ORNL Researchers Winners Of Five DOE SciDAC Awards

Oak Ridge, TN - Oak Ridge National Laboratory has the lead on five projects funded through the Department of Energy's Scientific Discovery through Advanced Computing program and has supporting roles in seven other projects.

SciDAC, begun in 2001, is an integrated program that will help create a new generation of scientific simulation computer programs. The programs will take full advantage of the extraordinary computing capabilities of computers capable of performing trillions of calculations per second to address increasingly complex problems.

The 30 SciDAC-2 computational science projects, announced by DOE, will receive a total of $60 million over the next three to five years to advance fundamental research in several areas related to the department's missions, including climate modeling, fusion energy sciences, chemical sciences, nuclear astrophysics, high-energy physics and high-performance computing.

Associate Lab Director Thomas Zacharia noted that ORNL is poised to make significant contributions.

"ORNL researchers will play a key role in supporting the SciDAC mission, leading five SciDAC projects and partnering in more than seven others," Zacharia said. "SciDAC represents a commitment to propel simulation science to petascale computing and new discoveries. I am pleased that ORNL is well represented in supporting DOE in advancing frontiers of science."

ORNL's awards were for the following projects:

• A Scalable and Extensible Earth System Model for Climate Change Science. The award is for $4.8 million per year for five years for a team led by John Drake of the Computer Science and Mathematics Division.

  The goal of this project is to transform an existing, state-of-the-science third generation global climate model, the Community Climate System Model, to create a first generation Earth system model that fully simulates the coupling between the physical, chemical, and biogeochemical processes in the climate system. The model will incorporate new processes necessary to predict future climates based on the specification of greenhouse gas emissions rather than specification of atmospheric concentrations, as is done in present models that make assumptions about the carbon cycle that are likely not valid.

  The model will include comprehensive treatments of the processes governing well-mixed greenhouse gases, natural and anthropogenic aerosols, the aerosol indirect effect and tropospheric ozone for climate change studies. Also, the model will improve the representation of carbon and chemical processes, particularly for treatment of greenhouse gas emissions and aerosol feedbacks in collaboration with the DOE Atmospheric Science Program, DOE Atmospheric Radiation Measurement Program and DOE Terrestrial Carbon Programs.

• Performance Engineering for the Next Generation Community Climate Model System. This three-year $500,000 per year project is an auxiliary to the project led by Drake and to a five-year, $3 million per year computer science project led by the University of Southern California.

  Goals of the project, led by Patrick Worley of the Computer Science and Mathematics Division, are to improve the computer performance of the Community Climate System Model on existing computer systems and to ready the model for efficient execution on next-generation massively parallel systems. These goals will require comprehensive performance analyses and improving the model scalability to many thousands of processors.

• Steady State Gyrokinetic Transport Code, headed by Mark Fahey of the Center for Computational Sciences.

  This $240,000 per year three-year project focuses on developing calculations describing the physics of fusion reactors. These computer simulations will help scientists understand new features of the complex behavior of the hot swirling fuel, or plasma, inside the donut-shaped reactor vessel, called a tokamak. The simulations will address a key problem of critical scientific importance, which is predicting the performance of the International Thermonuclear Experimental Reactor given an edge boundary condition. Fahey and colleagues believe this effort is crucial for the success of the ITER project.

• Advanced Mathematics for Electronic Structure, a $300,000 per year three-year project headed by George Fann of the Computer Science and Mathematics Division.

  Fann and colleague Robert Harrison are developing multi-resolution fast and scalable models for high-precision computational chemistry computations. Potential applications include clean energy innovations for automobiles and industry (catalysis, fuel cells, combustion), improvements to the efficiency of drug development in the pharmaceutical field, and analysis of the dynamics of molecules and electrons in powerful new laser fields.

• Center for Technology for Advanced Scientific Component Software, a $3 million per year five-year project headed by David Bernholdt of the Computer Science and Mathematics Division.

  This project continues DOE's investment in the development of component software technology for high-performance scientific computing begun in the first round of the SciDAC program. Component-based approaches provide the developers of large-scale simulation software tools to help organize and manage the ever-increasing complexity of their software. Through the tools developed in this project, researchers will be able to collaborate with software developers across geographic and disciplinary boundaries, and help connect software written in different programming languages, different approaches to parallelism and even different scientific domains - all of which are common in today's environment, Bernholdt said.

SOURCE: Department of Energy/Oak Ridge National Laboratory