University of Tennessee to establish next-generation data analysis center

Gregory Scott Jones

A visualization of the turbulent core of a simulated supernova. The high-resolution astrophysics simulation code, Chimera, was run on Kraken to explore stellar disruption.
Such visualizations are enabled by UT's new visualization center.
Image courtesy of David Pugmire, ORNL.
 Data courtesy of Bronson Messer, ORNL and the University of Tennessee.

To address this need, RDAV provides remote visualization and image generation, data and statistical analysis, workflow systems, and a variety of software services, making it a leading visualization and data analysis center with the nation’s broadest user community.

Much of RDAV relies on a new machine named Nautilus that employs the SGI shared-memory processing architecture. The machine features 1,024 cores, 4,096 gigabytes of memory, and 16 graphics processing units. The new SGI system can independently scale processor count, memory, and I/O to very large levels in a single system running standard Linux. This flexibility allows the RDAV team to configure a system uniquely capable of analyzing and visualizing petascale data sets, promising TeraGrid users new levels of scientific understanding.

Nautilus is used for three major tasks: visualizing data results from computer simulations with many complex variables, such as weather or climate modeling; analyzing large amounts of data coming from experimental facilities like ORNL’s Spallation Neutron Source; and aggregating and interpreting input from a large number of sensors distributed over a wide geographic region. The computer also has the capability to study large bodies of text and aggregations of documents.

“I believe this will be the largest shared-memory machine for analysis on the planet,” said the project’s Principal Investigator (PI) Sean Ahern, who is currently the visualization task lead at ORNL and serves as director of RDAV. “No one has ever done this before. The new system will handle data analysis algorithms that can’t be deployed on more traditional distributed memory systems.”

Shared-memory processing can be even more useful than the world’s most powerful computers for certain tasks, especially those aimed at visualization and data analysis. The system is complemented with a 1 petabyte file system and is fully connected to the TeraGrid, the nation’s largest computational network for open scientific research.

RDAV also features a support staff of visualization and data analysis experts adept at aiding researchers in interpreting their results and incorporating and developing new capabilities for understanding data.

TeraGrid XD is the next phase in the National Science Foundation's (NSF’s) ongoing effort to build a cyberinfrastructure that delivers high-end digital services, providing American researchers and educators with the capability to work with extremely large amounts of digitally represented information.

The Texas Advanced Computing Center at the University of Texas also received an XD award to operate another visualization and data analysis center.

"For many types of research, visualization provides the only means of extracting the information to understand complex scientific data," said Barry Schneider, program director in the NSF’s Office of Cyberinfrastructure. "The two awards, one to the Texas Advanced Computing Center at the University of Texas at Austin and the other to NICS at the University of Tennessee, will be deploying new and complementary computational platforms to address these challenges."

RDAV is fully leveraged with the resources of the National Institute for Computational Sciences (NICS), an NSF-funded organization managed by UT and home to Kraken, the world’s fastest supercomputer operated by an academic institution. RDAV began full operations in the second quarter of 2010 and is housed at NICS.

For more information, contact:

Sean Ahern, PI and RDAV director; ORNL visualization task lead; UT research associate professor

        (865) 241-3748,

Jian Huang, co-PI and UT associate professor in electrical engineering and computer science

        (865) 974-4398,

Gregory Scott Jones, Science Writer, Oak Ridge National Laboratory

        (865) 574-6944,