SPEC® CFP2006 Result

Copyright © 2006 Standard Performance Evaluation Corporation

Tyan

Thunder H2100R (S3993)

CPU2006 license # 94

Test sponsor: The Portland Group

Test date: May-2008

Hardware Avail: Apr-2008

Software Avail: Oct-2008

Benchmark results graph
Hardware
CPU Name: AMD Opteron 2352
CPU Characteristics:
CPU MHz: 2100
FPU: Integrated
CPU(s) enabled: 8 cores, 2 chips, 4 cores/chip
CPU(s) orderable: 1 to 2 chips
Primary Cache: 64 KB I + 64 KB D on chip per core
Secondary Cache: 512 KB I+D on chip per core
L3 Cache: 2 MB I+D on chip per chip
Other Cache: None
Memory: 16 GB (8 x 2 GB, PC5300, CL6, Reg, ECC)
Disk Subsystem: 1 x 130 GB SATA, 15000 RPM
Other Hardware: None
Software
Operating System: SLES 10 (Kernel 2.6.16.46-0.12-smp)
Compiler: The Portland Group (PGI)
PGI pgf95 7.1-1 Fortran Compiler
PGI pgcc 7.1-1 C Compiler
PGI pgCC 7.1-1 C++ Compiler
Auto Parallel: No
File System: ReiserFS
System State: Run Level 3 (Multi-user)
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 8 1500 72.6 1500 72.6 1500 72.6 8 1430 76.0 1430 76.0 1430 75.8
416.gamess 8 1540 102   1540 102   1540 102   8 1530 102   1530 102   1540 102  
433.milc 8 1480 49.5 1480 49.5 1480 49.5 8 1460 50.4 1460 50.4 1460 50.4
434.zeusmp 8 875 83.2 879 82.9 881 82.7 8 884 82.4 880 82.8 884 82.4
435.gromacs 8 723 79.0 724 78.9 724 78.9 8 583 98.0 585 97.6 583 98.0
436.cactusADM 8 1300 73.7 1300 73.6 1300 73.7 8 1300 73.7 1300 73.3 1300 73.4
437.leslie3d 8 1550 48.6 1540 48.7 1550 48.7 8 1550 48.6 1540 48.7 1550 48.7
444.namd 8 837 76.7 837 76.6 837 76.6 8 741 86.6 741 86.6 741 86.6
447.dealII 8 893 103   897 102   900 102   8 849 108   833 110   839 109  
450.soplex 8 1340 49.7 1340 49.8 1340 49.8 8 1340 49.7 1340 49.8 1340 49.8
453.povray 8 400 106   402 106   401 106   8 400 106   402 106   401 106  
454.calculix 8 606 109   606 109   606 109   8 607 109   609 108   609 108  
459.GemsFDTD 8 1860 45.6 1860 45.7 1860 45.6 8 1850 45.9 1850 45.9 1850 45.9
465.tonto 8 905 87.0 901 87.4 902 87.3 8 860 91.5 859 91.6 863 91.3
470.lbm 8 2400 45.8 2380 46.2 2370 46.4 8 2400 45.7 2380 46.2 2390 45.9
481.wrf 8 1080 82.6 1080 82.6 1090 82.2 8 1060 84.1 1060 84.6 1060 84.3
482.sphinx3 8 2120 73.4 2130 73.3 2130 73.3 8 1890 82.3 1890 82.4 1890 82.3

General Notes

Powersaved dameon was disabled in OS.
Environment stack size set to 'unlimited'
Max locked memory set to 2097152
Total number of huge pages available is 1200
numactl used to bind processes to CPUs

Compiler Invocation

C benchmarks:

pgcc

C++ benchmarks:

pgcpp

Fortran benchmarks:

pgf95

Benchmarks using both Fortran and C:

pgcc pgf95

Base Portability Flags

C benchmarks:

-DSPEC_CPU_LP64

C++ benchmarks (except as noted below):

-DSPEC_CPU_LP64
453.povray:
-DSPEC_CPU_LP64

Fortran benchmarks:

-DSPEC_CPU_LP64

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

-DSPEC_CPU_LP64 -Mnomain
436.cactusADM:
-DSPEC_CPU_LP64 -Mnomain
481.wrf:
-DSPEC_CPU_LP64 -DSPEC_CPU_LINUX -DSPEC_CPU_CASE_FLAG

Peak Portability Flags

C benchmarks:

-DSPEC_CPU_LP64

C++ benchmarks (except as noted below):

-DSPEC_CPU_LP64

Fortran benchmarks (except as noted below):

-DSPEC_CPU_LP64

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

-DSPEC_CPU_LP64 -Mnomain
436.cactusADM:
-DSPEC_CPU_LP64 -Mnomain
481.wrf:
-DSPEC_CPU_LP64 -DSPEC_CPU_LINUX -DSPEC_CPU_CASE_FLAG

Base Optimization Flags

C benchmarks:

-fast -Mfprelaxed -Mipa=fast -Mipa=inline -Msmartalloc=huge:150 -tp barcelona-64 -Bstatic_pgi

C++ benchmarks:

-fast --zc_eh -Mfprelaxed -Mipa=fast -Mipa=inline -Msmartalloc=huge:150 -tp barcelona-64 -Bstatic_pgi

Fortran benchmarks:

-fast -Mfprelaxed -Mipa=fast -Mipa=inline -Msmartalloc=huge:150 -tp barcelona-64 -Bstatic_pgi

Benchmarks using both Fortran and C:

-fast -Mfprelaxed -Mipa=fast -Mipa=inline -Msmartalloc=huge:150 -tp barcelona-64 -Bstatic_pgi

Peak Optimization Flags

C benchmarks:

433.milc:
-fast -Mfprelaxed -Mpfi(pass 1) -Mpfo(pass 2) -Mipa=noarg(pass 2) -Mipa=fast(pass 2) -Mipa=inline(pass 2) -Msmartalloc=huge:150 -Mdse -O4 -tp barcelona-64 -Bstatic_pgi
470.lbm:
-fastsse -Mfprelaxed -Mprefetch=distance:12 -Mprefetch=nta -Mipa=noarg -Mipa=fast -Mipa=inline -Msmartalloc=huge:150 -tp barcelona-64 -Bstatic_pgi
482.sphinx3:
-fastsse -Mfprelaxed -Mprefetch=distance:12 -Mprefetch=nta -Mpfi(pass 1) -Mpfo(pass 2) -Mipa=fast(pass 2) -Mipa=inline(pass 2) -Msmartalloc -tp barcelona-64 -Bstatic_pgi

C++ benchmarks (except as noted below):

basepeak = yes
444.namd:
-fast --zc_eh -Mfprelaxed -Mpfi(pass 1) -Mpfo(pass 2) -Mipa=fast(pass 2) -Mipa=inline(pass 2) -Msmartalloc=huge:150 -Munroll=n:4 -Munroll=m:8 -Mnodepchk -tp barcelona-64 -Bstatic_pgi
447.dealII:
-fast --zc_eh -Mfprelaxed -Mipa=fast -Mipa=inline -Msmartalloc=huge:150 -Mnovect -alias=ansi -tp barcelona-64 -Bstatic_pgi

Fortran benchmarks:

410.bwaves:
-fastsse -Mfprelaxed -Mprefetch=distance:12 -Mprefetch=nta -Mpfi(pass 1) -Mpfo(pass 2) -Mipa=fast(pass 2) -Mipa=inline(pass 2) -Msmartalloc -tp barcelona-64 -Bstatic_pgi
416.gamess:
-fast -Mfprelaxed -Mipa=fast -Mipa=inline -Msmartalloc=huge:150 -Mvect=noaltcode -tp barcelona-64 -Bstatic_pgi
434.zeusmp:
-fastsse -Mfprelaxed -Mipa=fast -Mipa=inline -Msmartalloc=huge:150 -tp barcelona-64 -Bstatic_pgi
437.leslie3d:
basepeak = yes
459.GemsFDTD:
-fast -Mfprelaxed -Mprefetch=nta -Mpfi(pass 1) -Mpfo(pass 2) -Mipa=fast(pass 2) -Mipa=inline(pass 2) -Msmartalloc=huge:150 -tp barcelona-64 -Bstatic_pgi
465.tonto:
-fast -Mfprelaxed -Mprefetch=distance:8 -Mipa=fast -Mipa=inline -Msmartalloc=huge:150 -Mvect=noaltcode -O4 -tp barcelona-64 -Bstatic_pgi

Benchmarks using both Fortran and C:

435.gromacs:
-fast -Mfprelaxed -Mfpapprox=rsqrt -Mipa=fast -Mipa=inline -Msmartalloc=huge:150 -tp barcelona-64 -Bstatic_pgi
436.cactusADM:
-fastsse -Mfprelaxed -Mipa=fast -Mipa=inline -Msmartalloc=huge:150 -Mdse -tp barcelona-64 -Bstatic_pgi
454.calculix:
-fastsse -Mfprelaxed -Mipa=fast -Mipa=inline -Msmartalloc=huge:150 -tp barcelona-64 -Bstatic_pgi
481.wrf:
-fast -Mfprelaxed -Mprefetch=distance:8 -Msmartalloc -Mvect=noaltcode -tp barcelona-64 -Bstatic_pgi

Other Flags

C benchmarks:

-w

C++ benchmarks:

-w

Fortran benchmarks:

-w

Benchmarks using both Fortran and C:

-w
W3C XHTML 1.0
W3C CSS