2)   -O3   -qarch=auto   -qtune=auto   -qsimd   -q64   -qnoprefetch   -D__extern_always_inline=inline   -lhugetlbfs 
453.povray:  -qinline=40   -qipa=threads   -qpdf1(pass 1)   -qpdf2(pass 2)   -O3   -qarch=auto   -qtune=auto   -qprefetch=dscr=0x93   -D__extern_always_inline=inline   -lhugetlbfs   -Wl,-q 

Fortran benchmarks:

410.bwaves:  -qipa=threads   -O5   -qsimd=noauto   -q64   -qfdpr   -qsmallstack=dynlenonheap   -lhugetlbfs   -Wl,-q 
416.gamess:  basepeak = yes 
434.zeusmp:  -qipa=threads   -O4   -qsimd=noauto   -q64   -qfdpr   -qxlf90=nosignedzero   -lhugetlbfs   -Wl,-q 
437.leslie3d:  -qipa=threads   -O5   -q64   -qassert=contig   -qprefetch=dscr=0x1D7   -qhot=novector   -qfdpr   -lhugetlbfs   -Wl,-q   -B/opt/at8.0/share/libhugetlbfs/   -tl   -Wl,--hugetlbfs-align 
459.GemsFDTD:  -qipa=threads   -qpdf1(pass 1)   -qpdf2(pass 2)   -O5   -q64   -qipa=partition=large   -qfdpr   -lhugetlbfs   -Wl,-q 
465.tonto:  -qipa=threads   -qpdf1(pass 1)   -qpdf2(pass 2)   -O5   -q64   -qipa=partition=large   -qalias=nostd   -qfdpr   -lhugetlbfs   -Wl,-q 

Benchmarks using both Fortran and C:

435.gromacs:  -qinline=40   -qipa=threads   -qpdf1(pass 1)   -qpdf2(pass 2)   -O4   -q64   -qipa=partition=large   -qfdpr   -lhugetlbfs   -Wl,-q 
436.cactusADM:  -qinline=40   -qipa=threads   -qpdf1(pass 1)   -qpdf2(pass 2)   -O4   -qipa=partition=large   -q64   -qfdpr   -lhugetlbfs   -Wl,-q 
454.calculix:  -qinline=40   -qipa=threads   -O3   -qarch=auto   -qtune=auto   -q64   -qhot   -qfdpr   -lhugetlbfs   -Wl,-q 
481.wrf:  -qinline=40   -O4   -q64   -qfdpr   -lhugetlbfs   -Wl,-q 

Peak Other Flags

C benchmarks:

433.milc:  -qipa=noobject   -qsuppress=1500-036 
470.lbm:  -qsuppress=1586-476(pass 2)   -qipa=noobject   -qsuppress=1500-036 
482.sphinx3:  -qsuppress=1500-036 

C++ benchmarks (except as noted below):

 -qsuppress=1586-476(pass 2)   -qipa=noobject   -qsuppress=1500-036 
444.namd:  -qipa=noobject   -qsuppress=1500-036 

Fortran benchmarks (except as noted below):

 -qipa=noobject   -qsuppress=1500-010   -qsuppress=cmpmsg   -qsuppress=1500-036 
459.GemsFDTD:  -qsuppress=1586-476(pass 2)   -qipa=noobject   -qsuppress=1500-010   -qsuppress=cmpmsg   -qsuppress=1500-036 
465.tonto:  -qsuppress=1586-476(pass 2)   -qipa=noobject   -qsuppress=1500-010   -qsuppress=cmpmsg   -qsuppress=1500-036 

Benchmarks using both Fortran and C (except as noted below):

 -qsuppress=1586-476(pass 2)   -qipa=noobject   -qsuppress=1500-010   -qsuppress=cmpmsg   -qsuppress=1500-036 
454.calculix:  -qsuppress=1500-010   -qsuppress=cmpmsg   -qsuppress=1500-036 
481.wrf:  -qipa=noobject   -qsuppress=1500-010   -qsuppress=cmpmsg   -qsuppress=1500-036 

The flags files that were used to format this result can be browsed at
http://www.spec.org/cpu2006/flags/IBM-XL.V13La.20151020.html,
http://www.spec.org/cpu2006/flags/IBM-Linux-V7.html.

You can also download the XML flags sources by saving the following links:
http://www.spec.org/cpu2006/flags/IBM-XL.V13La.20151020.xml,
http://www.spec.org/cpu2006/flags/IBM-Linux-V7.xml.