CSM 2007 Highlight Archive
Nagi Rao Elected IEEE Fellow
On November 15, 2007 Nageswara Rao was elected to be an IEEE Fellow.
Through its global membership of more than 370,000 members, IEEE is a leading authority on areas ranging from aerospace systems, computers and telecommunications to biomedical engineering, electric power and consumer electronics among others.
The prestigous title of Fellow is awarded to no more than one-tenth percent of the total voting Institute memebership.
Please click HERE to learn more about IEEE.
CSMD Researcher's Student Wins HPC Award
Dr. Sudharshan Vazhkudai's student intern, Chao Wang, has been awarded the ACM/IEEE-CS HPC Ph.D. Fellowship (aka Supercomputing Fellowship). The fellowship was awarded for Chao Wang's previous/planned work on HPC center storage and distributed storage research with Dr. Vazhkudai as well as his work dealing with fault tolerance he conducted with his advisor, Frank Mueller, at NCSU.
This is the first year of this extremely competitive program, honoring exceptional Ph.D. students throughout the world with the focus areas of High Performance Computing, Networking, Storage and Analysis.
The selection sub-committee included:
- William Kramer - Lawrence Berkeley National Laboratory
- Charles Koelbel - Rice University
- Scott Lathrop - Argonne National Laboratory
Nagi Rao wins Best Paper Award
UT-Battelle Corporate Fellow Nageswara Rao’s paper Bandwidth Scheduling and Path Computation Algorithms for Connection-Oriented Networks won the award for Best Paper at the Sixth International Conference on Networking ICN 2007. This paper was co-authored by S. Sahni, S. Ranka, Y. Li and E. Jung from University of Florida and N. Kamath of Advanced Algorithms and Systems.
The paper presented efficient scheduling algorithms for computing the dedicated connections with specified bandwidth and duration in support of high performance applications. These algorithms have been implemented and demonstrated first over DOE UltraScience Net (USN) at Supercomputing Conference in 2005. This was the first demonstration of an advanced bandwidth reservation and signaling system on an operational network.
The conference is part of a group of conferences that are sponsored by the International Academy, Research, and Industry Association (IARIA). IARIA is a non-profit organization founded in Silicon Valley in 2004. Its aim is to promote scientific and industrial interchanges between existing associations, standards, and fora, to create bridges between different scientific and industrial cultures, and to offer stages for young scientists to shine.
Click HERE to view the paper.
Click HERE to visit the IARIA web site.
John Drake Quoted in USA Today Article
John Drake (Researcher and Group Leader for Computational Earth Sciences) of Oak Ridge National Laboratory was quoted in the USA Today article Climate Model Predicts Hot Decade.
In the article Drake points out that "In the climate-modeling world, a short prediction is considerably harder than a long one. That's because natural variability in weather has a stronger effect in the short term than when averaged out over 100 years. But the ability to produce accurate, 10-year predictions will be important for world leaders charged with making climate-related decisions".
Click here to read the entire USA Today article.
Click here to read the Science article refered to in USA Today.
Click here to visit the Computational Earth Sciences Group page.
Phani Nukala wins the Emerald Award
Phani Nukala, Oak Ridge National Laboratory Senior Research Scientist and co-principal investigator of ASCR project, Low-rank updates in Statistical Physics Applications, has won the Senior Investigator Emerald Award, which recognizes demonstrated excellence and consistent leadership in advancing science and technology through discovering, developing, and implementing novel technologies. The Emerald Awards, sponsored by Science Spectrum magazine, are billed as the premier awards for African Americans, Hispanics, Asian Americans and Native Americans working in the research sciences.
Clear Path Forward for the SNS Power Upgrade
Researchers from the Accelerator Systems Division at Spallation Neutron Source (V. Danilov, A. Aleksandrov, S. Assadi, W. Blokland, D. Brown, C. Deibele, S. Henderson, J. Holmes, A. Shishlo, and A. Webster) and Computer Science and Mathematics Division (J. Barhen, Y. Braiman, W. Grice, and Y. Liu) of Oak Ridge National Laboratory published results from their experiment in the May 2007 issue of Physical Review. They report on the proof-of-principle demonstration of H- laser stripping scheme that they have designed to give high efficiency (around 90%) narrowband conversion of H- beam into protons at the Spallation Neutron Source (SNS) in Oak Ridge, TN. The proposed H- laser stripping approach may provide a clear path forward for the SNS Power Upgrade by eliminating foil lifetime issues and providing a lower-beamloss approach for charge exchange injection. Read the Physical Review Special Topics article.
Researchers Sadaf R. Alam, Pratul K. Agarwal, Melissa C. Smith, and Jeffrey S. Vetter of Oak Ridge National Laboratory and David Caliga of SRC Computers publish results from their test using field-programmable gate arrays in the March 2007 issue of IEEE Computer.
A field-programmable gate array implementation (FPGA) of a molecular dynamics simulation method reduces the microprocessor time-to-solution by a factor of three while using only high-level languages. FPGAs are a type of logic chip that can be programmed. They are especially popular for prototyping integrated circuit designs. Once the design is set, hardwired chips are produced for faster performance. The application speedup on FPGA devices increases as the problem size increases. Researchers from Compuational Science and Mathematics and SRC Computers have used a performance model to analyze the potential of simulating large-scale biological systems faster than many cluster-based supercomputing platforms.
Researchers used high-level languages to conduct an analysis and FPGA implementation of the particle-mesh Ewald (PME) method, a biomolecular algorithm that is part of Amber (http://amber.scripps.edu), a widely used molecular dynamics framework.
Amber (Assisted Model Building with Energy Refinement) refers to two things: a set of molecular mechanical force fields for the simulation of biomolecules (which are in the public domain, and are used in a variety of simulation programs); and a package of molecular simulation programs which includes source code and demos. Amber takes a list of macromolecular residues which it links using standard geometrical parameters. The user adjusts this skeleton and generates parameters for the force field. Then energy minimization, molecular dynamics, or free energy perturbation calculations may be made. http://www.psc.edu/general/software/packages/amber/amber.html
Using the task-based implementation approach developed at ORNL and SRC Computers, scientific application developers can exploit extremely powerful yet flexible FPGA devices to perform a diverse range of scientific calculations without learning new programming languages or using a familiar high-level programming interface to compromise achievable performance.
Trends indicate that FPGA capabilities are growing at a rate exceeding that for microprocessors. Using accurate models of our current implementation, the researchers predict that next-generation FPGA devices will reduce the time to solution by a factor greater than 15 for large-scale biological systems- a speedup that's greater than that available with many parallel-cluster systems.
Further, the researchers anticipate that numerous scientific applications will dramatically benefit from increased support for double-precision floating-point operations and high-level languages because these reconfigurable devices offer an ideal combination of performance, concurrency, and flexibility for a diverse range of numerical algorithms.
Click here to view the article.
Results in Nature: Discovery of the first plausible explanation for a pulsar's spin
The flow vectors highlight two strong rotational flows. On the right the flow
is moving clockwise along with the shock pattern, whereas at the bottom left the
post-shock flow is being diverted into a narrow stream moving anticlockwise, fueling
the accretion of angular momentum onto the PNS.
Researchers John M. Blondin of North Carolina State University and Anthony Mezzacappa of Oak Ridge National Laboratory explain results from the Terascale Supernova Initiatve (Alumni SciDAC project: Shedding New Light on Exploding Stars) in the January 4, 2007 issue of Nature (Letters).
They report a robust instability of the stalled accretion shock in core-collapse supernovae that is able to generate a strong rotational flow in the vicinity of
the accreting proto-neutron star (PNS). This explanation of pulsar spin is the first to match observations by astronomers.
Read the nature article (pdf)
The results are based on 3-D simulations run on the Leadership Computing Facility Cray X1E at ORNL. This research team has been awarded
7,300,000 processor-hours from the 2007 INCITE program.