|
Next
Steps in Using
Combustion Cyberinfrastructure
A Multi-agency Combustion Workshop
San Diego, California
March 28-29, 2007
Presentations and Discussions concerning the Workshop:
Agenda
Review of Workshop context,
agenda, and goals;
Phil Westmoreland, National Science Foundation
Perspectives from the TNF
Workshop;
Rob Barlow, Sandia National Laboratories
PrIMe – A Virtual Organization:
Part 1
Phil Smith, University of Utah
Part 2
Michael Frenklach, University of
California-Berkeley
Part 3
Greg Smith, SRI International
Developing
Cyberinfrastructure
for
Data-Oriented Science and Engineering
Fran Berman, Director, San Diego Supercomputing Center
Insights from the
Pittsburgh Combustion Simulation Workshop
Geo Richards, National Energy Technology Laboratory
Doug Talley, Air Force Research Laboratory
Discussions and Breakouts
Comments/Questions: Feel free to contact
Phil Westmoreland.
Comments for Discussion:
Michael Frenklach wrote (4/6/07):
In continuation of our discussion last week in San Diego, about extremely large-size data records,
I would like to propose to undertake an exploratory exercise: to select a ~1 GB size data record,
typical of the type of data collected in turbulent flames, and see if the current PrIMe Data Model
can accommodate it, and/or what modifications are needed to do so.
If this sounds something that you would like to do, I would suggest the following plan of action:
- To form a small work-group of people interested (it would be good to include among them also
those who know object-oriented programming and/or XML).
- The work group can identify/create a "representative" sample file of data (it would be easier
to start with no larger than 1-10 GB size).
- Zoran and I will create a first-cut XML example of this data
file/record following the present PrIMe Data Model, and identify
what additional features are possibly needed.
- We all will meet (actually or virtually) to discuss the results and consider further actions.
~ ~ ~ ~ ~
|
