MSC.Nastran V2001.0.9 Distributed Memory Parallel Test Results

This page contains the results of several jobs running on different hardware platforms. The times listed in each table are elapsed time. The plots show speedup, defined as the serial time divided by the parallel time. Each job was run by hardware partners.

Below is a summary of the test problems. Note that resource requirements (memory, scratch disk used, and total I/O) are distributed across processors. Per-process figures may be roughly determined by dividing the totals below by the number of processors used.

Name	Ndof	Description	SOL	MEM	SCR Disk Used	Total I/O	Comments
LGQDF	31,125	Cube w/ interior	108	100 Mb	0.6 Gb	700 Gb	76 Frequency Increments
XLEMF	654,560	Car Body	111	400 Mb	11 Gb	328 Gb	448 + 34 Roots
XLTDF	525,027	Car Body	108	450 Mb	5 Gb	209 Gb	32 Frequency Increments
XXCMDA	1,584,622	Car Body	103	400 Mb	31 Gb	730 Gb	1073 Roots (PACMS)
XXCMDF	1,584,622	Car Body	103	800 Mb	43 Gb	2400 Gb	1073 Roots (FDMODES)
XXCMDG	1,584,622	Car Body	103	400 Mb	43 Gb	2400 Gb	1073 Roots (GDMODES)

Below is a description of the hardware used. Detailed computer, disk, network and OS configuration is listed the bottom of this page.

Vendor	Hardware	O/S
HP/Compaq	GS1280 1.15GHz 21364	Tru64UNIX v5.1B
Cray	SV1ex	UNICOS 10.0
Fujitsu	VPP5000 9.6 Gflops	UXP.V V29L10
HP	HP SuperDome 875 MHz PA8700	HP-UX 11.11
IBM	IBM p655 1300 MHz POWER4	AIX 5.1 ML03
Intel HPUX IA64	IPF-2 1000 MHz	HP-UX 11.22
Intel Linux IA64	Intel Itanium2 1000 MHz (NEC TX7)	RedHat 7.2 (LINUX64)
Intel Linux IA32	HP X4000 P4 Xeon 1700 MHz	MSC.Linux 2.4.6-1msc-smp
NEC	SX-6	SuperUX 12.1
SGI	Origin 3000 / 600 MHz	IRIX64 6.5
Sun	UltraSPARC 900 MHz	Solaris 8

• DMP is recognized in SOL 101, 103, 108, 111, and 112.
• Each task requires enough memory for the size of executable, the amount specified by the MEM parameter, and the amount needed for system I/O buffers. On true distributed systems, this amount is on each node. On SMP systems, all the tasks are on the same node.
• PACMS (example XXCMDA) also runs in serial. The serial ACMS result for XXCMDA is shown in the "dmp=1" column. A DMP license is not required for serial ACMS.
• This page is maintained by Peter Schartz.

Hardware Details:

• Compaq

      Model:            HP/HP Alphaserver GS1280 (16 CPUs)
      CPU:              1.15GHz 21364
      OS Level:         Tru64UNIX v5.1B
      Memory:           4GB memory per cpu
      Disk:             8 - I/O shelves hooked to 8 I/O port. Each I/O shelf had 40 disks on 8 scsi3 controllers.
      

• Cray

      Model:            SV1ex (32 cpus)
      CPU:              SV1ex 500MHz
      OS Level:         UNICOS 10.0.0lf
      Memory:           2 Gbyte (4 GW) memory
      Disk:             96 Gbyte (12 GW) SSDI, configured with:
                        /ssd   4 Gbytes (0.5 GW)
                        swap  32 Gbytes (4.0 GW)
                        sds   60 GBytes (7.5 GW)
      

• Fujitsu

      Model:            Fujitsu VPP5000 9.6 Gflops, VX
      OS Level:         UXP/V V20L10
      Memory:           16Gb
      

• HP

      Model:            Superdome 9000/800/S16K-A (16 CPUs)
      CPU:              PA-RISC 8700 875MHz
      OS Level:         HP-UX 11.11
      Memory:           32 GB Real memory
                        50 GB Swap space
      Disk:             8 fiberchannel controller/40 fc disk scratch stripe
      

• IBM

      Model:            p655 (4 CPUs)
      CPU:              POWER4 1300 MHz
      OS Level:         AIX 5.1 ML03
      Memory:           32 GB memory, 16 GB set as large pages
      Disk:             JFS striped at 128 KB on 16 (36GB) SCSI disks on 4 Ultra160 adapters.
                        32-bit V2001.0.9 executable enabled for large pages
      

• Intel HP IA64

      Model:            rx5670 (4 CPUs)
      CPU:              IPF-2 1000 Mhz
      OS Level:         HP-UX 11.22
      Memory:           48 gb Real memory
                        50 gb Swap space
      Disk:             2 fiber channel controller / 10 fc disk scratch stripe
      

• Intel Linux IA64

      Model:            NEC Express5800/1000 (TX7) (16 CPUs)
      CPU:              Intel Itanium2 1000Mhz
      OS Level:         RedHat 7.2 (LINUX64)
      Memory:           32 Gb
      Disk:             Two 4-way Stripped Ultra 320 SCSI Disks
      

• Intel Linux IA32

      Model:            HP 4000
      CPU:              P4 Xeon (1700 MHz)
      OS Level:         MSC.Linux 2.4.6-1.msc-smp
      Memory:           3 Gb
      Disk:             3X36Gb Software striped IBM Ultrastar DDYS-T36950
      

• NEC

      Model:            SX-6
      CPU:              SX-6 Vector processor (Max 8 GFLOPs per processor)
      OS Level:         SuperUX 12.1
      Memory:           64 GB (Real)
      Disk:             Two 4-way striped RAID disks
      

• SGI

      Model:            Origin 3000 / 600 MHz
      CPU:              R1400A 600MHz
      OS Level:         IRIX64 6.5.14f
      Memory:           32 Gb
      Disk:             8 lun TP9400 RAID file system.
      

• SUN

      Model:            UltraSPARC 900 MHz
      OS Level:         Solaris 8