From: Lori Freitag Diachin <diachin2@llnl.gov>

Date: Thu Sep 11, 2003  8:31:37 AM US/Pacific

To: kohlja@ornl.gov, John Shalf <jshalf@lbl.gov>

Cc: diva@lbl.gov

Subject: Re: DiVA Survey (Please return by Sept 10!)

 

Hi John and Jim,

 

Let me weigh in a bit on the TSTT item below as I am helping

coordinate that effort - so I can describe exactly what it is we're

up to...

 

At 07:59 AM 9/11/2003, James Kohl wrote:

Hi John,

 

  On Wed, Sep 10, 2003 at 04:49:42PM -0700, John Shalf wrote:

  > On Wednesday, September 10, 2003, at 03:32 PM, James Kohl wrote:

  > >>What do you consider the most elegant/comprehensive implementation for

  > >>data representations that you believe could form the basis for a

  > >>comprehensive visualization framework?

  > >

  > >Sounds like the "Holy Grail" to me...  If anything even remotely close

  > >to this already existed, we'd all be using it already...

  > >(Unless of course it's the dreaded NIH syndrome...)

 

  > I wanted to check to see if someone had already found the "Holy Grail"

  > and just forgot to tell the rest of us schmucks that game was up... :-)

 

I doubt it...  :-)  As I've mentioned before, the SciDAC TSTT center is

working on parts of this, but they are early in the progres....

 

The TSTT center is not interested in defining a data representation

per se - that is dictating what the data structure will look like.  Rather,

we are interested in defining how data can be accessed in a uniform

way from a wide variety of different data structures (for both structured

and unstructured meshes).  This came about because we recognize

that

  1.  there are a lot of different meshing/data frameworks out there,

       that have many man years of effort behind their development,

       that are not going to change their data structures very easily

       (if at all).  Moreover, these infrastructures have made their

       choices for a reason - if there was a one-size-fits-all answer,

       someone probably would have found it by now :-)

  2.  Because of the difference in data structures - it has been very

       difficult for application scientists (and tool builders) to experiment

       with and/or support different data infrastructures which has

       severely limited their ability to play with different meshing strategies,

       discretization schemes, etc.

 

We are trying to address this latter point - by developing common

interfaces for a variety of infrastructures applications can easily

experiment with different techniques and supporting tool developers

(such as mesh quality improvement and front tracking codes) and

write their tools to a single API and automatically support multiple

infrastructures.

 

We are also experimenting with the language interoperability tools

provided by the Babel team at LLNL and have ongoing work to

evaluate it's performance (and the performance of our interface in

general) for fine and course grained access to mesh (data) entities -

something that I suspect will be of interest to this group as well.

 

I know that you'll be having a lot of discussion of data structures

(perhaps interfaces would make sense for diva as well, rather than

data structures?) at the meeting next week, and I really hoped that

I could be there.  Unfortunately it conflicted with another conference

that I was already committed to.  If it's of interest, I would be happy

to provide more information about the TSTT effort - perhaps if the

data discussion spills over to the next meeting - I can talk about it

then.  Or I can send additional info in email.

 

Lori