Re: X-FEM mini-course

Dear Stéphane,

Yes, you answered my questions and no, I don't think at all that you 
bothered somebody by posting the announcement in this mailing list. My 
questions had another motivation. Maybe you might have seen on the web 
page that David Abrahams and I have started to develop a  new version 
of MTL. One of my goals is that the MTL is widely applicable. Therefore 
I am vividly interested in what it is used for and also willing to 
extend the functionality to better support scientific computing. There 
are serious plans to integrate MTL into PETSc. Part of this 
collaboration with Argonne will be to provide a matrix format for FEM 
codes, which is a list of element matrices that are dense and for 
higher order elements large enough to benefit from dense matrix 
performance enhancements.

I would go even a step further: in scientific computing one has either 
complete systems with there own data structure and a set of algorithms 
that solves problems of a more or less narrow domain or one has 
general-purpose numeric libraries that usually have the attitude to 
pass the message "give me a matrix (dense or sparse) and a rhs vector 
and I solve you a linear system (eventually a non-linear system) but 
how do you get to this linear system is none of my business". (Maybe I 
have bothered some people now.) I am thinking that a library like MTL 
can be also useful in the process of setting up linear or non-linear 
systems, i.e. FEM. Probably, the MTL is already helpful for this but I 
have the feeling that an intense communication with FEM software 
developer and their users can improve this further. I am also playing 
with the idea to come to your mini-course and eventually present 
something, if you are interested.

Best Regards,

Peter

P.S.: La commune des utilisateurs de MTL est d'ailleurs assez 
francophone. J'avais récemment des communications intensives avec 
Nantes et Rennes et je suis sûr ce-qu'il y'a encore des autres. Un 
projet (où je m'engage aussi) pour établir des concepts et de logiciel 
générique pour des systèmes d'algèbre linéaire ce-que est pour ça relié 
très proche avec MTL (cf glas.sourceforge.net) est gérer en Belgique.

On Jul 25, 2005, at 8:32 AM, Stephane BORDAS wrote:

> Dear Peter,
>
> Thank you for your reply. I believe indeed that people who use the MTL 
> are likely to do computational work. Among these people, there is a 
> good chance that a good share work on finite element methods, and may 
> be interested in learning how the finite element method can be 
> "extended" into the "Extended Finite Element Method" to solve complex 
> problems such as crack growth, fluid-structure interaction, etc. In 
> addition, the X-FEM allows to increase computational efficiency, when 
> compared to FEM. People who use MTL are interested in efficiency, 
> therefore, they probably are, also, interested in X-FEM.
>
> The MTL will not be presented, but I am working with some 
> collaborators on a new X-FEM implementation in which MTL may be used 
> to handle the linear algebra.
>
> Did I answer your question? I hope that I have not bothered too many 
> people by posting this message here, but I truly feel that users of 
> the MTL are likely to take advantage of the course we are offering.
>
> I would welcome any comment, by anyone, on any of the above.
>
> Thank you, and best regards to you, Peter, and to the whole list (many 
> people, that is, :),
>
> Stephane
>
>
>
> Peter Gottschling wrote:
>
>> Dear Stéphane,
>>
>> Thank you for this information. Can you please say a few words how  
>> tutorial and/or software is related to the MTL. Is it used in the  
>> framework, is it a potential application for your framework, will the 
>>  MTL be presented in the mini-course, or is it just that you think 
>> that  the same people are interested in your FEM work and the MTL? 
>> Thank you  in advance.
>>
>> Best Regards,
>>
>> Peter
>>
>> On Jul 19, 2005, at 5:52 AM, Stephane BORDAS wrote:
>>
>>> Dear colleagues,
>>>
>>> This is to announce an eXtended Finite Element mini-course at EPFL,
>>> Lausanne, Switzerland, December 7-9 2005
>>>
>>> http://continuing-education.epfl.ch/Jahia/site/continuing-education/ 
>>> op/preview/lang/en/pid/12861
>>>
>>> This method allows to treat free-boundary, singularities, strong and
>>> weak discontinuities seamlessly within an FEM framework.
>>>
>>> Please contact me if you are interested in attending, and pass this
>>> information around you if you see fit.
>>>
>>> Thank you, and best regards,
>>>
>>> Stephane
>>>
>>> -- 
>>> _________________________________________________________________
>>> Stephane Bordas
>>> http://www.tam.northwestern.edu/sbordas
>>>
>>> Laboratory of structural and continuum mechanics
>>> Laboratoire de mécanique des structures et milieux continus (LSC)
>>> EPFL/ENAC/IS/LSC GC A2 454
>>> Station 18 EPFL-LSC, Ecublens CH-1015 Lausanne
>>> Suisse/Switzerland/Schweiz
>>>
>>> Tél: +41 21 693 24 03 Fax: +41 21 693 63 40
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> This list is archived at 
>>> http://www.osl.iu.edu/MailArchives/mtl-devel/
>>>
>> ------------
>> Peter Gottschling
>> Research Associate
>> Open Systems Laboratory
>> Indiana University
>> 301i Lindley Hall
>> Bloomington, IN 47405
>> Tel.: +1 812 855-8898   Fax: +1 812 856 0853
>> http://www.osl.iu.edu/~pgottsch
>>
>>
>
> -- 
> _________________________________________________________________
> Stephane Bordas
> http://www.tam.northwestern.edu/sbordas
>
> Laboratory of structural and continuum mechanics
> Laboratoire de mécanique des structures et milieux continus (LSC)
> EPFL/ENAC/IS/LSC GC A2 454
> Station 18 EPFL-LSC, Ecublens CH-1015 Lausanne
> Suisse/Switzerland/Schweiz
>
> Tél: +41 21 693 24 03 Fax: +41 21 693 63 40
>
>
>
>
------------
Peter Gottschling
Research Associate
Open Systems Laboratory
Indiana University
301i Lindley Hall
Bloomington, IN 47405
Tel.: +1 812 855-8898   Fax: +1 812 856 0853
http://www.osl.iu.edu/~pgottsch


_______________________________________________
This list is archived at http://www.osl.iu.edu/MailArchives/mtl-devel/