D.D, Koelling , Argonne National Laboratory
*(The answer might even be NO-but it's worth considering,)
We have been through a period of rapid advances in photon sources.
The real experimental effort is now more in the end stations (what to really
do with these new capabilities), Theory has continued "as usual" during this
build up. If one inquires what would be interesting to improve theory, the
answer is, of course, to build up all components: fundamental theory; modelling;
and analysis. We are skilled on all fronts and, at the same time, highly
deficient on all fronts.
A short shopping list ---
- Fundamental
- - Improvements in potentials (TD-DFT, excited state DFT or whatever outdoes DFT)
- - Really setting up and calculating cross sections, getting away from muffin-tinning
dipole approximations and angular momentum limitations
- - Electronic structure technologies (solid, cluster, embedding, film, scattering
state, Many body, ...)
- - Better handling of lifetime, scattering, and those "background" effects.
- Modelling
- - Find candidate large scale structures via dynamics (MD, MC)
- - Create statistical Ensembles
- - Mimic the dynamic behavior of the actual processes taking place in the experiment.
- Analysis
- Develop the capability to go from experiment to conclusion (where possible).
Experimentalists are doing much of this activity but organized theoretical input
can at least reduce "looks like a duck, must be a duck" kind of analysis.
This workshop is to help seek support for the endeavor:
- We could encourage and even coordinate applications for individual grants.
- This Can only accomplish small, incremental gains; may actually loose
ground: the facilities grants are now expiring!
- We can seek to establish a larger coordinated center: in all likelihood a
geographically dispersed center.
- DOE 2000, (tools)
- Network advances. (more tools)
- Coordination (US!)
Three possible roads ---
- Organize at a facility (probably *not* a good idea).
- Propose a virtual center somewhat along the Center for Synthesis and
Processing. (Probably todays effort),
- Propose a MAJOR project for the envisioned SSP effort. (This is what I want
to consider here.)
Scientific Simulation P...
- What is it?
- - Five year type project possibly starting in about two years,
- - Involves TERRflop computing,
- - Addresses a major scientific (and social) issue: global climate and
combustion will be prominent efforts following the Kyoto conference!
- - Involves major software - networking problem issue.
- To Play ---
- - Must pull together a large working group (virtual center).
- - Must identify a central thrust problem. (Spin off will be important!)
- The Buzzword: KILLER-APP
- -IT MUST BE BIG & IMPRESSIVE
- -IT MUST BE TIME-CRITICAL
- -IT MUST BE REAL