[CIG-LONG] [CIG-SHORT] Short-Term Crustal Dynamics priorities - mesh generation

[Thorsten Becker]

Carl,

Thanks for your comments and offer to help! I forward this to the long
term deformation group who might be interested as well. Right now,
element types in the GALE code, for example, are a bit limited and
issues such as remeshing would required tighter integration. However,
mesh generation issues have come up in our discussions as well. 

Cheers

Thorsten

On Thu, 2008-05-15 at 11:14 -0600, Carl W. Gable wrote:
> RE: mesh generation
> 
> Brad and community,
> 
> I'm interested in community feedback both in terms of CIG priorities but 
> also, if people have feedback for me it will help me focus what I bring 
> to the table at the Golden workshop in June. Comments welcome.
> 
> It is my impression that mesh generation for systems with complex 
> geometry is still a steep hurdle for many people doing short-term 
> crustal dynamics.
> 
> I would like to try to help people overcome these hurdles in any way I 
> can. For me this means development of methods and algorithms for LaGriT. 
> I do not advocate LaGriT as 'the' solution, just the solution in which I 
> have experience and expertise and can offer something to the community.
> 
> Up to now I have produced various demo meshes and made them available to 
> the community ( http://meshing.lanl.gov/proj/#Crustal_Dynamics_open ) 
> and created output file structures specifically for PyLith and Geofest.
> 
> A recent addition to LaGriT that should be available by the time of the 
> workshop is the ability to read/write netCDF format files. This is a 
> more general exchange format and it allows one to read a mesh produced 
> by Cubit into LaGriT for modification and/or setting of IC and BC.
> 
> I believe the problem of meshing the SCEC CFM (community fault model) is 
> solved, but the steps are rather difficult and require a lot of tweaking.
> 
> Rather than saying 'this is what the community needs' I would ask the 
> questions:
> 
> Do people want mesh generation:
> A) That is easier to use for simple problems
> or
> B) New algorithms for solving complex problems (non-manifold geometry, 
> mesh quality improvement).
> or
> C) More demo problems relevant to crustal dynamics
> or
> D) Data structures and algorithms that are efficient for very large 
> problems (10 million - 100 million element).
> or
> E) something else ?
> 
> My contributions have been supported by SCEC but not CIG. Are there 
> wants/needs that I could contribute to and the community wants? If 
> possible, I would do them without CIG resources. If there are larger 
> software development efforts in this area, could/should CIG resources ($ 
> or people) be used for that development?
> 
> Cheers,
> Carl
> 
> Brad Aagaard wrote:
> > Hi all:
> > 
> > The Science Steering Committee will be meeting in two weeks to update the CIG 
> > five-year rolling strategic plan. At this meeting, I will present our 
> > priorities for software development by CIG. I have created two pages on the 
> > CIG website to document our desires 
> > (http://www.geodynamics.org/cig/workinggroups/short/workarea/planning/priorities2008/). 
> > One page is devoted to development plans for PyLith while the other describes 
> > our priorities year-by-year over the next five years. All content posted on 
> > these pages is up for discussion.
> > 
> > Some questions to help organize your thoughts:
> > 
> > (1) What obstacles inhibit your abilities to create realistic models?
> > (2) What modeling tools would eliminate/reduce these obstacles?
> > (3) If you are using PyLith, what features do wish it had?
> > (4) If you are not using PyLith, why? Are you waiting for a particular set of 
> > features to be added? Is it too difficult to learn? Is it too slow? Is it too 
> > inefficient?
> > (5) Are you satisfied with the pace of PyLith development? Would you be 
> > willing to work on PyLith development? What sort of training (if any) would 
> > you need?
> > (6) What other types of modeling tools, besides PyLith, do we want developed?
> > (7) Are there useful semi-analytic codes that would be of great use if they 
> > were more portable? documented? open-source? more efficient? Should we divert 
> > resources from PyLith development to support this task?
> > 
> > Please help define our needs and prioritize them for the coming years by 
> > participating in this discussion. Send comments/suggestions to this mailing 
> > list or visit the pages at the link above and add comments. NOTE: If you 
> > respond directly to me with suggestions/comments, I will forward your 
> > comments to this mailing list!
> > 
> > If you are interested in participating in a 1 hour teleconference towards the 
> > end of next week (May 21-23) to finalize this list of priorities, please let 
> > me know what times on those days you are available.
> > 
> > Thanks,
> > Brad
> > _______________________________________________
> > CIG-SHORT mailing list
> > CIG-SHORT at geodynamics.org
> > http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short
> 
>