SPEC® CFP2006 Result

Copyright 2006-2014 Standard Performance Evaluation Corporation

Supermicro

Supermicro Processor Blade SBA-7142G-T4 (BHQGE,
Opteron 6338P)
AMD Opteron 6338P

CPU2006 license: 001176 Test date: Oct-2013
Test sponsor: Supermicro Hardware Availability: Jan-2014
Tested by: Supermicro Software Availability: Aug-2012
Benchmark results graph
Hardware
CPU Name: AMD Opteron 6338P
CPU Characteristics: AMD Turbo CORE technology up to 2.80 GHz
CPU MHz: 2300
FPU: Integrated
CPU(s) enabled: 48 cores, 4 chips, 12 cores/chip
CPU(s) orderable: 2,4 chips
Primary Cache: 384 KB I on chip per chip,
64 KB I shared / 2 cores;
16 KB D on chip per core
Secondary Cache: 12 MB I+D on chip per chip, 2 MB shared / 2 cores
L3 Cache: 16 MB I+D on chip per chip, 8 MB shared / 6 cores
Other Cache: None
Memory: 256 GB (32 x 8 GB 2Rx4 PC3-12800R-11, ECC)
Disk Subsystem: 1 x 500 GB SATA II, 7200 RPM
Other Hardware: None
Software
Operating System: Red Hat Enterprise Linux Server release 6.4,
Kernel 2.6.32-358.el6.x86_64
Compiler: C/C++/Fortran: Version 4.5.2 of x86 Open64
Compiler Suite (from AMD)
Auto Parallel: No
File System: ext4
System State: Run level 3 (Full multiuser with network)
Base Pointers: 64-bit
Peak Pointers: 32/64-bit
Other Software: None

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 48 1146 569 1120 582 1146 569 48 1105 590 1124 580 1106 590
416.gamess 48 2023 465 1992 472 2007 468 48 1826 515 1832 513 1824 515
433.milc 48 864 510 864 510 867 508 48 748 589 747 590 749 589
434.zeusmp 48 692 631 700 624 700 624 48 667 655 668 654 672 650
435.gromacs 48 594 577 595 576 595 576 48 481 712 481 713 482 711
436.cactusADM 48 750 765 753 761 754 761 48 680 844 665 863 664 863
437.leslie3d 48 1090 414 1089 414 1088 415 48 921 490 932 484 929 486
444.namd 48 838 459 837 460 836 460 48 690 558 692 556 695 554
447.dealII 48 515 1070 507 1080 520 1060 48 488 1130 482 1140 482 1140
450.soplex 48 874 458 831 482 832 481 48 783 511 773 518 773 518
453.povray 48 409 625 407 628 407 628 48 354 721 353 722 354 722
454.calculix 48 435 910 427 928 428 924 48 408 970 410 967 413 959
459.GemsFDTD 48 1306 390 1306 390 1307 390 48 1171 435 1168 436 1170 435
465.tonto 48 928 509 841 561 890 531 48 768 615 766 617 780 605
470.lbm 48 858 768 847 779 851 775 48 854 773 850 776 851 775
481.wrf 48 799 671 817 657 794 675 48 792 677 778 689 781 686
482.sphinx3 48 1599 585 1595 587 1591 588 48 1312 713 1312 713 1316 711

Submit Notes

The config file option 'submit' was used.
'numactl' was used to bind copies to the cores.
See the configuration file for details.

Operating System Notes

'ulimit -s unlimited' was used to set environment stack size
'ulimit -l 2097152'  was used to set environment locked pages in memory limit

Set transparent_hugepage=never as a boot parameter in /boot/grub/menu.lst

Set vm/nr_hugepages=43008 in /etc/sysctl.conf
mount -t hugetlbfs nodev /mnt/hugepages

Platform Notes

C State mode set to disabled under Processor & Clock Options in BIOS.
HPC Mode set to enabled under Processor & Clock Options in BIOS.
C1E Support set to disabled under Processor & Clock Options in BIOS.
Power Down Enable set to disabled in Memory Configuration in BIOS.

General Notes

Environment variables set by runspec before the start of the run:
HUGETLB_LIMIT = "896"
LD_LIBRARY_PATH = "/home/spec/amd1206-rate-libs-revA/32:/home/spec/amd1206-rate-libs-revA/64"

The x86 Open64 Compiler Suite is only available from (and supported by) AMD at
http://developer.amd.com/cpu/open64

Binaries were compiled on a system with 2x AMD Opteron 6386SE chips + 128GB Memory using RHEL 6.3

Base Compiler Invocation

C benchmarks:

 opencc 

C++ benchmarks:

 openCC 

Fortran benchmarks:

 openf95 

Benchmarks using both Fortran and C:

 opencc   openf95 

Base Portability Flags

410.bwaves:  -DSPEC_CPU_LP64 
416.gamess:  -DSPEC_CPU_LP64 
433.milc:  -DSPEC_CPU_LP64 
434.zeusmp:  -DSPEC_CPU_LP64 
435.gromacs:  -DSPEC_CPU_LP64 
436.cactusADM:  -DSPEC_CPU_LP64   -fno-second-underscore 
437.leslie3d:  -DSPEC_CPU_LP64 
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 
459.GemsFDTD:  -DSPEC_CPU_LP64 
465.tonto:  -DSPEC_CPU_LP64 
470.lbm:  -DSPEC_CPU_LP64 
481.wrf:  -DSPEC_CPU_LINUX   -DSPEC_CPU_CASE_FLAG   -DSPEC_CPU_LP64   -fno-second-underscore 
482.sphinx3:  -DSPEC_CPU_LP64 

Base Optimization Flags

C benchmarks:

 -Ofast   -OPT:malloc_alg=1   -HP:bd=2m:heap=2m   -IPA:plimit=8000   -IPA:small_pu=100   -mso   -march=bdver1 

C++ benchmarks:

 -Ofast   -static   -CG:load_exe=0   -OPT:malloc_alg=1   -INLINE:aggressive=on   -HP:bd=2m:heap=2m   -D__OPEN64_FAST_SET   -march=bdver1 

Fortran benchmarks:

 -Ofast   -LNO:blocking=off   -LNO:simd_peel_align=on   -OPT:rsqrt=2   -OPT:unroll_size=256   -HP:bd=2m:heap=2m   -mso   -march=bdver1 

Benchmarks using both Fortran and C:

 -Ofast   -OPT:malloc_alg=1   -HP:bd=2m:heap=2m   -IPA:plimit=8000   -IPA:small_pu=100   -mso   -march=bdver1   -LNO:blocking=off   -LNO:simd_peel_align=on   -OPT:rsqrt=2   -OPT:unroll_size=256 

Peak Compiler Invocation

C benchmarks:

 opencc 

C++ benchmarks:

 openCC 

Fortran benchmarks:

 openf95 

Benchmarks using both Fortran and C:

 opencc   openf95 

Peak Portability Flags

410.bwaves:  -DSPEC_CPU_LP64 
416.gamess:  -DSPEC_CPU_LP64 
433.milc:  -DSPEC_CPU_LP64 
434.zeusmp:  -DSPEC_CPU_LP64 
435.gromacs:  -DSPEC_CPU_LP64 
436.cactusADM:  -DSPEC_CPU_LP64   -fno-second-underscore 
437.leslie3d:  -DSPEC_CPU_LP64 
444.namd:  -DSPEC_CPU_LP64 
453.povray:  -DSPEC_CPU_LP64 
454.calculix:  -DSPEC_CPU_LP64 
459.GemsFDTD:  -DSPEC_CPU_LP64 
465.tonto:  -DSPEC_CPU_LP64 
470.lbm:  -DSPEC_CPU_LP64 
481.wrf:  -DSPEC_CPU_LINUX   -DSPEC_CPU_CASE_FLAG   -DSPEC_CPU_LP64   -fno-second-underscore 

Peak Optimization Flags

C benchmarks:

433.milc:  -Ofast   -CG:movnti=1   -CG:locs_best=on   -HP:bdt=2m:heap=2m   -IPA:plimit=7000   -IPA:callee_limit=1200   -OPT:struct_array_copy=2   -OPT:alias=field_sensitive   -mso   -march=bdver1 
470.lbm:  -Ofast   -CG:cmp_peep=on   -OPT:keep_ext=on   -HP:bdt=2m:heap=2m   -IPA:plimit=8000   -IPA:small_pu=100   -march=bdver1   -mso 
482.sphinx3:  -fb_create fbdata(pass 1)   -fb_opt fbdata(pass 2)   -Ofast   -m32   -IPA:plimit=1000   -OPT:malloc_alg=2   -CG:cmp_peep=on   -CG:p2align=0   -CG:load_exe=1   -CG:dsched=on   -INLINE:aggressive=on   -LNO:prefetch=2   -LNO:prefetch_ahead=4   -mso   -march=bdver2 

C++ benchmarks:

444.namd:  -Ofast   -IPA:plimit=3000   -LNO:ignore_feedback=off   -CG:local_sched_alg=0   -CG:load_exe=0   -OPT:unroll_size=256   -fno-exceptions   -HP:bdt=2m:heap=2m   -LNO:if_select_conv=1   -OPT:alias=disjoint   -LNO:psimd_iso_unroll=ON   -march=bdver1 
447.dealII:  -Ofast   -D__OPEN64_FAST_SET   -static   -INLINE:aggressive=on   -LNO:opt=1   -LNO:simd=2   -fno-emit-exceptions   -m32   -OPT:unroll_times_max=8   -OPT:unroll_size=256   -OPT:unroll_level=2   -HP:bdt=2m:heap=2m   -GRA:unspill=on   -CG:cmp_peep=on   -CG:movext_icmp=off   -TENV:frame_pointer=off   -march=bdver1 
450.soplex:  -fb_create fbdata(pass 1)   -fb_opt fbdata(pass 2)   -O3   -LNO:ignore_feedback=off   -INLINE:aggressive=on   -OPT:RO=1   -OPT:IEEE_arith=3   -OPT:IEEE_NaN_Inf=off   -OPT:fold_unsigned_relops=on   -fno-exceptions   -CG:p2align=0   -m32   -mno-fma4   -HP:bdt=2m:heap=2m   -WOPT:sib=on   -march=bdver1 
453.povray:  -fb_create fbdata(pass 1)   -fb_opt fbdata(pass 2)   -Ofast   -CG:pre_local_sched=off   -CG:p2align=0   -CG:p2align_split=on   -CG:dsched=on   -INLINE:aggressive=on   -HP:bd=2m:heap=2m   -OPT:transform=2   -OPT:alias=disjoint   -WOPT:aggcm=0   -march=bdver2 

Fortran benchmarks:

410.bwaves:  -fb_create fbdata(pass 1)   -fb_opt fbdata(pass 2)   -Ofast   -OPT:Ofast   -OPT:treeheight=on   -LNO:blocking=off   -LNO:ignore_feedback=off   -LNO:fu=4   -LNO:loop_model_simd=on   -LNO:simd_rm_unity_remainder=on   -WOPT:aggstr=0   -HP:bdt=2m:heap=2m   -CG:cmp_peep=on   -march=bdver1 
416.gamess:  -fb_create fbdata(pass 1)   -fb_opt fbdata(pass 2)   -Ofast   -LNO:fu=6   -LNO:blocking=0   -LNO:simd=2   -OPT:ro=3   -OPT:recip=on   -CG:local_sched_alg=1   -HP:bdt=2m:heap=2m   -WOPT:sib=on   -march=bdver1 
434.zeusmp:  -fb_create fbdata(pass 1)   -fb_opt fbdata(pass 2)   -Ofast   -LNO:blocking=off   -LNO:interchange=off   -IPA:plimit=1500   -HP:bdt=2m:heap=2m   -march=bdver1 
437.leslie3d:  -Ofast   -CG:pre_minreg_level=2   -LNO:simd=0   -LNO:fusion=2   -HP:bdt=2m:heap=2m   -mso   -march=bdver1 
459.GemsFDTD:  -Ofast   -IPA:plimit=1500   -OPT:unroll_size=1024   -OPT:unroll_times_max=16   -LNO:fission=2   -CG:local_sched_alg=2   -HP   -march=bdver1 
465.tonto:  -Ofast   -OPT:alias=no_f90_pointer_alias   -LNO:blocking=off   -CG:load_exe=1   -CG:local_sched_alg=3   -IPA:plimit=525   -HP:bdt=2m:heap=2m   -march=bdver1 

Benchmarks using both Fortran and C:

435.gromacs:  -Ofast   -OPT:rsqrt=2   -HP:bdt=2m:heap=2m   -CG:local_sched_alg=2   -CG:load_exe=3   -GRA:unspill=on   -march=bdver1   -LNO:simd=3 
436.cactusADM:  -fb_create fbdata(pass 1)   -fb_opt fbdata(pass 2)   -Ofast   -LNO:blocking=off   -LNO:prefetch=2   -LNO:pf2=0   -LNO:prefetch_ahead=4   -HP   -CG:locs_shallow_depth=1   -CG:load_exe=0   -CG:dsched=on   -WOPT:sib=on   -march=bdver1 
454.calculix:  -Ofast   -OPT:unroll_size=256   -OPT:alias=disjoint   -GRA:optimize_boundary=on   -CG:dsched=on   -HP:bdt=2m:heap=2m   -march=bdver1 
481.wrf:  -Ofast   -LNO:blocking=off   -LANG:copyinout=off   -IPA:callee_limit=5000   -GRA:prioritize_by_density=on   -HP   -WOPT:sib=on   -march=bdver1 

The flags files that were used to format this result can be browsed at
http://www.spec.org/cpu2006/flags/x86-open64-452-flags-rate-revA-III.html,
http://www.spec.org/cpu2006/flags/amd-Supermicro-platform-rate-revC-I.html.

You can also download the XML flags sources by saving the following links:
http://www.spec.org/cpu2006/flags/x86-open64-452-flags-rate-revA-III.xml,
http://www.spec.org/cpu2006/flags/amd-Supermicro-platform-rate-revC-I.xml.