Berkeley Lab Scientists to Lead, Support 14 New SciDAC Projects
September 26, 2012
Jon Bashor, [email protected], 510-486-5849
When the Department of Energy announced the series of projects under the latest Scientific Discovery through Advanced Computing (SciDAC) program, Berkeley Lab scientists, mathematicians and computer scientists were listed as key contributors in three institutes and 11 science application partnerships. Funding for the projects is expected to total about $6 million annually over the next three to five years.
The SciDAC Institutes and Scientific Computation Application Partnerships are key components of the program. While the Institutes form the foundation for efforts by applied mathematicians and computer scientists to systematically address technical challenges inherent to the scale of new architectures and common across a wide range of science applications, the partnerships enable scientists to conduct complex scientific and engineering computations at a level of fidelity needed to simulate real-world conditions in targeted science application projects. A critical aspect of the SciDAC program is the collaboration between domain scientists and mathematicians and computer scientists to ensure that the applications are both scientifically accurate and computationally efficient.
Through the Institutes, Berkeley Lab researchers will help improve the accuracy and fidelity of simulations on next-generation supercomputers, develop tools so that scientists can fully exploit the capabilities of the most powerful supercomputers, and help tackle the Big Data problem with new tools for managing, sharing and visualizing massive datasets. The partnerships will address a wide range of scientific challenges, such as understanding climate change, designing new accelerators, investigating the 95 percent of the universe known as dark energy and dark matter, improving our understanding of nuclear physics, chemistry and molecular dynamics, and modeling the characteristics of new materials.
The SciDAC program, first launched in 2001, brings together teams of some of the nation’s top researchers at national laboratories and universities to create the software and infrastructure needed to help scientists effectively utilize the next generation of supercomputers for tacking the toughest scientific challenges – some of which can only be studied through high performance computation and simulation. Berkeley Lab scientists have played key roles in projects funded under the previous two SciDAC programs, and will again contribute expertise in applied mathematics, computer science, physics, climate science and chemistry.
“The fact that so many of the projects are either led by or rely on the expertise of Berkeley Lab scientists is a strong endorsement of our leadership in computational science,” said David Brown, director of the Computational Research Division. “A number of our staff members are involved in multiple projects, which further underscores our capabilities.”
The SciDAC program is funded by programs in the DOE Office of Science: Advanced Scientific Computing Research, Basic Energy Sciences, Biological and Environmental Research, Fusion Energy, High Energy Physics and Nuclear Physics.
FASTMath (Frameworks, Algorithms, and Scalable Tachnologies for Mathematics): As the complexity of computer architectures and the range of physical phenomena that can be numerically simulated for important DOE applications continue to grow, application scientists have to continue to improve the quality of their simulations by increasing accuracy and fidelity and make their software and algorithms more reliable and robust. They will also have to adapt their computations to make effective use of the high-end computing facilities being acquired by DOE over the next five years. This challenge will necessitate million-way parallelism and implementations that are efficient on many-/multi-core nodes. The FASTMath SciDAC Institute will help DOE application scientists address both of these challenges by focusing on the interactions among mathematical algorithms, software design and computer architectures. FASTMath is led by Lori Diachin of Lawrence Livermore National Laboratory and CRD’s Phil Colella and Esmond Ng are on the executive council. CRD researchers contributing to FASTMath are Ann Almgren, John Bell, Jim Demmel (joint appointment with UC Berkeley and LBNL), Dan Graves, Sherry Li, Peter McCorquodale, Brian Van Straalen and Chao Yang. FASTMath is funded by DOE’s Advanced Scientific Computing Research Program.
SUPER (The Institute for Sustained Performance, Energy, and Resilience): The SUPER project is a broadly based SciDAC institute with expertise in compilers and other system tools, performance engineering, energy management and resilience. The goal of the project is to ensure that DOE’s computational scientists can successfully exploit the emerging generation of high performance computing (HPC) systems, which will increase in performance potential from tens to hundreds of petaflop/s (quadrillions of operations per second) over the next few years, and will evolve significantly from those in use today. This goal will be met by providing application scientists with strategies and tools to productively maximize performance, conserve energy, and attain resilience. SUPER is lead by Bob Lucas of the University of Southern California and Berkeley Lab participants are David Bailey, Leonid Oliker and Samuel Williams. SUPER is funded by DOE’s Advanced Scientific Computing Research Program.
SDAV (The Scalable Data Management, Analysis, and Visualization Institute): As scientists around the world address some of society’s biggest challenges, they increasingly rely on tools ranging from powerful supercomputers to one-of-a-kind experimental facilities to dedicated high-bandwidth research networks. But the scientists all face a common problem: massive amounts of data which must be stored, shared, analyzed and understood. Building on the success of previous SciDAC projects, SDAV will provide comprehensive expertise aimed at transferring state-of-the-art techniques in data management, analysis and visualization into operational use by application scientists, as well as working with researchers to further improve those tools. The SDAV Director is Arie Shoshani, head of Berkeley Lab’s Scientific Data Management Group. Wes Bethel is the principal investigator for Berkeley Lab and other LBNL participants are Hank Childs and John Wu. SDAV is funded by DOE’s Advanced Scientific Computing Research Program.
Science Application Partnerships
PISCEES (Predicting Ice Sheet and Climate Evolution at Extreme Scales): During the past decade, loss of mass from ice sheets has raised the global mean sea level by 1 millimeter per year. If recent trends continue, ice sheets will make a dominant contribution to 21-st century sea-level rise, far exceeding expert projections. Not only could this raise sea level, but also could affect other parts of the climate system. Building on recent successes of SciDAC and the Ice Sheet Initiative for CLimate ExtremeS (ISICLES), PISCEES will develop improved models and new tools will be implemented in the Community Ice Sheet Model (CISM) and the Community Earth System Model (CESM), providing a coherent structure for ongoing collaboration among glaciologists, climate modelers, and computational scientists. PISCEES is funded by DOE’s Biological and Environmental Research Program and will be led by Bill Lipscomb of Los Alamos National Laboratory (LANL). Berkeley Lab participants are Dan Martin, Esmond Ng and Sam Williams.
Multiscale Methods for Accurate, Efficient, and Scale-Aware Models of the Earth System: Some of the greatest challenges in projecting the future of the Earth’s climate result from the signiﬁcant and complex interactions among small-scale features and large-scale structures of the ocean and atmosphere. The project’s primary goal is to produce better models for these critical processes and constituents from ocean-eddy and cloud-system to global scales through improved physical and computational implementations. An integrated team of climate and computational scientists will accelerate the development and integration of multiscale atmospheric and oceanic parameterizations into the Community Earth System Model (CESM). Funded by DOE’s Biological and Environmental Research Program, the project is led by Bill Collins, co-leader of LBNL’s Climate and Carbon Sciences Program. Other Berkeley Lab participants are David Romps, Lenny Oliker, Michael Wehner and Sam Williams.
ComPASS (Community Project for Accelerator Science and Simulation): Particle accelerators are critical to scientific discovery both nationally and worldwide. The development and optimization of accelerators is essential for advancing our understanding of the fundamental properties of matter, energy, space and time. ComPASS will develop the HPC tools and applications necessary to design next-generation accelerators. It will build on the successful HPC accelerator modeling tools developed under SciDAC1 and SciDAC2. Funded by DOE’s High Energy Physics Program, ComPASS is led by Panagiotis Spentzouris of Fermilab. Esmond Ng of LBNL’s Computational Research Division is the project’s co-director for Computation. Other Berkeley Lab participants are Phil Colella, Miguel Furman, Cameron Geddes, Daniel Graves, Sherry, Li, Peter Schwartz, Brian Van Straalen, Jean-Luc Vay and Chao Yang.
Computation-Driven Discovery for the Dark Universe: As part of a new Department of Energy collaboration aimed at illuminating the 95 percent of the universe known as dark matter and dark energy, researchers in Berkeley Lab’s Computational Research Division will apply their scientific computing expertise in simulation and analysis to boost the success of next-generation cosmology experiments. Using Nyx, a new simulation code developed by LBNL’s Center for Computational Sciences and Engineering the Berkeley Lab members of the team will simulate gas and dark matter in a huge chunk of space – about 500 million light years on each side. Adaptive mesh refinement will enable resolution fine enough to resolve statistical information about structures like galaxies and galaxy clusters (the Milky Way is about 100,000 light years across). The high resolution enabled by Nyx will allow the team to study the low density gas found between galaxies. As light from distant quasars passes through these clouds, some of the light energy may be absorbed, which can indicate the presence of hydrogen and free electrons in space. Funded by DOE’s High Energy Physics Program, the project is led by Salman Habib of Argonne National Laboratory. Berkeley Lab participants are Ann Almgren, Zarija Lukic and Peter Nugent.
NUCLEI (Nuclear Computational Low Energy Initiative): The NUCLEI project will build upon recent successes in large-scale computations of atomic nuclei to provide results critical to nuclear science and nuclear astrophysics, and to nuclear applications in energy and national security. The envisioned large-scale computations will transform the fields of low-energy nuclear physics and astrophysics. Physics topics to be addressed include nuclear interactions and their uncertainties, ab-initio studies of light nuclei and their reactions, and studies of nucleonic matter and its astrophysical properties. The project will fundamentally advance the studies of neutron-rich nuclei and the fission of heavy nuclei, and the key nuclear physics issues in neutron star and tests of fundamental symmetries. NUCLEI is funded by DOE’s Nuclear Physics Program and led by Joe Carlson of LANL. Berkeley Lab participants are Hasan Metin Aktulga, Esmond Ng and Chao Yang.
A Multiscale Approach to Nuclear Structure and Reactions: Forming the Computational Bridge between Lattice QCD and Nonrelativistic Many-Body Theory: The project’s goal is to develop procedures by which a non-relativistic eﬀective theory of nuclear physics can be linked to the exact theory of the strong interaction, quantum chromodynamics (QCD), by connecting the low-energy constants of the former to lattice QCD calculations of nucleon-nucleon scattering parameters. The algorithms developed under this SciDAC project will have very general applicability and will beneﬁt multiple domain science applications. Development of these algorithms on both homogeneous and heterogeneous architectures ensures that the impact of this research remains relevant as the progress toward exascale platforms continues. The project is funded by DOE’s Nuclear Physics Program and led by LBNL’s Wick Haxton. Other Berkeley Lab participants are Esmond Ng, André Walker-Loud, Sam Williams and Chao Yang.
Simulating the Generation, Evolution and Fate of Electronic Excitations in Molecular and Nanoscale Materials with First Principles Methods: There are strong existing limitations on calculations of bound excited states in large molecules, and for resonance states in any polyatomic molecule. These limitations partly reflect deficiencies of existing software and algorithms. But more fundamentally they also reflect limitations of existing methods and models, particularly for resonances, or where multiple electrons are excited, or strong electron correlations are in play. The overall goal of this project is to make meaningful progress by coupling new and improved models for bound and metastable excited states from physical scientists with advances on underlying methodological challenges in applied mathematics, and practical realization via high performance computing. The project is funded by DOE’s Basic Energy Sciences Program and led by LBNL’s Martin Head-Gordon. Other Berkeley Lab participants are Evgeny Epifanovsky, Daniel Haxton, Sherry Li, William McCurdy, Esmond Ng, Sam Williams and Chao Yang.
Developing Advanced Methods for Excited State Chemistry in the NWChem Software Suite: A suite of new theoretical methods will be developed and implemented in the NWChem computational chemistry software suite in order to provide improved capabilities for excited-state dynamics in the gas phase and to add the capability to perform electronically excited-state dynamics in solution. Successful implementation will be transformative for the study of photochemical reactions with levels of accuracy similar to those commonly available for ground-state thermal reactions. Developed methods will have a broad impact as they will be implemented in a robust, widely available, actively supported software environment. The project is funded by DOE’s Basic Energy Sciences Program and led by Christopher Cramer of the University of Minnesota. Berkeley Lab participants are Esmond Ng, Chao Yang and Lenny Oliker.
Advanced Modeling of Ions in Solutions, on Surface, and in Biological Environments: This project combines expertise from academia and national labs to advance the state of ab initio molecular dynamics (AIMD) simulation in handling hydrated ions in situations relevant for future research applications dealing with energy and the environment. To do so, community AIMD codes will be elaborated and augmented to incorporate new theoretical concepts and recent algorithmic advances, and to run more efficiently on the DOE Leadership Class computing platforms. This project complements the ongoing experimental effort at the DOE Labs to probe the microscopic structure of water solutions using advanced spectroscopic techniques. Building the capacity of predicting from basic quantum theory ionic processes and spectroscopic observations in solution is a grand challenge for Materials and Chemical Sciences that will be addressed by our interdisciplinary team. The project is funded by DOE’s Basic Energy Sciences Program and led by Roberto Car of Princeton University. Berkeley Lab participants are Lin Lin, Esmond Ng and Chao Yang.
Discontinuous Methods for Accurate, Massively Parallel Quantum Molecular Dynamics:Transformational changes in energy storage technologies are critically needed to enable the eﬀective use of renewable resources such as solar and wind, and to make possible the transition from hybrid-electric vehicles to plug-in hybrids to all-electric vehicles, eliminating dependence on fossil fuels. Lithium ion cells have played a key role in the wireless revolution and have the potential to do the same for transportation and electrical distribution. To realize this potential, however, substantial increases in performance, lifetime, and safety will be required. This project will address this critical need by bringing to fruition a breakthrough in quantum mechanical simulations, discontinuous Garlerkin (DG) quantum molecular dynamics (QMD), and applying it to accomplish detailed simulations and analysis on an unprecedented scale. The project is funded by DOE’s Basic Energy Sciences Program and led by John Pask of Lawrence Livermore National Laboratory. Berkeley Lab participants are Lin Lin and Chao Yang.
Scalable Computational Tools for Discovery and Design -- Excited State Phenomena in Energy Materials: The objective of this project is to develop and implement new methods and theories to elucidate and predict excited electronic state phenomena in energy-related materials, such as those used in photovoltaic and photocatalysis applications. Understanding excited state phenomena requires knowledge of both the ground state properties and the related many-electron interactions with excited states, which may involve structural and orbital relaxations. As such, quantitative predictions for excited state phenomena are at the leading edge of current theories for the electronic structure of materials. A unified approach will be employed to describe simultaneously complex ground state structures and excited state phenomena. The project is funded by DOE’s Basic Energy Sciences Program and led by James Chelikowsky of the University of Texas at Austin. Steven Louie of Berkeley Lab’s Materials Science Division is the LBNL lead. Other Berkeley Lab participants are Andrew Canning, Jack Deslippe, Jeffrey Neaton, and Chao Yang.
About Computing Sciences at Berkeley Lab
High performance computing plays a critical role in scientific discovery, and researchers increasingly rely on advances in computer science, mathematics, computational science, data science, and large-scale computing and networking to increase our understanding of ourselves, our planet, and our universe. Berkeley Lab’s Computing Sciences Area researches, develops, and deploys new foundations, tools, and technologies to meet these needs and to advance research across a broad range of scientific disciplines.
Founded in 1931 on the belief that the biggest scientific challenges are best addressed by teams, Lawrence Berkeley National Laboratory and its scientists have been recognized with 13 Nobel Prizes. Today, Berkeley Lab researchers develop sustainable energy and environmental solutions, create useful new materials, advance the frontiers of computing, and probe the mysteries of life, matter, and the universe. Scientists from around the world rely on the Lab’s facilities for their own discovery science. Berkeley Lab is a multiprogram national laboratory, managed by the University of California for the U.S. Department of Energy’s Office of Science.
DOE’s Office of Science is the single largest supporter of basic research in the physical sciences in the United States, and is working to address some of the most pressing challenges of our time. For more information, please visit energy.gov/science.