A publication of the Advanced Simulation & Computing Division, NA-121.2, NNSA Defense Programs
March 2009
NA-ASC-500-09—Issue 10Printable Version
(~1.6MB PDF)
The Meisner Minute
Editorial by Bob Meisner
… ASC now plays the central integrating role previously performed by nuclear tests, and is the only arena in which all aspects of the program are tested together.—Defense Science Board Task Force on the National Nuclear Security Administration (NNSA) Strategic Plan for Advanced Computing, January 2, 2009
Our newsletter this month focuses on the variety of contributions by ASC to national defense. This is a timely theme and it has been echoed in several external studies that have been completed in the past few months (including the one cited above). A recent demonstration of the central role of ASC is that the first refurbished W76 nuclear warhead has been accepted into the U.S. nuclear weapon stockpile by the Navy, culminating a ten-year effort that made extensive use of ASC-provided simulations and computing http://nnsa.energy.gov/news/2286.htm
DoD/DOE Strategic Alliance Workshop Held on Computational Simulation
On November 20, 2008, the first in a series of DoD/DOE strategic alliance workshops was held on the topic of verification and validation methods for computational simulation. Attendees included representatives from DoD research organizations (Air Force Research Laboratory [AFRL], Army Research Laboratory, Office of Naval Research), along with representatives from the DOE/NNSA weapons labs (SNL, LANL, LLNL) and NNSA HQ.
High-End Computing Research Needs Investment in I/O and Storage R&D
A multiagency working group—with representation from ASC—to help manage overall government investments in high-end computing research & development predicts that in the near future, gaps and open problems in the file systems and I/O in high-end computing will be formidable. These challenges present areas in need of new and continued investment in R&D and standardization that the government should pursue.
Roadrunner to Expand Hybrid Computing Applications
On December 22, 2008, Los Alamos National Laboratory (LANL) officially accepted the Roadrunner Phase 3 system from IBM. Having met this significant contractual milestone, LANL is moving to the next step of system and code stabilization. See the timeline graphic showing Roadrunner’s implementation schedule from August 2008 to January 2010. Weapons and Open Science code runs are being used to accomplish the stabilization.
NNSA Awards IBM Contract to Build Next Generation Supercomputer
Two new IBM supercomputing systems will help continue to ensure the safety and reliability of the nation’s aging nuclear deterrent. Sequoia will be a 20 petaFLOPS system based on future BlueGene technology, with delivery in 2011 and deployment in 2012. An initial delivery system, Dawn, a 500 teraFLOPS BlueGene/P system, is being delivered and installed. Dawn will lay the applications foundation for multi-petaFLOPS computing on Sequoia.
Kansas City Plant Acquires Tri-Lab Capacity Cluster Platform
As reported in the December 2008 issue of this newsletter, Kansas City Plant (KCP) has leveraged the work done by the three NNSA defense laboratories in an effort to bring forward the next generation of capacity platforms to the Nuclear Weapons Complex.
SIERRA Successfully Deployed at KCP in Support of Defense Programs Applications
The latest SIERRA engineering simulation framework has been deployed on the KCP Tri-Lab Capacity Cluster (TLCC) platform in support of various Defense Programs (DP) applications. The codes have been very successful, both on implementation and operations. The projects performed since last fall include various W76 applications–supporting current production build schedules. Examples include a Lightning Arrestor Connector abnormal loading review, Trajectory Strong Link feed-thru header sealing evaluations, encapsulation flow of electronics, header welding operations, vendor forging operations, safety reviews of mixing operations, and high-voltage abnormal drop loadings. Advanced development support for internal PDRD projects is also being performed with the tools, focused on fluid/solid interactions and coupling methods.
Structural Simulation Toolkit Mitigates Risks in Supercomputer Deployments
Taking chances with supercomputing is a very unhealthy business. Costs are in the tens to hundreds of millions. Complexity is high with numerous commodity and customized parts requiring integration. Innovative architectures like multicore come with extreme uncertainty. The simulation codes to be run are diverse. Potential unknowns are everywhere. Sandia has responded to this high risk by developing the Structural Simulation Toolkit (SST) to provide early insight into how future supercomputers will perform. The SST enables researchers to explore the performance of supercomputers running complex simulations virtually, all before funds and resources are committed. The SST has already had dramatic impact on the success of supercomputer deployments. It has also identified significant issues in scaling multicore hardware and pinpointed differences between current performance tests and real-world applications.
Study of Turbulence Mixing Showcases Science and Computation
Some of the largest fully resolved simulations of turbulence mixing to date showcase the ASC Program’s scientific and computational science capabilities. New and unexpected physics have been revealed in the study of mixing driven by strong pressure gradients. Applications of interest include inertial confinement fusion targets, laser induced launching of a flyer plate, stellar pulsations, and supernova explosions.
The Dawn of Petascale Computing at Lawrence Livermore
IBM recently delivered the final seven racks of the Dawn Supercomputer to LLNL. All 36 racks of the system are being installed in the Terascale Simulation Facility. Dawn is a predecessor to the gargantuan 20-quadrillion-operations-per-second Sequoia system, which will be delivered in 2011.
Application-Level Services Enable In Situ Analysis of Shock Physics Codes
The ability to perform meaningful, real-time analysis and visualization of scientific data, either simulated or real, is a significant advance that has the potential to fundamentally change the way computational scientists work. In situ analysis provides immediate feedback that allows the scientist to quickly detect bugs, identify areas of interest, or interactively modify the behavior of a running application. Such a capability will dramatically improve productivity of the application scientist and significantly reduce time-to-solution for time-critical ASC applications.
Report Now Available from ASC Risk Management Workshop for High-Performance Computing Centers
The RMTAP workshop, held Sept. 17 and 18, 2008, in San Francisco, CA, convened to assess current and emerging techniques, practices, and lessons learned for effectively identifying, understanding, managing, and mitigating risks associated with acquiring leading-edge computing systems at high-performance computing centers (HPCC). The report from the workshop detailing discussions and findings is now available online.
Seager Honored by Federal Computer Week
Lawrence Livermore’s Mark Seager was selected by Federal Computer Week magazine as one of this year’s “Federal 100” top executives from government, industry, and academia who had the greatest impact on government information systems in 2008.
LANL Computing Platforms Available for PSAAP Centers
As of February 2009, Los Alamos National Laboratory has announced a large increase in the amount of computing power available for use at Los Alamos by the Predictive Science Academic Alliance Program (PSAAP) Centers. To read more about the PSAAP Centers, see the announcement in the March 2008 ASC eNews.
ASC Salutes Sue Kelly
For the past seven years, Sue Kelly has been a driving technical force behind the acquisition, development, bring-up, and sustainment of the Red Storm massively parallel processing supercomputer located in Albuquerque. Sue began working on the Red Storm project soon after the contract was awarded to Cray, Inc., in 2002. The project involved considerable engineering R&D, some of which was performed by Sandia and was led by Sue. Sandia provided the scalable run time system software, which included the operating system, its libraries, the application launch utility and the node allocator.
DOE Privacy Disclaimer | Sandia Privacy Disclaimer |
Developed and maintained by Sandia National Laboratories for NA-121.2