Science Stories

From TeraGrid Wiki

Jump to: navigation, search

The following stories and images were merged to create the above nuggets.

New 1.16.07 (12:50 central) Updated version of science stories text for the 2006 annual report can be found here.

This file contains one-paragraph versions of the science stories from the SC06 science highlights book, as well as additional contributions from Resource Providers. Sorted by directorate. Broken into main and appendix sections.

Changes and additions can be sent to Bill Bell for inclusion.

Images for use in annual report:

Nuclear pore protein binding to the surface of importin-beta. Image courtesy of the University of Illinois at Urbana-Champaign's Theoretical and Computational Biophysics Group.[1]

Geometry of the aorta and adjacent branches, colored to represent differences in pressure on the arterial walls. These computations were performed on NCSA and PSC systems as a cross-site simulation. Image courtesy of George Karniadakis, Brown University.[2]

Experiments have been unable to show how drugs gain access to the active site near the center of the protease, but simulations run on an SGI Altix system at NCSA revealed transient events in which the binding site opens and an inhibitor can enter. Image courtesy of Carlos Simmerling, Stony Brook University. [3]

The covering matrix for the football pool problem. The optimization instance is to find the smallest set of columns of the matrix such that every row is covered by the set of chosen columns. Image Courtesy Jeff Linderoth, Lehigh University. [4]

The TeraGrid network traffic map, which monitors the transfer rates on various networks, showing high transfer rates between Purdue and San Diego during the LHC Service Challenge. Image courtesy TeraGrid.[5]

Hercules, earthquake simulations snapshot from QuakeShow - simulation shows amplified seismic waves in the San Fernando Valley. Image courtesy PSC. [6]

This closeup on the bizinc MβL active site shows the conformation of the reactants (in stick representation) from the QM/MM simulation. Image courtesy Matteo Dal Peraro, University of Pennsylvania. [7]

A cross-sectional slice of an iron nanoparticle embedded in an iron-aluminide matrix, shows charge distribution on the atoms (blue to red, positive to negative). Image courtesy PSC. [8]

Black Holes with a pronounced filamentary structure. Yellow circles indicate black holes (diameter increasing with mass). Image courtesy Tiziana DiMatteo. [9]

PREDICTING PROTEIN STRUCTURES Discovering the 3-D structure of proteins opens the door to understanding their function. In a Strategic Applications Collaboration (SAC), SDSC computational scientist Ross Walker helped HHMI investigator David Baker and his group at the University of Washington modify their Rosetta code to utilize tens of thousands of processors, all working in parallel to compute the high-resolution protein structure prediction shown. The computation on SDSC’s Blue Gene supercomputer was part of a large number of Rosetta calculations run at SDSC for the CASP7 competition. Red is the native X-ray structure, blue is the Rosetta prediction, and green is a low resolution NMR model. Image: R.C. Walker, SDSC; S. Raman, U. Washington. [10]

VIRTUAL DEFORESTATION LiDAR Light Detection And Ranging (LiDAR) data is one of the hottest tools in the Geosciences for studying the earth’s surface, generating digital elevation models ten times more accurate than before. These images of the Northern San Andreas Fault near Fort Ross, California were produced using tools developed in the GEON project. The images show full land cover as well as the “bare earth” revealed by a “virtual deforestation” algorithm. Red lines show traces of the 1906 San Francisco earthquake. Note offset in drainage (center) due to cumulative motion along the fault. Credit: J. R. Arrowsmith, C. Crosby, J. Conner ASU/GEON; E. Frank, A. Memon, V. Nandigam, SDSC/UCSD. [11]

VIRTUAL JET IN A CROSSFLOW An exact simulation, without approximations, of a turbulent jet using Direct Numerical Simulations on SDSC DataStar supercomputer. Credit: Muppidi and Mahesh. [12]

ACCURATELY PREDICTING THE SUN’S CORONA Solar physicist Zoran Mikic and colleagues in the Solar Physics Group at SAIC ran their solar model for four days on SDSC’s DataStar supercomputer to compute the most accurate predictions to date of the sun’s corona, a capability that can lead to better forecasting of space weather impacts on satellites, power grids, etc. Their predictions for the total solar eclipse on March 29th, 2006, allowing easier observations of the corona, show the simulated magnetic field lines in the solar corona (upper right). Note the similarity to observations during the eclipse (full image). Solar north is up. Simulations: Solar Physics Group, SAIC. Observations: composite of Williams College Eclipse Expedition (NSF/NASA/National Geographic); and SOHO (NASA, ESA). [13]

EXPLODING STAR: TYPE IA SUPERNOVA Covering a representative one eighth of the star’s volume, the simulation on the TeraGrid DataStar resource at SDSC shows the distribution of burned material (which has undergone nuclear fusion). Dark red corresponds to the unburned carbon and oxygen mix, and white corresponds to fully burned material. Note the complex 3-D fractal geometry of the burning surface. The left panel shows the star two seconds after ignition and the right panel after 77 minutes. Credit: Alexei Poludnenko and Alexei Khokhlov, ASC Flash Center and the Department of Astronomy and Astrophysics, University of Chicago. [14]

SIMULATIONS OF TURBULENT MIXING Turbulent flows, which are very challenging to predict, include such important practical problems as how well different chemical species will mix and possibly react in a fluid. This 3-D volume rendering shows the scalar dissipation rate for a weakly diffusive passive contaminant computed on SDSC’s DataStar supercomputer on a 2,048 3 grid. The very high resolution of these direct numerical simulations in an SDSC SAC collaboration with NSF-supported research by P.K. Yeung and his group at Georgia Tech reveals the complex sheet-like topology of high activity regions where the most intense turbulent mixing and chemical reactions occur. Image: A. Chourasia, SDSC; D. Donzis. Georgia Tech. [15]

Continue to suggest additional items for inclusion on this page.

Please include:

Your name, institution, e-mail info. The science impact information should include PI, institution, e-mail, links to images, and address the following:

  • Why is this science important?
  • Why is the computational/cyberinfrastructure part hard?
  • What did the TeraGrid do to help accomplish the computational/cyberinfrastructure part?
Personal tools