A publication of the Advanced Simulation & Computing Division, NA-121.2, NNSA Defense Programs
December 2008
NA-ASC-500-08—Issue 9Printable Version
(~1.6MB PDF)
Change Is in the Air
Editorial by Bob Meisner
Change is in the air in Washington, as it is across the nation…
It is risky to speculate on the next Administration’s expectations for NNSA and to what level the new team will support our efforts in the service of stockpile stewardship, but the early indications are certainly positive. Secretary Gates, who has agreed to remain a part of the President-elect Obama’s cabinet for at least a year, said at a speech in Minot Air Force Base in North Dakota that America’s security depends on a reliable and credible nuclear deterrent. “As long as others have nuclear weapons, we must maintain some level of these weapons ourselves to deter potential adversaries and to reassure over two dozen allies and partners who rely on our nuclear umbrella for their own security, making it unnecessary for them to develop their own.” Clearly, the Secretary of Defense understands our mission, and all indications are that he will support it energetically. For the position of Secretary of Energy, President-elect Obama has nominated a world-renowned physicist, the Director of Lawrence Berkeley labs, Stephen Chu. We know from public statements that a major priority for him will be alternate forms of energy production, and we believe that he will be an effective spokesperson for the scientific underpinnings and the necessity for a deep understanding of the performance of nuclear weapons as well as the avoidance of technological surprise from our potential adversaries. Who better to serve as a champion for science than a Nobel Prize winner in Physics who has been director of a premier science lab in the Department.
Impact Simulations on Red Storm Address National Security Problem
At the request of the Missile Defense Agency, Sandia researchers used ASC’s Red Storm supercomputer to address a complex, unprecedented mission. The objective was to protect against potential loss of life due to the uncontrolled reentry of a 2270 kg satellite and destroy its 450 kg hydrazine propellant tank. Preparation required countless hours of collaborative analysis to determine the feasibility of the SM3 interceptor to achieve this goal. Sandia was responsible for defining hit point requirements to achieve this objective using advanced modeling and simulation tools. In addition, Sandia assisted in predicting post-intercept satellite debris, and assessing kill/no-kill based on pre-event modeling. For more details, please go to http://www.sandia.gov/LabNews/081010.html
ASC Simulation Capabilities Used to Address Aircraft Bomb Issues
Application of ASC's high-performance computing capabilities to support areas other than the nation’s nuclear weapons complex is becoming more common. The latest beneficiary of DOE's/NNSA's investment is the Transportation Security Administration, which has received numerous briefings from an ASC tri-lab team on how computational modeling can help identify aircraft vulnerabilities. For more on this story, please go to: http://www.sandia.gov/LabNews/081010.html
Lawrence Livermore Teams with Computing Industry Leaders to Develop an Advanced Technology Cluster Testbed
Lawrence Livermore National Laboratory (LLNL) has teamed with 10 computing industry leaders to accelerate the development of powerful next-generation Linux clusters in a project dubbed Hyperion.
Hyperion brings together Dell, Intel, Supermicro, QLogic, Cisco, Mellanox, DDN, Sun, LSI and RedHat to create a large-scale testbed for high-performance computing technologies critical to NNSA’s work to maintain the aging U.S. nuclear weapons stockpile without underground nuclear testing, and industry’s ability to make petaFLOPS (quadrillion floating operations per second) computing and storage more accessible for commerce, industry, and research and development.
The Power of Interactive Simulation: SC08 Roadrunner Demo
The Los Alamos exhibit booth at SC08 featured a rack of 48 Cell processors of ASC's Roadrunner hardware to run demonstrations. A demonstration code was run using a new version of the PPM gas dynamics code with a special treatment of the multifluid flow using the Piecewise-Parabolic Boltzmann scheme (PPB). The small interactive runs performed in the demonstrations at SC08 ran to completion in just 7 minutes.
Kansas City Plant Moves to TLCC Platforms to Support W76 AF&F Production
The Kansas City Plant (KCP) has leveraged the work done by the three NNSA defense labs in an effort to bring forward the next generation of capacity platforms to the Nuclear Weapons Complex. One of the latest Tri-Lab Capacity Cluster (TLCC) platforms is now positioned at KCP for weapons manufacturing support. KCP’s mission is to produce many of the non-nuclear components for systems and is constantly challenged with maintaining high quality output with small production runs. Simulation technologies only enhance this mission, with the TLCC platform now enabling more coupled-physics and parametric studies.
A New Warm-Dense-Matter Computational Modeling Development Announced
Sandia ASC-funded researchers are developing novel kinetic algorithms to study the equilibration, stopping power, and fusion reactivity of dense, radiating, burning multi-component, multi-species plasmas.
Fully relativistic, electromagnetic simulation of moderately-to-strongly coupled plasmas and warm-dense-matter makes use of nonlinear inter-particle potentials calculated by a scalable implicit field solver. Validation of the simulations, within the LSP particle-in-cell code framework, against two-component hypernetted chain models enable the study of fundamental physical properties of coupled plasmas, including electron-ion temperature equilibration, and slowing-down of energetic charged particles in dense plasmas. Other applications include the study of the impact of impurities on various thermodynamic properties in warm-dense-matter, thermonuclear burn, and mix.
Latest version of DAKOTA Released
Version 4.2 of the DAKOTA software toolkit was released and deployed in November 2008, and offers substantial advancements that enable efficient, robust analysis and design of critical systems in the presence of uncertainty (See complete story for illustration).
DAKOTA, which is currently in use at Sandia, Los Alamos, and Lawrence Livermore National Laboratories to support the nuclear weapon stockpile stewardship program and other national security programs, is also used broadly by academic, government, and corporate institutions for sensitivity analysis, uncertainty quantification, parameter estimation, and design optimization studies. (see: http://www.cs.sandia.gov/dakota)
New 3-D Ion Beam Simulation Capabilities Explored via Sensitivity Analysis
Funded by the ASC Program, initial sensitivity studies of ion beam performance have been completed, using the new three-dimensional modeling capabilities of Sandia’s Aleph physics simulation code.
Protection Against Erroneous Monte Carlo Calculations
Note that Monte Carlo calculations are statistical by nature. Every time a Monte Carlo calculation of a problem is done, a different answer (estimated mean) for the problem results. Without knowledge of a confidence interval, “the” answer from a Monte Carlo calculation is almost meaningless. It is like a political poll, the answer depends on the random choice of the particular people polled. The average result of the poll is meaningful only if there is an estimate of the error in the poll numbers. Just as no sophisticated political campaign would base actions on a poll with a huge margin of error, no scientist or engineer wants to base the analysis of nuclear systems on calculations with unknown margins of error.
Roadrunner Maintains Top Supercomputing Spot
Roadrunner, the world’s first petaFLOPS system, ushers in a new era of computing for all applications of national interest. A revolutionary change, Roadrunner puts the nation on a different path to discovery. Though the revolution is not necessarily about speed, Roadrunner maintained its lead on the TOP500 list of supercomputers. The 32nd edition of the world’s top supercomputers was released on November 14, 2008, showing Roadrunner in the number one position with 1.105 petaFLOPS. Also cited for extraordinary energy efficiency, Roadrunner is number seven on the November 2008 Green500 list. The full Roadrunner system is now installed at Los Alamos.
Red Storm Regains Position in Top 10 Fastest Computers
The prior quarter’s newsletter article, “Red Storm Supercomputer Upgraded to 284 teraFLOPS” http://www.sandia.gov/NNSA/ASC/enews/0908/0908.html reported on the technical features of the latest upgrade to the Red Storm supercomputer.
It described the architectural design that has allowed this machine to evolve and remain viable as hardware technology advances. Since last quarter, a new Top 500 list was published in November. In its new configuration, Red Storm achieved 204.2 teraFLOPS when running the HPL (High Performance Linpack) benchmark. This result moved it from the 12th position to the 9th position in the ranking.
Los Alamos Hosts Predictive Science Panel
On November 12 through November 14, 2008, the Predictive Science Panel (PSP) met at Los Alamos. The PSP meets twice annually alternating between Lawrence Livermore (LLNL) and Los Alamos national laboratories to evaluate progress toward the goal of a credible predictive capability, answering the question: How do you know it’s right?
ASC Hosts Risk Management Workshop for High-Performance Computing Centers
At the request of the DOE Office of Science, ASC at LLNL hosted a two-day Risk Management, Theory and Practice (RMTAP) workshop Sept. 18–19 in San Francisco.
The purpose of the workshop, which was sponsored by the Office of Science/Advanced Scientific Computing Research and the NNSA/ASC program, was to assess current and emerging techniques, practices, and lessons learned for effectively identifying, understanding, managing, and mitigating the risks associated with acquiring leading-edge computing systems at high-performance computing centers (HPCCs).
ASC Booth Showcases Numerous Achievements at SC08 in Austin, TX
With a theme of “Leading HPC—Past, Present and Future,” the ASC Program once again demonstrated the numerous tri-lab advances in high performance computing at the International Conference on High-Performance Computing, Networking, Data Storage, and Analysis. Sponsored by the Association for Computing Machinery (ACM) and the Institute of Electrical and Electronics Engineers (IEEE) Computer Society, this year’s conference was held in Austin, TX, from November 15th to the 21st.
Visualization Team Upgrades ASC PowerWalls
Lawrence Livermore National Lab (LLNL) recently completed upgrades to the ASC visualization theaters to improve overall picture quality and reliability or add the ability to project native Powerpoint files. Built and maintained by the ASC Program, the Lab’s PowerWalls are extremely large multiprojector displays used for demonstrations, press conferences, visits, scientific collaborations, and other events that require high-resolution display capabilities.
ASC Salutes Paul Henning
If you can imagine how it would be to be faced with demonstrating performance on a radically new computer architecture and to write code without having the real processors, memory, or interconnect, then you can understand how exciting Paul Henning’s life has been while working on the new Roadrunner supercomputer. A technical staff member in the Computing, Computation, and Statistical Sciences Division at LLNL, Paul has been involved with the ASC Program since January 1999.
DOE Privacy Disclaimer | Sandia Privacy Disclaimer | 2009-0056 W
Developed and maintained by Sandia National Laboratories for NA-121.2