[Trilinos-Users] Trilinos-Users Digest, Vol 66, Issue 6

[rrossi at cimne.upc.edu]

dear alan,

i just reread your email (mpi coding was getting quickly very nasty :D).
what you are proposing makes a lot of sense to me. i will have a look to it
and see if i can modify it locally
in my installation. the clean way would be to add a new argument which
defaults to ADD but that could be whatever. ideally a user defined
operation could be added, but this could present problems.

thank you for the hint in any case.

Riccardo




On Mon, 07 Feb 2011 12:00:02 -0700,
trilinos-users-request at software.sandia.gov wrote:
> Send Trilinos-Users mailing list submissions to
> 	trilinos-users at software.sandia.gov
> 
> To subscribe or unsubscribe via the World Wide Web, visit
> 	http://software.sandia.gov/mailman/listinfo/trilinos-users
> or, via email, send a message with subject or body 'help' to
> 	trilinos-users-request at software.sandia.gov
> 
> You can reach the person managing the list at
> 	trilinos-users-owner at software.sandia.gov
> 
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Trilinos-Users digest..."
> 
> 
> Today's Topics:
> 
>    1. Proposal concerning the ReplaceLocalValues (Riccardo Rossi)
>    2. Re: Proposal concerning the ReplaceLocalValues (Williams, Alan B)
>    3. Re: Proposal concerning the ReplaceLocalValues (Heroux, Michael A)
> 
> 
> ----------------------------------------------------------------------
> 
> Message: 1
> Date: Mon, 07 Feb 2011 12:48:47 +0100
> From: "Riccardo Rossi" <rrossi at cimne.upc.edu>
> Subject: [Trilinos-Users] Proposal concerning the ReplaceLocalValues
> To: trilinos-users at software.sandia.gov
> Message-ID: <1297079327.28887.7.camel at rrossi-desktop>
> Content-Type: text/plain; charset=utf-8
> 
> Dear List,
> 
>     as pointed out last week, the ReplaceGlobalValues right now performs
> a sum of the non_local values to be replaced.
> 
>  Would it be possible to customize such behaviour so that for exampel
> one could prescribe the "maximum" or "minimum" value to be preserved?
> 
> in this way the result would be deterministic, and still an operation
> closer to the replace would be performed.
> 
> aside of this, i had a look to the code inside, and honestly i don't
> understand the current behaviour since a REPLACE is used inside the
> implementation. Can anyone point me to where the actual sum is
> performed?
> 
> thank you
> Riccardo