SPEC® CFP2006 Result

Hewlett-Packard Company

HP Integrity BL890c i2
(1.73 GHz/24MB Quad-Core Intel Itanium 9350)

SPECfp®_rate2006 = 520

SPECfp_rate_base2006 = 508

CPU2006 license:	03	Test date:	Apr-2010
Test sponsor:	Hewlett-Packard Company	Hardware Availability:	Jul-2010
Tested by:	Hewlett-Packard Company	Software Availability:	Mar-2010

Hardware
CPU Name:	Intel Itanium 9350
CPU Characteristics:	Intel Turbo Boost Technology up to 1.86 GHz
CPU MHz:	1730
FPU:	Integrated
CPU(s) enabled:	32 cores, 8 chips, 4 cores/chip, 2 threads/core
CPU(s) orderable:	4, 8 chips
Primary Cache:	16 KB I + 16 KB D on chip per core
Secondary Cache:	512 KB I + 256 KB D on chip per core
L3 Cache:	6 MB I+D on chip per core
Other Cache:	None
Memory:	512 GB (64 x 8GB 2Rx4 PC3-10600R)
Disk Subsystem:	4 x 73 GB 15K RPM SAS (1 OS, 3 SPEC)
Other Hardware:	None

Software
Operating System:	HP-UX 11i v3 Data Center Operating Environment B.11.31.1003
Compiler:	HP C/aC++ Developer's Bundle C.11.31.05 HP Fortran 90 Compiler B.11.31.11
Auto Parallel:	No
File System:	vxfs
System State:	Multi-user
Base Pointers:	32-bit
Peak Pointers:	32-bit
Other Software:	MallocNextGen B.11.31.0903.02

Results Table

Benchmark	Base							Peak
Benchmark	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	32	703	618	703	618	704	618	32	703	618	703	618	704	618
416.gamess	32	1868	335	1869	335	1868	335	32	1815	345	1827	343	1816	345
433.milc	32	692	424	693	424	693	424	32	710	414	709	414	709	414
434.zeusmp	32	563	517	564	517	563	517	32	563	517	564	517	563	517
435.gromacs	32	442	517	440	519	442	517	32	396	576	397	576	397	576
436.cactusADM	32	323	1180	323	1180	324	1180	32	323	1180	323	1180	324	1180
437.leslie3d	32	577	521	577	521	578	521	32	577	521	577	521	578	521
444.namd	32	290	885	290	886	290	886	32	290	885	290	886	290	886
447.dealII	32	542	675	538	681	536	683	32	542	675	538	681	536	683
450.soplex	32	963	277	962	277	961	278	32	939	284	933	286	934	286
453.povray	32	521	327	520	327	519	328	32	410	415	410	415	410	415
454.calculix	32	471	561	466	566	466	566	32	471	561	466	566	466	566
459.GemsFDTD	32	941	361	943	360	941	361	32	933	364	933	364	933	364
465.tonto	32	653	483	653	482	653	482	32	655	481	654	481	654	481
470.lbm	32	1170	376	1173	375	1170	376	32	1170	376	1173	375	1170	376
481.wrf	32	670	533	671	533	670	534	32	670	533	671	533	670	534
482.sphinx3	32	977	638	973	641	972	641	32	981	635	974	641	968	645

Submit Notes

The config file option 'submit' was used.
 The following config file entry was used to bind
 processes to cores using the HP-UX "mpsched" utility:
 submit = let "MYCPU=\$SPECCOPYNUM*2" ;mpsched -c \$MYCPU $command

Operating System Notes

   The following kernel tunables were set, in addition to the
   defaults set by the Base Operating Environment:

   filecache_max=25%
   filecache_min=25%
   maxdsiz=3221225472
   fcache_fb_policy=1
   base_pagesize=64
   pagezero_daemon_enabled=0
   vxfs_ifree_timelag=-1
   maxssiz=0x17f00000
   lcpu_attr=0
   numa_policy=4

Platform Notes

 Use of Hardware Threading by the OS was disabled via kctune

Base Compiler Invocation

Base Portability Flags

453.povray:	-DSPEC_CPU_NEED_INVHYP
481.wrf:	-DNOUNDERSCORE +noppu

Peak Compiler Invocation

C benchmarks:

/opt/ansic/bin/cc -AC99

C++ benchmarks:

/opt/aCC/bin/aCC -Aa

Fortran benchmarks:

/opt/fortran90/bin/f90

Benchmarks using both Fortran and C:

/opt/ansic/bin/cc -AC99 /opt/fortran90/bin/f90

Peak Portability Flags

453.povray:	-DSPEC_CPU_NEED_INVHYP
481.wrf:	-DNOUNDERSCORE +noppu

Peak Optimization Flags

C benchmarks:

433.milc:	+Oprofile=collect:all(pass 1) +Oprofile=use(pass 2) +Ofaster +Otype_safety=ansi -Wl,-a,archive_shared -Wl,+pd,64M -Wl,+pi,64M +Onoparmsoverlap -Wl,-N
470.lbm:	basepeak = yes
482.sphinx3:	Same as 433.milc

C++ benchmarks:

444.namd:	basepeak = yes
447.dealII:	basepeak = yes
450.soplex:	+Oprofile=collect:all(pass 1) +Oprofile=use(pass 2) +Ofaster +Otype_safety=ansi -Wl,-a,archive_shared -Wl,+pd,64M -Wl,+pi,64M +Onoparmsoverlap -Wl,-N -lmallocng
453.povray:	+Oprofile=collect:all(pass 1) +Oprofile=use(pass 2) +Ofaster +Otype_safety=ansi -Wl,-a,archive_shared -Wl,+pd,64M -Wl,+pi,64M -Wl,-N -lmallocng

Fortran benchmarks:

410.bwaves:	basepeak = yes
416.gamess:	+Ofaster -Wl,-a,archive_shared -Wl,+pd,64M -Wl,+pi,64M +Odataprefetch=direct -Wl,-N
434.zeusmp:	basepeak = yes
437.leslie3d:	basepeak = yes
459.GemsFDTD:	+Oprofile=collect:all(pass 1) +Oprofile=use(pass 2) +Ofaster -Wl,-a,archive_shared -Wl,+pd,64M -Wl,+pi,64M +Odataprefetch=direct -Wl,-N
465.tonto:	Same as 459.GemsFDTD

Benchmarks using both Fortran and C:

435.gromacs:	+Oprofile=collect:all(pass 1) +Oprofile=use(pass 2) +Ofaster +Otype_safety=ansi -Wl,-a,archive_shared -Wl,+pd,64M -Wl,+pi,64M +Onoparmsoverlap -Wl,-N
436.cactusADM:	basepeak = yes
454.calculix:	basepeak = yes
481.wrf:	basepeak = yes

The flags file that was used to format this result can be browsed at
http://www.spec.org/cpu2006/flags/Itanium-HPUX-1003-flags.20100511.html.

You can also download the XML flags source by saving the following link:
http://www.spec.org/cpu2006/flags/Itanium-HPUX-1003-flags.20100511.xml.