XANSYS Message: 278 [Go back to message list]
[bookmark on del.icio.us]
No rating yet
Rate item:

Subject: Re: ANSYS and parallel processing...
Author: John Swanson
Date: 1999-01-15 19:46:00

I am afraid I must differ. The NT system is a multi-threaded
operating system, which
means that if you have more than one processor, and the program you are
running (such as
ANSYS) is designed to run in a multi-thread shared memory environment,
then you can get
parallel processing.
Perhaps I should clarify the two types of parallel:

1. Shared memory, multi thread.
Several processors
One memory, seen by all processors
Multiple threads (from 1 to some number greater than the
number of processors)
Usually best parallel performance is obtained with n-1
processors, where n is the number
of installed processors.
This form of parallel processing is supported by ANSYS on
several systems, including NT.
(That said, because of bugs in the parallel operating system,
only the frontal solver was supported
in parallel at 5.4. More complete support is available at
5.5)

2. Distributed memory multiprocessing
Several processors
Private memory for each processor
Usually more processors that shared memory systems, because
the private memory
removes the memory bottleneck in the system performance
Message passing (MPI, for example) between processes.
Each process has its own code and data.
MUCH more complex to develop software, and much more
expensive hardware systems.
Domain decomposition is the standard approach for mapping
the finite element task on this
type of system. ANSYS does not (yet) support this type of
hardware, except for the trivial
case of a thru-put machine, where many copies of ANSYS, each
doing a different task, run
on the many processors.

John.Swanson

> -----Original Message-----
> From: Breaux, Brian [SMTP:bbreaux1@e...]
> Sent: Friday, January 15, 1999 5:13 PM
> To: 'xansys@o...'
> Subject: [xansys] Re: ANSYS and parallel processing...

> From: "Breaux, Brian"

> Viktor,

> From information I have been given, Intel for the standard pc with NT
> can
> not do true parallel processing. It does symmetric multiprocessing
> which
> does not evenly distribute a problem across all processors.
> Typically,
> processors that are not as intensive will utilize the second or less
> tasked
> processor.

> Applications such as ansys for pc's have a simple multiprocessing
> algorithm
> where by simple summations can be divided between available processors
> and
> then re-assembled for a final solution. This is different than a
> truly
> parallel system such as a Cray which will have multiple processors
> solve the
> same algorithm.

> However, multiple processors generally benefit smaller computers as at
> no
> time is ansys the only application working. General background tasks
> will
> be handled by the less tasked processor. But this is not the same as
> "parallel" processing as we have been discussing.

> Brian D. Breaux
> Design Engineer, Central Engineering, JEC.
> 281-228-7716

> > From: Viktor Doundakov[SMTP:Viktor.Doundakov@R...]
> > Sent: Friday, January 15, 1999 3:51 PM
> > To: 'xansys@o...'
> > Subject: [xansys] Re: ANSYS and parallel processing...

> > From: Viktor Doundakov

> > Hello Brian,

> > Can you elaborate a little bit more?

> > Viktor Doundakov
> > R&D Engineer
> > Read-Rite Corporation
> > Direct: 510-683-7495 Main: 510-683-6100 FAX: 510-683-7065
> > e-mail: Viktor.Doundakov@r...

> > -----Original Message-----
> > From: Breaux, Brian [mailto:bbreaux1@e...]
> > Sent: Friday, January 15, 1999 12:05 PM
> > To: 'xansys@o...'
> > Subject: [xansys] Re: ANSYS and parallel processing...

> > From: "Breaux, Brian"

> > You can not do true parallel processing on a Pentium II
> > Brian D. Breaux
> > Design Engineer, Central Engineering, JEC.
> > 281-228-7716

> > > From: Viktor Doundakov[SMTP:Viktor.Doundakov@R...]
> > > Sent: Friday, January 15, 1999 1:51 PM
> > > To: 'xansys@o...'
> > > Subject: [xansys] ANSYS and parallel processing...

> > > From: Viktor Doundakov

> > > Hi Joseph,

> > > I've heard that the ANSYS parallel processing is not very good
> (and in
> > > general I am told that it holds for the FEA since you work on
> matrixes).
> > I
> > > have a dual Pentium II processor PC and if I am wrong on the
> subject and
> > I
> > > can take advantage of parallel processing it will help me a lot
> since I
> > am
> > > starting to run relatively big jobs.

> > > Anybody with info on the subject will help.

> > > Viktor Doundakov
> > > R&D Engineer
> > > Read-Rite Corporation
> > > Direct: 510-683-7495 Main: 510-683-6100 FAX: 510-683-7065
> > > e-mail: Viktor.Doundakov@r...

> > > -----Original Message-----
> > > From: Metrisin, Joseph T. [mailto:metrisin@p...]
> > > Sent: Friday, January 15, 1999 9:34 AM
> > > To: 'xansys@o...'
> > > Subject: [xansys] Re: Multiple CPU's in SUN environment

> > > From: "Metrisin, Joseph T."

> > > Geno,

> > > If you are looking at the CPU time printed at the bottom of the
> > > ANSYS output, this is the TOTAL CPU time accumulated by
> > > all four processors. In other words, your job still takes nearly
> > > the same amout of CPU time to solve, it's just spread out over
> > > the various processors. If you look at the elapsed time, it
> > > will probably show that it solved in roughly 1/4 of the wall clock
> > > time.

> > > Joseph T. Metrisin

> > > Structural Methods
> > > Pratt & Whitney/Unitetd Technologies
> > > M/S: 714-03
> > > P.O. Box 109600
> > > West Palm Beach, FL. 33410-9600

> > > Phone: (561) 796-5967
> > > Fax: (561) 796-8993

> > > > From: Geno Vallejos[SMTP:gvalle@c...]
> > > > Sent: Friday, January 15, 1999 12:01 PM
> > > > To: xansys@o...
> > > > Subject: [xansys] Multiple CPU's in SUN environment

> > > > From: Geno Vallejos

> > > > To all my fellow ANSYS users:

> > > > I have had some interesting results when solving the same
> structural
> > > > problem using the same model but on two different SUN machines.
> > > > Below are the specifics;

> > > > The structural model is a part created in ProENGINEER 18
> > > > I am using ANSYS ProFEA 5.4 and ProENGINEER 18.

> > > > Solution #1:
> > > > Machine: SUN E-3000 server
> > > > Processors: 4 each 360 MHz CPU's
> > > > Operating System: Solaris 2.6
> > > > CP Time (from the file.lst summary) = 10 minutes 28 seconds

> > > > The first time I requested four CPU's using the config.ans file,
> with
> > > the
> > > > following line added;
> > > > NUM_PROC = 4
> > > > The config.ans file was located in the /docu directory.
> > > > The file.lst indicated that 4 CPU's will be used

> > > > The second time I requested four CPU's using a .inp file. The
> .inp
> > > file
> > > > had
> > > > the following line;
> > > > /config,nproc,4

> > > > The solution times (CP Time listed above) was 10 minutes 28
> seconds

> > > > Solution #2:
> > > > Machine: SUN Ultra 60 workstation
> > > > Processors: 1 each 360 MHz CPU
> > > > Operating System: Solaris 2.6
> > > > CP Time (from the file.lst summary) = 10 minutes 53 seconds

> > > > Can anyone help explain why the E-3000 server and the Ultra 60
> > > > workstation solved the structural analysis problem by taking the
> same
> > > > amount of time????

> > > > I have heard some ideas....but nothing concrete. Some here at
> ATL say
> > > > that the four SUN processors will work as hard as they need to
> to
> > solve
> > > > the problem. I say that I would like them to solve the problem
> as
> > fast
> > > as
> > > > possible. Observing the CPU activity bar during the solution
> phase
> > > > indicated that at no time were any of the four CPU's working any
> > higher
> > > > than 40-50% of maximum capacity. Was this due to some SUN or
> Solaris
> > > > setting "chocking" the CPU speed or efficiency; or some ANSYS
> setting
> > > > not getting the full support of multi-tasking from the SUN
> server??

> > > > Thanks in advance!

> > > > Geno Vallejos
> > > > ATL Ultrasound - A Philips Company
> > > > Bothell, Washington 98021
> > > > www.atl.com

Posts possibly associated with message #278AuthorDateScore
263ANSYS and parallel processing...Viktor Doundakov1999/01/15 
265Re: ANSYS and parallel processing...John Swanson1999/01/15 
267Re: ANSYS and parallel processing...Brian Breaux1999/01/15 
268Re: ANSYS and parallel processing...Brian Murphy1999/01/15 
271Re: ANSYS and parallel processing...Joe Metrisin1999/01/15 
273Re: ANSYS and parallel processing...Viktor Doundakov1999/01/15 
275Re: ANSYS and parallel processing...Brian Breaux1999/01/15 
278Re: ANSYS and parallel processing...John Swanson1999/01/15 
286Re: ANSYS and parallel processing...Terje Fredriksen1999/01/18