ASCR provides supercomputing facilities and advanced scientific networks along with programs to utilize these facilities.
Supercomputers or massively parallel high-performance computers, by every definition describes machines that employ very large numbers of processors in parallel to address scientific and engineering challenges. "Massively parallel processing" splits up a task or problem and parcels it out to multiple processors that work simultaneously but in concert to quickly supply results. High-performance computers link hundreds or thousands of processors - the computer chips that perform most calculations - in massively parallel configurations. Some of the processors in high-performance computers are specially designed for their purpose. In others, the processors are similar to the commercially available ones found in home computers, but they're linked by custom-made hardware and software that lets them work together quickly and efficiently.
High-performance computers carry out trillions or even quadrillions of calculations each second - powerful enough to simulate the most complex physical, biological and chemical phenomena. Supercomputers help scientists understand these processes at unprecedented levels - from individual atoms for nanoscale engineering to the entire planet for global climate studies. High-performance computers also are powerful enough to provide insight into systems and process by simulating them with great detail over relatively long times.
The primary facilities funded and managed by ASCR:
- Oak Ridge Leadership Computing Facility
Home to Titan, a Cray XK7 capable of about 20 thousand trillion calculations a second—or 20 petaflops—the OLCF combines world-class staff with cutting-edge facilities and support systems. The center serves scientists from all areas of the research community through programs such as the Department of Energy’s Innovative and Novel Computational Impact on Theory and Experiment (INCITE) program, ensuring it will be a computing powerhouse for the foreseeable future. In 2013, over 1.8 billion processor hours were awarded to 32 INCITE projects from universities, private industry, and government research laboratories, representing a wide array of scientific inquiry, from combustion to climate to chemistry.
Titan is the first major supercomputing system to utilize a hybrid-architecture, or one that utilizes conventional CPUs and unconventional graphics processing units (GPUs). In particular, Titan consists of 18,688 AMD 16-core Opteron 6274 CPUs (for a total of 299,008 processing cores) and 18,688 NVIDIA Kepler GPUs. It also features 710 terabytes of memory and a 10-petabyte file system that can move data at more than 240 gigabytes per second. Titan is the United States' fastest and most powerful supercomputer dedicated to open scientific research according to the November 2012 Top 500 list, a semiannual ranking of computing systems around the world making it a national asset.
- Argonne Leadership Computing Facility
The Argonne Leadership Computing Facility (ALCF) provides the computational science community with a world-leading computing capability dedicated to breakthrough science and engineering. The ALCF provides resources that make computationally intensive projects of the largest scales possible.
The National Energy Research Scientific Computing Center (NERSC) is the primary scientific computing facility for the Office of Science in the U.S. Department of Energy. As one of the largest facilities in the world devoted to providing computational resources and expertise for basic scientific research, NERSC is a world leader in accelerating scientific discovery through computation. NERSC is a division of the Lawrence Berkeley National Laboratory, located in Berkeley, California.
As one of the largest facilities in the world devoted to providing computational resources and expertise for basic scientific research, NERSC is a world leader in accelerating scientific discovery through computation. NERSC is a division of the , located in Berkeley, California.
More than 5,000 scientists working on 700 projects and running 600 different codes rely on NERSC systems and services to perform basic scientific research across a wide range of disciplines, including climate modeling, sustainable energy, research into new materials, simulations of the early universe, analysis of data from high energy physics experiments, investigations of protein structure, and a host of other scientific endeavors.
NERSC’s flagship supercomputer is Edison, a 124,608-core Cray XC30 system designed for scientific productivity. The system was designed with a high-speed interconnect, lots of memory bandwidth, lots of memory per node and very high input/output speeds to the file system and disk system, meaning that users’ data-intensive applications typically run faster than on other supercomputers. Edison is the first supercomputer at NERSC to rely solely on outside air for cooling, benefitting from the San Francisco Bay Area’s mild climate.
The Energy Sciences Network, or ESnet, is the Department of Energy’s high-speed network that provides the high-bandwidth, reliable connections that link scientists at national laboratories, universities and other research institutions, enabling them to collaborate on some of the world's most important scientific challenges.
Managed and operated by the ESnet staff at Lawrence Berkeley National Laboratory, ESnet provides direct connections to more than 40 DOE sites at speeds up to 100 gigabits per second, allowing scientists to manage, share and analyze massive datasets that are the hallmark of 21st century science. Connectivity to the global Internet is maintained through "peering" arrangements with more than 100 other research and education networks. Funded principally by DOE's Office of Science, ESnet allows scientists to use unique DOE research facilities and computing resources independent of time and location with state-of-the-art performance levels. ESnet derives its effectiveness from the extensive cooperation it enjoys with its user community. It is one of the most widely based and successful cooperative efforts within the Department of Energy.
The Innovative & Novel Computational Impact on Theory and Experiment (INCITE) program was conceived specifically to seek out computationally intensive, large-scale research projects with the potential to significantly advance key areas in science and engineering.
The ASCR Leadership Computing Challenge (ALCC) program allocates up to 30% of the computational resources at NERSC and the Leadership Computing Facilities at Argonne and Oak Ridge for special situations of interest to the Department.