ML-DDDAS Group
Introduction
The multilevel, dynamic data-driven application
simulation (ML-DDDAS) group is an outgrowth of Craig Douglas' research in multigrid
methods, hardware assisted (i.e., cache aware) multigrid methods, and DDDAS.
Our research has been generously supported over more than 20 years by the National Science
Foundation, Sandia National Laboratories, the North Atlantic Treaty
Organization, the University of Kentucky, Yale
University, Duke University, CERFACS (France), the Johannes Kepler Universität Linz (Austria),
LNCC (Brasil), Hewlett-Packard, Intel, and International Business Machines.
We strive to bring fast, accurate algorithms to
computational scientists using the latest algorithmic and data-driven
techniques. Application fields that we have addressed or are addressing include
- Wildland fire simulation
- Contaminant transport simulation
- Ocean modeling
- Flame simulation
Current Group Members
- Dr. Deng Li, senior scientist (joined group in August, 2004)
- Wei Li, Ph.D. student (joined group January, 2005)
- Ryan McKenzie, Ph.D. student (joined group January, 2003)
- Adam Zornes, Ph.D. student (joined group September, 2003)
- Prof. Gundolf Haase, frequent visitor from University
of Graz (2 months during 2004-2006)
- Prof. Walter Zulehner, frequent visitor from Johannes Kepler
University Linz (2 months during 2004-2006)
Recent Ph.D. Dissertations
The following are recent graduates of the group:
- Daniel T. Thorne (Ph.D. advisor C. C. Douglas),
Multigrid with Cache Optimizations on Adaptive Mesh Refinement
Hierarchies,
University of Kentucky Computer Science Department, December, 2003.
- Karen B. Walters (Ph.D. advisor C. C. Douglas),
Cascadic Multigrid Algorithms,
University of Kentucky Mathematics Department, April, 2003.
- Michael Sukop (Ph.D. advisor E. Perfect),
Porosity, Percolation Thresholds, and Water Retention Behavior of
Random Fractal Porous Media,
University of Kentucky Soils Department, February, 2001.
- Jonathan J. Hu (Ph.D. advisor C. C. Douglas),
Cache Based Multigrid on Unstructured Grids in Two and Three Dimensions,
University of Kentucky Mathematics Department, September, 2000.
Projects
Current projects include the following:
- NSF OISE-0405349 (9/1/2004 -
8/31/2007, $57K total) International: Fast Solvers for
Computational Problems Arising in Pharmacy, Life Sciences, Mathematics,
Physics, and the Environment. International partners Johannes
Kepler Universität Linz (Ulrich Langer and
Walter Zuhlener) and Karl Friedrichs Universität
Graz (Gundolf Haase and Alfio Borzì).
- NSF EIA-0324876
(9/15/2003 - 8/31/2007, $2.062M total) ITR/NGS: DDDAS: Data Dynamic
Simulation for Disaster Management. Joint project with
University of Colorado at Denver (J. Mandel), National Center for
Atmospheric Research (J. Coen), Texas A&M (W. Zhao), and Rochester
Institute of Technology (A. Vodacek).
- NSF ACI-0305466 (7/1/2003 - 6/30/2006, $253K total) Multiscale,
Multicolor, Multigrid-like Solvers for High Performance Technical
Computing.
- NSF ACI-0219627
(9/15/2002 - 8/31/2005, $500K total) ITR: Predictive Contaminant
Tracking Using Dynamic Data Driven Application Simulation (DDDAS)
Techniques. Joint project with Texas A&M (R. E. Ewing) and the
University of Utah (C. R. Johnson).
- Sandia Projects (ASCI and CSRI),
2001-2004.
- Intel-HP Itanium Exploration, 2001-2003.
Tangible Results
DDDAS Virtual Proceedings
|