From YKRETCHM at altera.com  Thu Jan 26 13:51:22 2006
From: YKRETCHM at altera.com (Yaron Kretchmer)
Date: Thu, 26 Jan 2006 11:51:22 -0800
Subject: Petsc and LSF?
Message-ID: <D500AB0F5C034F48AB08127D359BEA1F0D094E@SJ-ISMSG01.altera.priv.altera.com>

Hi All

Has anyone used LSF to control processor assignments of Petsc/MPI
programs? Any information of doing that would be most welcome

 

Yaron

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20060126/0a8f3596/attachment.htm>

From bsmith at mcs.anl.gov  Thu Jan 26 15:16:28 2006
From: bsmith at mcs.anl.gov (Barry Smith)
Date: Thu, 26 Jan 2006 15:16:28 -0600 (CST)
Subject: Petsc and LSF?
In-Reply-To: <D500AB0F5C034F48AB08127D359BEA1F0D094E@SJ-ISMSG01.altera.priv.altera.com>
References: <D500AB0F5C034F48AB08127D359BEA1F0D094E@SJ-ISMSG01.altera.priv.altera.com>
Message-ID: <Pine.OSX.4.61.0601261515470.21804@barry-smiths-computer.local>


  Yaron,

    It should be the same as using LSF with any MPI programs
(that is PETSc shouldn't introduce any differences). 

  Barry

On Thu, 26 Jan 2006, Yaron Kretchmer wrote:

> Hi All
> 
> Has anyone used LSF to control processor assignments of Petsc/MPI
> programs? Any information of doing that would be most welcome
> 
>  
> 
> Yaron
> 
>  
> 
> 



From YKRETCHM at altera.com  Mon Jan 30 00:44:33 2006
From: YKRETCHM at altera.com (Yaron Kretchmer)
Date: Sun, 29 Jan 2006 22:44:33 -0800
Subject: usage of MatPreallocInitialize/MatPreallocFinalize
Message-ID: <D500AB0F5C034F48AB08127D359BEA1F0D0961@SJ-ISMSG01.altera.priv.altera.com>

Hi All
How does one use MatPreallocInitialize/MatPreallocFinalize? Any code
examples? The documentation is a bit sketchy

Regards
Yaron





From knepley at mcs.anl.gov  Mon Jan 30 06:01:43 2006
From: knepley at mcs.anl.gov (Matthew Knepley)
Date: Mon, 30 Jan 2006 06:01:43 -0600
Subject: usage of MatPreallocInitialize/MatPreallocFinalize
In-Reply-To: <D500AB0F5C034F48AB08127D359BEA1F0D0961@SJ-ISMSG01.altera.priv.altera.com> (Yaron
 Kretchmer's message of "Sun, 29 Jan 2006 22:44:33 -0800")
References: <D500AB0F5C034F48AB08127D359BEA1F0D0961@SJ-ISMSG01.altera.priv.altera.com>
Message-ID: <y8hslr5apqw.fsf@harley.mcs.anl.gov>

"Yaron Kretchmer" <YKRETCHM at altera.com> writes:

> Hi All
> How does one use MatPreallocInitialize/MatPreallocFinalize? Any code
> examples? The documentation is a bit sketchy

  These are internal routines. You would use Mat<type>SetPreallocation(),
like MatSeqAIJSetPreallocation().

     Matt
-- 
"Failure has a thousand explanations. Success doesn't need one" -- Sir Alec Guiness



From bsmith at mcs.anl.gov  Mon Jan 30 10:31:07 2006
From: bsmith at mcs.anl.gov (Barry Smith)
Date: Mon, 30 Jan 2006 10:31:07 -0600 (CST)
Subject: usage of MatPreallocInitialize/MatPreallocFinalize
In-Reply-To: <y8hslr5apqw.fsf@harley.mcs.anl.gov>
References: <D500AB0F5C034F48AB08127D359BEA1F0D0961@SJ-ISMSG01.altera.priv.altera.com>
 <y8hslr5apqw.fsf@harley.mcs.anl.gov>
Message-ID: <Pine.LNX.4.58.0601301026220.30945@harley.mcs.anl.gov>


  Actually they are not really internal routines; though their
documentation is a bit short. You can see uses of them in
src/dm/da/utils/fdda.c

  General statement. Etags, (see
http://www-unix.mcs.anl.gov/petsc/petsc-as/snapshots/petsc-dev/docs/codemanagement.html
and search for etags)
are an enormously powerful tool to get around the PETSc source code. In
this case you could simply use etags to search for MatPreallocInitialize()
and immediately locate all of its uses. There is also a TAGS for VI, check
VI's docs.

   Barry

On Mon, 30 Jan 2006, Matthew Knepley wrote:

> "Yaron Kretchmer" <YKRETCHM at altera.com> writes:
>
> > Hi All
> > How does one use MatPreallocInitialize/MatPreallocFinalize? Any code
> > examples? The documentation is a bit sketchy
>
>   These are internal routines. You would use Mat<type>SetPreallocation(),
> like MatSeqAIJSetPreallocation().
>
>      Matt
> --
> "Failure has a thousand explanations. Success doesn't need one" -- Sir Alec Guiness
>
>



From balay at mcs.anl.gov  Mon Jan 30 10:34:03 2006
From: balay at mcs.anl.gov (Satish Balay)
Date: Mon, 30 Jan 2006 10:34:03 -0600 (CST)
Subject: usage of MatPreallocInitialize/MatPreallocFinalize
In-Reply-To: <Pine.LNX.4.58.0601301026220.30945@harley.mcs.anl.gov>
References: <D500AB0F5C034F48AB08127D359BEA1F0D0961@SJ-ISMSG01.altera.priv.altera.com>
 <y8hslr5apqw.fsf@harley.mcs.anl.gov> <Pine.LNX.4.58.0601301026220.30945@harley.mcs.anl.gov>
Message-ID: <Pine.LNX.4.64.0601301032270.13733@asterix>

MatPreallocInitialize appears to be a typo [for
MatPreallocateInitialize] Will fix this in petsc-dev.

Satish

On Mon, 30 Jan 2006, Barry Smith wrote:

> 
>   Actually they are not really internal routines; though their
> documentation is a bit short. You can see uses of them in
> src/dm/da/utils/fdda.c
> 
>   General statement. Etags, (see
> http://www-unix.mcs.anl.gov/petsc/petsc-as/snapshots/petsc-dev/docs/codemanagement.html
> and search for etags)
> are an enormously powerful tool to get around the PETSc source code. In
> this case you could simply use etags to search for MatPreallocInitialize()
> and immediately locate all of its uses. There is also a TAGS for VI, check
> VI's docs.
> 
>    Barry
> 
> On Mon, 30 Jan 2006, Matthew Knepley wrote:
> 
> > "Yaron Kretchmer" <YKRETCHM at altera.com> writes:
> >
> > > Hi All
> > > How does one use MatPreallocInitialize/MatPreallocFinalize? Any code
> > > examples? The documentation is a bit sketchy
> >
> >   These are internal routines. You would use Mat<type>SetPreallocation(),
> > like MatSeqAIJSetPreallocation().
> >
> >      Matt
> > --
> > "Failure has a thousand explanations. Success doesn't need one" -- Sir Alec Guiness
> >
> >
> 
>