SciDAC at Berkeley Lab
The U.S. Department of Energy's Scientific Discovery through Advanced Computing (SciDAC) program has brought together 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 tackling the toughest scientific challenges — some of which can only be studied through high performance computation and simulation.
Lawrence Berkeley National Laboratory is leading five of the current SciDAC projects and playing a key partnership role in nine others. A hallmark of the SciDAC program is the creation of multidisciplinary teams, so many of the projects involve collaborations with researchers at other national laboratories and universities.
Here is a brief summary of the projects centered at Berkeley Lab:
The Scientific Data Management Center for Enabling Technologies is led by Arie Shoshani of the Computational Research Division (CRD). This five-year project is improving methods for helping scientists manage and analyze the increasing volume and complexity of data produced by ultra-scale simulations and high-throughput experiments. The goal is to help researchers spend less time worrying about data and more time focused on science.
The Visualization and Analytics Center for Enabling Technologies, led by Wes Bethel of CRD, will address one of the greatest scientific and engineering challenges of the 21st century — to effectively understand and make use of the growing wealth of data to scientific advantage. This center will address this situation by adapting, extending, creating when necessary, and deploying technologies that will enable scientists to visualize and understand the copious data now available.
The Applied Partial Differential Equations Center for Enabling Technologies, led by Phil Colella of CRD, is developing a collection of algorithmic and software components that can be assembled to simulate a broad range of complex multicomponent physical systems in which partial differential equations play a central role. Their goal is to enable the agile development of high-performance simulation codes for complex multiphysics and multiscale applications by providing a flexible, capable, expressive, and high performance toolset.
The SciDAC Outreach Center, under the leadership of David Skinner at NERSC, is a clearinghouse for SciDAC activities and resources. The Center provides information and services in support of SciDAC objectives, including outreach, training, and research; fosters communication within the High Performance Computing (HPC) community; delivers SciDAC technologies to new communities; and helps disseminate SciDAC accomplishments. Additionally, the center gathers data to understand the needs of the HPC community to identify workshops, summer schools, institutes and research topics to meet those needs.
The Predicting the Function of Proteins for Newly Sequenced Organisms project, led by Steven Brenner of the Physical Biosciences Division, is refining algorithms to automate a process for predicting the functions of proteins in a family of microbes. By better understanding the function of these proteins, DOE hopes to harness the potential of microbial communities for the remediation of contaminated sites, development of smart sensors, and the bio-generation of hydrogen and ethanol.
Berkeley Lab Computing Sciences staff also serve as co-investigators in nine other SciDAC projects:
Computational Astrophysics Consortium: Supernovae, Gamma Ray Bursts, and Nucleosynthesis, led by Stan Woosley of UC Santa Cruz, includes Ann Almgren, John Bell, and Peter Nugent of CRD as co-investigators.
Sustaining and Extending the Open Science Grid: Science Innovation on a Petascale Nationwide Facility, led by Miron Livny of the University of Washington, includes Doug Olson of CRD as a co-investigator.
A Scalable and Extensible Earth System Model for Climate Change Science, led by John Drake of Oak Ridge National Laboratory, includes Michael Wehner of CRD as a co-investigator.
The Scaling the Earth System Grid to Petascale Data Center for Enabling Technologies, led by Dean Williams of Lawrence Livermore National Laboratory, includes Arie Shoshani of CRD as a co-investigator.
The Center for Enabling Distributed Petascale Science, led by Ian Foster of Argonne National Laboratory, includes Keith Jackson and Brian Tierney of CRD as co-investigators.
The Center for Scalable Application Development Software for Advanced Architectures, led by John Mellor-Crummey of Rice University, includes NERSC Director Kathy Yelick as a co-investigator.
The Performance Engineering Research Institute, led by Robert Lucas of the Institute for Information Sciences, includes co-investigators David Bailey and Daniel Gunter of CRD and NERSC Director Kathy Yelick.
The Petascale Data Storage Institute, led by Garth Gibson of Carnegie Mellon University, includes NERSC General Manager Bill Kramer as a co-investigator.