A-Z Index | Phone Book | Careers

National Energy Research Scientific Computing Center Announces Acceptance of One of the World's Largest Supercomputers

Department of Energy supercomputing center to advance scientific research with powerful Cray XT4™ system

November 1, 2007

Contact: Jon Bashor, jbashor@lbl.gov, 510-486-5849

SetWidth230-CrayXT4.10.JPG

CrayXT4

BERKELEY, CA and SEATTLE, WA — The U.S. Department of Energy’s (DOE) Lawrence Berkeley National Laboratory and Cray Inc. today announced the successful completion of the acceptance test of one of the world’s largest supercomputers. Installed at the DOE’s National Energy Research Scientific Computing Center (NERSC), the powerful Cray XT4™ system contains nearly 20,000 processor cores and has a top processing speed of more than 100 teraflops. 

The next-generation supercomputer will be used to advance a broad range of scientific research. Named “Franklin” in honor of the first internationally recognized American scientist, Benjamin Franklin, the Cray XT4 system enables researchers to tackle the most challenging problems in science by conducting more frequent and increasingly detailed simulations and analyses of massive sets of data.

“With Franklin, we are increasing the computational power available to our 2,900 NERSC users by a factor of six, providing them with access to one of the world’s fastest supercomputers dedicated to open scientific research,” said Michael Strayer, associate director of DOE’s Office of Advanced Scientific Computing Research, which funds NERSC. “We have high expectations that NERSC’s proven track record of scientific productivity will provide many new discoveries and understandings.”

The highly scalable Cray XT4 system is capable of running applications across a wide range of scientific disciplines, including astrophysics, fusion, climate change prediction, combustion, energy and biology. Franklin will enable researchers at Berkeley Lab to address such problems as developing better models of the Earth’s climate and using it to predict the impact of carbon dioxide emissions and global warming. The powerful system will also allow researchers to explore clean energy technologies and validate theories that attempt to uncover evidence that explains the origin of the universe.

“Our new Cray XT4 system has demonstrated that it can deliver a high sustained performance on a demanding scientific workload in a rigorous production environment while at the same time permitting users to explore scaling to 20,000 cores,” said Horst Simon, director of the NERSC Division at Lawrence Berkeley National Laboratory. “NERSC is proud to accept one of the largest ‘broad impact science’ supercomputers in the world for its demanding user community.”

Franklin has a theoretical peak speed in excess of 100 teraflops (100 trillion floating point operations per second).  In assessing proposed systems, the Cray XT4 scalable architecture promised to deliver high sustained performance, which is critical to NERSC’s 24x7 operation to meet users’ supercomputing demands.

“We are very excited to see one of the largest supercomputers in the world opened up to the expansive user community at NERSC,” said Peter Ungaro, president and CEO of Cray.  “The Cray XT4 system will provide the computational power to enable researchers who compute at NERSC to efficiently tackle some of the most important problems we face today.  With high sustained performance, scalability and upgradeability to petaflops capacity as its key attributes, the Cray XT4 supercomputer will help enable major advances in a number of scientific fields now and in the future.”  

As part of an extensive testing program, a number of NERSC users were given early access to Franklin to ensure that the system could handle the most demanding scientific applications.

“I am extremely impressed with Franklin,” said Robert Harkness, an astrophysicist at the San Diego Supercomputer Center working on a project aimed at precisely measuring the cosmological parameters that describe the shape, matter-energy contents, and expansion history of the universe. Running an application called ENZO, the project seeks to increase our understanding of the dark energy and dark matter thought to make up more than nine-tenths of the universe. “We have run the largest instances of ENZO ever, anywhere, and found that the performance and scaling on Franklin are both strong and better than other computer platforms we have used at other computing centers.”

Another project, led by Julian Borrill of Berkeley Lab, leveraged Franklin’s computing power to prepare for analyzing the massive amounts of data to be sent back to Earth by the Planck satellite, set for launch in 2008. A joint U.S.-European project, Planck will use 74 detectors to measure the cosmic microwave background, the residual radiation from the actual “Big Bang.”  Last scattered some 400,000 years after the Big Bang, it provides the earliest possible image of the universe, including encoded signatures of the fundamental parameters of all matter.

“I am delighted to report that we have just successfully created a map of the entire Planck Full Focal Plane one-year simulation,” Borrill said.  “This is the first time that so many data samples — three terabytes of data in 50,000 files, representing all the information collected by Planck during one mission year — have been analyzed simultaneously, a primary goal of our group's early Franklin efforts.”

Franklin contains 9,672 AMD dual-core Opteron 2.6 GHz processors with 39 terabytes of memory.  Running on 16,384 processor cores, the group was able to complete the run in just 45 minutes.


About Computing Sciences at Berkeley Lab

The Lawrence Berkeley National Laboratory (Berkeley Lab) Computing Sciences organization provides the computing and networking resources and expertise critical to advancing the Department of Energy's research missions: developing new energy sources, improving energy efficiency, developing new materials and increasing our understanding of ourselves, our world and our universe.

ESnet, the Energy Sciences Network, provides the high-bandwidth, reliable connections that link scientists at 40 DOE research sites to each other and to experimental facilities and supercomputing centers around the country. The National Energy Research Scientific Computing Center (NERSC) powers the discoveries of 6,000 scientists at national laboratories and universities, including those at Berkeley Lab's Computational Research Division (CRD). CRD conducts research and development in mathematical modeling and simulation, algorithm design, data storage, management and analysis, computer system architecture and high-performance software implementation. NERSC and ESnet are DOE Office of Science User Facilities.

Lawrence Berkeley National Laboratory addresses the world's most urgent scientific challenges by advancing sustainable energy, protecting human health, creating new materials, and revealing the origin and fate of the universe. Founded in 1931, Berkeley Lab's scientific expertise has been recognized with 13 Nobel prizes. The University of California manages Berkeley Lab for the DOE’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 science.energy.gov.