2010 DOE INCITE projects allocated at ORNL

Oak Ridge National Laboratory researchers will lead projects that have been awarded a total of 251 million processor hours of computing time on supercomputers located at Oak Ridge National Laboratory and Argonne National Laboratory. These awards were made through the U.S. Department of Energy (DOE)’s Innovative and Novel Computational Impact on Theory and Experiment (INCITE) program. This year, the INCITE program will make approximately 1.6 billion processor hours available to projects designed to facilitate breakthroughs in areas such as climate change, alternative energy, life sciences and materials science. The projects were selected based on peer review and evaluations of their potential to advance scientific discovery.

Jeremy Smith, a biophysicist, was awarded 25 million processor hours to run simulations that will help reveal the inner workings of lignocellulosic biomass, a raw material for biofuel production. Cellulose is a complex carbohydrate that forms the cell walls of plants and gives leaves, stalks, stems and trunks their rigidity. Figuring out how to unlock its sugar subunits, which can be fermented to produce ethanol, is an engineering challenge. Meeting that challenge could enable full use of plants for cellulosic ethanol. The simulations are designed to provide a picture of biomass that will help experimentalists design plants with new, less resistant cell walls and enzymes that break cellulose down more efficiently.

Terry Jones, computer scientist, was awarded 4 million processor hours to investigate improvements in system software on leadership class computer systems. By the end of this decade exascale computers with unprecedented processor counts and complexity will require significant new levels of scalability and fault management. As system software designed decades ago is applied to computers that comprise hundreds of thousands of processors, leadership class systems and application software face significant challenges in the areas of providing reliability, availability and serviceability, as well as in removing load imbalances and scaling bottlenecks. This project tackles these problems using adaptive system software that dynamically adjusts as needs change, thereby reducing the demands placed on application developers. More information on this project is available at: http://www.hpc-colony.org.

Robert Harrison, a computational chemist, was awarded 75 million processor hours to investigate the rational design of catalysts using the reliable and accurate prediction of the electronic structure of large molecules and surfaces. Catalysts are crucial to many clean energy sources and for the development of new manufacturing processes with improved activity and selectivity. Additionally, catalytic processes are directly involved in the synthesis of 20 percent of all industrial products. The project will be used to develop a fundamental understanding of chemical transformation in several areas, including catalytic transformation of hydrocarbons; clean energy, including hydrogen production and storage; and the chemistry of transition metal clusters.

Markus Eisenbach, a computational scientist, was awarded 21 million processor hours to analyze magnetic systems and, in particular, the effect of temperature on these systems using first principles methods. By accurately revealing the magnetic properties of specific materials, the project promises to boost the understanding of magnetism in both scientifically and technologically relevant materials. This research will ultimately contribute to advances in such areas as magnetic storage and the development of motors for electric vehicles. The application that will be used to conduct this research—known as WL-LSMS—received the 2009 ACM Gordon Bell Prize for the world’s highest-performing scientific computing application.

A team of computational astrophysicists from ORNL, the University of Tennessee, Florida Atlantic University, and North Carolina State University, led by Anthony Mezzacappa, was awarded 34 million processor hours to develop three-dimensional models of core collapse supernovae. The simulations, which will include nearly all the factors likely to be important to the supernova explosion mechanism, will mark a significant step forward in core collapse supernova theory and will position the ORNL-led team to make far more realistic predictions of key observables associated with this class of stellar explosions, such as their neutrino and gravitational wave emissions and the production of elements, such as oxygen and calcium, necessary for the origin of life on Earth.

Patrick Worley leads a team of computer scientists that was awarded 20 million processor hours to maximize the utility of leadership-class computer systems. This research focuses on four primary goals: (1) updating and extending performance evaluation of all systems; (2) making performance tools available to high-end computing users, and further developing the tools to take into account the scale and unique features of the leadership-class systems; (3) validating the effectiveness of performance-prediction technologies and modifying them as necessary to improve their utility for predicting resource requirements for production runs on the leadership-class systems; and (4) analyzing and helping to optimize current or candidate leadership-class application codes.

Researchers in ORNL’s Nuclear Science and Technology Division were awarded 8 million processor hours to perform nuclear reactor simulations on Boiling Water Reactor (BWR) assemblies. This allocation will enable them to develop simulations at resolutions that were previously unobtainable. The equations governing the physics in a nuclear reactor core are very complex and require tremendous computational resources, both in terms of memory and processing speed. The proposed simulations will be the most detailed computer models ever developed for these types of assemblies. The results of these calculations will be used to determine the level of fidelity required to produce predictive reactor simulations, which is the next phase of technology development for nuclear energy modeling and simulation.

Jack Wells, a physicist, has been awarded 24 million processor hours to investigate materials that could be used to develop lithium air batteries capable of powering a car for 500 miles on a single charge. Lithium-ion batteries used in today’s emerging plug-in hybrid electric vehicles currently have a range of approximately 40 to 100 miles. This research project is focused on understanding (1) the mechanisms of Lithium/Air cell discharge and recharge reactions, (2) the role and selection of catalyst and cathode surface properties, (3) the solubility of lithium ion and lithium oxides and optimization of electrolyte and (4) the reactions occurring at the electrode-electrolyte interface.

To learn more about the 2010 INCITE program, visit http://www.er.doe.gov/ascr/incite/index.html.

About INCITE: The Innovative and Novel Computational Impact on Theory and Experiment program promotes cutting-edge research that can be conducted only with state-of-the-art supercomputers. Leadership Computing Facilities at Argonne and Oak Ridge national laboratories operate the program. The Leadership Facilities award sizeable allocations (typically tens of millions of processor hours per project) on supercomputers to researchers from academia, government and industry. The projects address "grand challenges" in science and engineering such as developing new energy solutions and gaining a better understanding of climate change resulting from energy use. The INCITE program, which was recently recognized by HPCwire for high-performance computing collaborations between government and industry, also includes several supercomputing allocations for companies such as General Electric and IBM to help develop technologies beneficial to the U.S. economy.

About America’s Leadership Computing Facilities: The Department of Energy provides a portfolio of national high-performance computing facilities housing some of the world’s most advanced supercomputers. For more information about the Leadership Class Facilities resources at Oak Ridge National Laboratory see http://www.ornl.gov/info/press_releases/get_press_release.cfm?ReleaseNumber=mr20091116-02. For information about resources at Argonne National Laboratory see http://www.anl.gov/Media_Center/News/2008/news080722.html.

ORNL is managed by UT-Battelle for the U.S. Department of Energy Office of Science.