SPEC® CFP2006 Result

Copyright 2006-2014 Standard Performance Evaluation Corporation

Fujitsu Limited

Fujitsu SPARC Enterprise T5240

CPU2006 license: 19 Test date: Mar-2008
Test sponsor: Fujitsu Limited Hardware Availability: May-2008
Tested by: Sun Microsystems Software Availability: Feb-2008
Benchmark results graph
Hardware
CPU Name: UltraSPARC T2 Plus
CPU Characteristics:
CPU MHz: 1415
FPU: Integrated
CPU(s) enabled: 16 cores, 2 chips, 8 cores/chip, 8 threads/core
CPU(s) orderable: 2 chips
Primary Cache: 16 KB I + 8 KB D on chip per core
Secondary Cache: 4 MB I+D on chip per chip
L3 Cache: None
Other Cache: None
Memory: 128 GB (32 x 4 GB)
Disk Subsystem: 737 GB RAID 5 using Sun StoreEdge
6120 with 12x 73 GB 10K RPM disks
(Seagate ST373307
2 Gbps Fibre Channel)
Other Hardware: None
Software
Operating System: Solaris 10 8/07 + patch 127111-08
Compiler: Sun Studio 12 and gccfss V4.2.0
(see additional detail below)
Auto Parallel: No
File System: ufs
System State: Default
Base Pointers: 32-bit
Peak Pointers: 32-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 127 11499 150   11345 152   11458 151   127 11499 150   11345 152   11458 151  
416.gamess 127 24264 102   24264 102   24261 102   127 23493 106   23483 106   23478 106  
433.milc 127 7980 146   7979 146   7987 146   127 7980 146   7979 146   7987 146  
434.zeusmp 127 10866 106   10851 107   10856 106   127 10834 107   10824 107   10843 107  
435.gromacs 127 8920 102   8916 102   8919 102   127 8719 104   8712 104   8706 104  
436.cactusADM 127 14707 103   14652 104   14742 103   127 14509 105   14492 105   14491 105  
437.leslie3d 127 12480 95.7 12525 95.3 12468 95.8 96 8882 102   8825 102   8842 102  
444.namd 127 8827 115   8836 115   8792 116   127 8736 117   8739 117   8675 117  
447.dealII 127 8194 177   8261 176   8221 177   127 7476 194   7509 193   7517 193  
450.soplex 127 9081 117   9037 117   9049 117   127 9081 117   9037 117   9049 117  
453.povray 127 4584 147   4484 151   4613 146   127 3378 200   3386 200   3404 198  
454.calculix 127 13523 77.5 13482 77.7 13430 78.0 127 13523 77.5 13482 77.7 13430 78.0
459.GemsFDTD 127 15163 88.9 15354 87.8 15416 87.4 127 15163 88.9 15354 87.8 15416 87.4
465.tonto 127 11250 111   11262 111   11253 111   112 8657 127   8712 127   8667 127  
470.lbm 127 35440 49.2 35485 49.2 35417 49.3 16 2579 85.3 2577 85.3 2576 85.3
481.wrf 127 11107 128   11092 128   11169 127   127 10989 129   10994 129   10993 129  
482.sphinx3 127 17207 144   17102 145   17121 145   127 16757 148   16792 147   16759 148  

Compiler Invocation Notes


 Sun Studio compiler patches are available at
 http://developers.sun.com/sunstudio/downloads/patches/ss12_patches.jsp
 The tested configuration included patch 124867-02, 124861-04,
 124863-01, 127000-01

 Peak also uses "GCC for SPARC Systems", which combines gcc
 with the Sun Code Generator for SPARC systems.  It is invoked
 as "gcc", and accepts source code compatible with GCC 4.2.
 For more information, including support, see
      http://cooltools.sunsource.net/gcc/

Operating System Notes


 Processes were bound to cores using "submit" and "pbind".
 A processor set was created using
   psrset -c 1-127
 and the runspec process was placed into the set using
   psrset -e 1

 These shell commands request use of local 4MB pages:
      export LD_PRELOAD=madv.so.1:mpss.so.1
      export MPSSHEAP=4MB
      export MPSSSTACK=4MB
      export MADV=access_lwp
 'access_lwp' means that the next light weight
 process to touch the specified address range
 will access it the most heavily.

 ulimit -s 131072 was used to allow the stack to grow
 up to 131072 KB (aka 128 MB).  Note that saying "131072"
 is preferable to "unlimited", because there is a tradeoff
 between space for the stack vs. space for the heap.

 The open file limit was set to 1300 with ulimit -n

 /etc/system parameters
    autoup=600
       Causes pages older than the listed number of seconds to
       be written by fsflush.
    bufhwm=3000
       Memory byte limit for caching I/O buffers
    segmap_percent=1
       Set maximum percent memory for file system cache
    tune_t_fsflushr=10
       Controls how many seconds elapse between runs of the
       page flush daemon, fsflush.
    tsb_rss_factor=128
       Suggests that the the size of the TSB (Translation Storage Buffer)
       may be increased if it is more than 25% (128/512) full.  Doing so
       may reduce TSB traps, at the cost of additional kernel memory.

 The "webconsole" service was turned off using
     svcadm disable webconsole

 The system had 144 GB of swap space.

Platform Notes


 This result was measured on a Sun SPARC Enterprise T5240.
 The Sun SPARC Enterprise T5240 and the Fujitsu SPARC
 Enterprise T5240 are electrically equivalent.

Base Compiler Invocation

C benchmarks:

 cc 

C++ benchmarks:

 CC 

Fortran benchmarks:

 f90 

Benchmarks using both Fortran and C:

 cc   f90 

Base Optimization Flags

C benchmarks:

 -g   -fast   -xipo=2   -xpagesize=4M   -xprefetch_level=2   -xalias_level=std   -xprefetch_level=3   -xprefetch_auto_type=indirect_array_access   -M /usr/lib/ld/map.bssalign 

C++ benchmarks:

 -g0   -library=stlport4   -fast   -xipo=2   -xpagesize=4M   -xprefetch_level=2   -xdepend   -xalias_level=compatible   -M /usr/lib/ld/map.bssalign 

Fortran benchmarks:

 -g   -fast   -xipo=2   -xpagesize=4M   -xprefetch_level=2   -M /usr/lib/ld/map.bssalign 

Benchmarks using both Fortran and C:

 -g   -fast(cc)   -fast(f90)   -xipo=2   -xpagesize=4M   -xprefetch_level=2   -xalias_level=std   -xprefetch_level=3   -xprefetch_auto_type=indirect_array_access   -M /usr/lib/ld/map.bssalign 

Base Other Flags

C benchmarks:

 -xjobs=32   -V   -# 

C++ benchmarks:

 -xjobs=32   -verbose=diags,version 

Fortran benchmarks:

 -xjobs=32   -V   -v 

Benchmarks using both Fortran and C:

 -xjobs=32   -V   -#   -v 

Peak Compiler Invocation

C benchmarks:

 cc 

C++ benchmarks (except as noted below):

 CC 
447.dealII:  g++ 

Fortran benchmarks:

 f90 

Benchmarks using both Fortran and C:

 cc   f90 

Peak Optimization Flags

C benchmarks:

433.milc:  basepeak = yes 
470.lbm:  -g   -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -fast   -xpagesize=4M   -M /usr/lib/ld/map.bssalign   -xprefetch_level=3   -xipo=2   -xrestrict 
482.sphinx3:  -g   -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -fast   -xpagesize=4M   -M /usr/lib/ld/map.bssalign   -xinline=   -xprefetch_level=2   -Wc,-Qlp-ol=1   -xrestrict   -xalias_level=strong   -fsimple=1   -xlinkopt=2   -lfast 

C++ benchmarks:

444.namd:  -g0   -library=stlport4   -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -fast   -xpagesize=4M   -xdepend   -xalias_level=compatible   -M /usr/lib/ld/map.bssalign   -xprefetch_level=1   -xlinkopt=2 
447.dealII:  -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -fast   -xpagesize=4M   -xdepend   -Wl,-M,/usr/lib/ld/map.bssalign   -xipo=2   -xrestrict   -xalias_level=std 
450.soplex:  basepeak = yes 
453.povray:  -g0   -library=stlport4   -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -fast   -xpagesize=64K   -xdepend   -xalias_level=compatible   -xipo=2   -xrestrict   -xlinkopt=2 

Fortran benchmarks:

410.bwaves:  basepeak = yes 
416.gamess:  -g   -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -fast   -xpagesize=4M   -M /usr/lib/ld/map.bssalign   -xlinkopt=2 
434.zeusmp:  -g   -fast   -xpagesize=4M   -M /usr/lib/ld/map.bssalign   -xipo=1   -qoption cg -Qeps:enabled=1   -qoption cg -Qeps:ws=8   -lmopt 
437.leslie3d:  -g   -fast   -xpagesize_heap=4M   -xpagesize_stack=64K   -M /usr/lib/ld/map.bssalign   -xprefetch_level=3   -xprefetch=latx:1.6   -qoption cg -Qlp=1   -qoption cg -Qlp-fa=0   -qoption cg -Qlp-fl=1   -qoption cg -Qlp-av=448   -qoption cg -Qlp-t=4 
459.GemsFDTD:  basepeak = yes 
465.tonto:  -g   -fast   -xpagesize=4M   -M /usr/lib/ld/map.bssalign   -xipo=2   -lfast 

Benchmarks using both Fortran and C:

435.gromacs:  -g   -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -fast(cc)   -fast(f90)   -xpagesize=4M   -M /usr/lib/ld/map.bssalign   -xipo=1   -xinline=   -xarch=generic   -xchip=generic   -fsimple=0 
436.cactusADM:  -g   -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -fast(cc)   -fast(f90)   -xpagesize=4M   -M /usr/lib/ld/map.bssalign   -xipo=2   -fsimple=1   -xlinkopt=2 
454.calculix:  basepeak = yes 
481.wrf:  -g   -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -fast(cc)   -fast(f90)   -xpagesize=4M   -M /usr/lib/ld/map.bssalign   -xlinkopt=2 

Peak Other Flags

C benchmarks:

 -xjobs=32   -V   -# 

C++ benchmarks (except as noted below):

 -xjobs=32   -verbose=diags,version 
447.dealII:  -v 

Fortran benchmarks:

 -xjobs=32   -V   -v 

Benchmarks using both Fortran and C:

 -xjobs=32   -V   -#   -v 

The flags file that was used to format this result can be browsed at
http://www.spec.org/cpu2006/flags/Sun-Solaris-Studio12-and-gccfss4.2.20090714.html.

You can also download the XML flags source by saving the following link:
http://www.spec.org/cpu2006/flags/Sun-Solaris-Studio12-and-gccfss4.2.20090714.xml.

SPEC and SPECfp are registered trademarks of the Standard Performance Evaluation Corporation. All other brand and product names appearing in this result are trademarks or registered trademarks of their respective holders.

For questions about this result, please contact the tester.
For other inquiries, please contact webmaster@spec.org
Copyright 2006-2014 Standard Performance Evaluation Corporation
Tested with SPEC CPU2006 v1.0.1.
Report generated on Tue Jul 22 16:50:18 2014 by SPEC CPU2006 HTML formatter v6932.
Originally published on 29 April 2008.